Volume 5 - Annexes IV-X to the Expert Reports

STATUS AND MEMORIAL OF CHILEMEMORIAL AND EXPERT REPORTSVOLUME 1OF INTERNATIONAL COURT OF JUSTICEDISPUTE OVER THE STATUSAND USE OF THE WATERS OF THE SILALA(CHILE v. BOLIVIA)MEMORIALOF THEREPUBLIC OF CHILEANNEXES IV-XTO THE EXPERT REPORTSVOLUME 5OF 63 JULY 2017

ANNEX NO TITLE PAGE NO
VOLUME 5
ANNEXES TO THE EXPERT REPORTS (ANNEXES IV-X)
Annex IV
Latorre, C. and Frugone, M., 2017. Holocene
Sedimentary History of the Río Silala (Antofagasta
Region, Chile)
1
Annex V Mao, L., 2017. Fluvial Geomorphology of the Silala
River, Second Region, Chile 39
Annex VI McRostie, V., 2017. Archaeological First Baseline
Study for the Silala River, Chile 89
Annex VII Muñoz, J.F., Suárez, F., Fernández, B., Maass, T., 2017.
Hydrology of the Silala River Basin 163
Annex VIII SERNAGEOMIN (National Geology and Mining
Service), 2017. Geology of the Silala River Basin 279
Annex IX
Suárez, F., Muñoz, J.F., Maass, T., Mendoza, M., 2017.
Evapotranspiration Estimation in the Silala River Basin
- Methods Review and Estimation of Wetland
Evaporation
435
Annex X
Suárez, F., Sandoval, V., Sarabia, A., 2017. River-
Aquifer Interactions Using Heat as a Tracer in the
Transboundary Basin of the Silala River
475
Data CD CD-ROM containing supporting data to Annexes I–X 577

Annex IV
Latorre, C. and Frugone, M., 2017. Holocene Sedimentary History of the Río Silala (Antofagasta Region, Chile)
1
2
Annex IV
HOLOCENE SEDIMENTARY HISTORY OF THE RÍO SILALA
(ANTOFAGASTA REGION, CHILE)
Claudio Latorre Hidalgo (PhD)
Professor, Pontificia Universidad Católica de Chile
and
Matías Frugone Álvarez (PhD)
Associate Researcher, Pontificia Universidad Católica de Chile
May, 2017
Annex IV
3
GLOSSARY The following terms used in this report have been defined in accordance with the International Glossary of Hydrology, elaborated by the World Meteorological Organization (WMO-No. 385) and the International Peatland Society. Aggradation: (as a land surface process) Process of raising a land surface by the deposition of sediment. Aquifer: Geological formation capable of storing, transmitting and yielding exploitable quantities of water. Black mat: A typically well-sorted and fine grained, and sedimentary layer that range from thin (ca. 1 cm) strata with organic matter concentrations of several percent to thick (up to 1 m or more), more diffuse strata that may contain less than one percent organic matter (Pigati et al., 2014). Deposition: Deposition is the geological process in which sediments, soil and rocks are added to a landform or land mass. Discharge: Volume of water flowing per unit time, for example through a river cross-section or from a spring or a well. Erosion: The action of surface processes that remove sediments from one location and transport them it away to another location. Groundwater: Subsurface water occupying the saturated zone (i.e. where the pore spaces (or open fractures) of a porous medium are full of water). Infiltration: The movement of water from the surface of the land into the subsurface. Peat: Heterogeneous mixture of decomposed plant (humus) remains that has accumulated in a water-saturated environment and in the absence of oxygen. Pedogenesis: Process of soil formation and development by soil forming factors: climate (mainly temperature and precipitation), parent material, living organisms (plants and biota), topography, time, water and man. Perched Aquifer: Groundwater body, generally of moderate dimensions, supported by a relatively impermeable stratum and which is generally located between a deeper water table and the ground surface. Recharge Area: Area which contributes water to an aquifer, either by direct infiltration or by runoff and subsequent infiltration. 4
Annex IV
Rhizoliths: Organosedimentary structures resulting in the preservation of roots of
higher plants, or remains thereof, in mineral matter (Klappa, 1980).
Silt: Loose particles of rock or mineral (sediment) that range in size from 0.002 -
0.0625 millimetres in diameter. Silt is finer than sand, but coarser than clay.
Stratigraphy: Branch of geology dealing with the succession and age of rock strata.
Annex IV
5
TABLE OF CONTENTS 1 INTRODUCTION ..................................................................................................... 1 Presentation ........................................................................................................ 1 1.1 Geographic context of the Silala River and location of study sites.................... 2 1.2 Main objectives of this report ............................................................................. 3 1.3 Structure of the present report ............................................................................ 3 1.42 SUMMARY OF THE MAJOR FINDINGS.............................................................. 4 3 A FIRST APPROXIMATION TO THE GEOLOGIC HISTORY OF THE SILALA RIVER AND REGIONAL PALAEOCLIMATE DURING THE LATE QUATERNARY ............................................................................................................... 8 4 MATERIALS AND METHODS ............................................................................ 10 Field methods ................................................................................................... 10 4.1 Initial description of sites used in this study (see Fig. 1 for locations) ............ 11 4.2 SIL16-P01: S 22.02395°; W 68.03601°; 4241 m.a.s.l. (meters above mean 4.2.1sea level). ................................................................................................................ 11 SIL16-P02: S 22,02610°; W 68,03855°; 4223 m.a.s.l. ............................. 11 4.2.2 SIL16-P03: S 22,02690°; W 68,03942°; 4216 m.a.s.l. ............................. 12 4.2.3 Laboratory methods .......................................................................................... 12 4.3 Radiocarbon dating ................................................................................... 12 4.3.1 Calibration of radiocarbon ages ................................................................ 13 4.3.2 Radiocarbon analysis ................................................................................ 14 4.3.35 RESULTS ................................................................................................................ 15 Chronology ....................................................................................................... 15 5.1 Stratigraphy, Sedimentary Units and Depositional Environments ................... 19 5.2 Sedimentological description of stratigraphic units .................................. 21 5.2.16 DISCUSSION .......................................................................................................... 26 Holocene sedimentary infill of the Silala River ............................................... 26 6.1 Regional palaeoclimate and correlation with the Silala record ........................ 27 6.27 REFERENCES ........................................................................................................ 29 6
Annex IV
1
1 INTRODUCTION
1.1 Presentation
The present report was commissioned by the Dirección Nacional de Fronteras y Límites
del Estado (DIFROL) of the Ministry of Foreign Affairs of Chile to Professor Claudio
Latorre Hidalgo to address the Holocene (the most recent geologic epoch spanning the
last 11,500 years) sedimentary history of the Río Silala. This commission is to address
the current dispute between Chile and Bolivia regarding the waters of the Silala, and
focuses on the nature of its fluvial origins. Thus, the time during which this system has
existed as a river and the impacts that past natural climatic variability has had on its
sedimentary history are relevant antecedents to answer the origin of the waters that flow
through the present-day ravine and for how long these have done so.
The following report was researched and written by a team of two published experts in
late Quaternary (the last 50,000 years) palaeoenvironments (the study of the evolution
of Earth’s past environments) and palaeoclimate (the study of how climate has changed
in periods older than the historical record) based at the Department of Ecology, School
of Biological Sciences, Pontificia Universidad Católica de Chile. Dr. Matías Frugone
(PhD Geology, Universidad de Zaragoza, Spain) is an expert on limnogeology, a
discipline that studies the geological evolution of lake and sedimentary flood deposits
and how these relate to palaeoclimate. Dr. Claudio Latorre (PhD in Ecology and
Evolutionary Biology, Universidad de Chile, Chile) is an expert on the Quaternary
palaeoecology and hydroclimate history of the Atacama Desert. He has over 20 years
research experience in the Atacama and other arid regions using multiple different
geohistorical archives, including rodent middens (organic-rich nest deposits formed by
rock-dwelling rodents that contain plant, animal and insect remains preserved for
millennia encased in crystallized urine in caves and rock shelters- see Latorre et al.,
2002), plant macrofossils, palaeowetland and lacustrine deposits and human prehistory.
The research presented in this report was carried out in close collaboration with two
international experts on hydrology and hydrogeology, Dr. Howard Wheater (Director,
Global Institute for Water Security, University of Saskatchewan, Canada) and Dr. Denis
Peach (British Geological Survey and Imperial College, London, UK). The maps
included were specifically drafted by Coalter Lathrop (Sovereign Geographic) for this
report based upon our source materials.
Annex IV
7
2 Geographic context of the Silala River and location of study sites 1.2Figure 1. Map showing the sampled sites and major summits near the Silala River. Contours are every 50m. Points labeled “SIL16” are the sampling sites used in this report. Local intakes are also shown. The Silala River originates from at least two different sets of springs located in Bolivia. The water in these springs is most likely to originate by infiltration of the precipitation (rain and snow) at high elevations in the Andes, mainly within Bolivia but there are also many natural springs along the Silala that feed into the river from the Chilean side. A recent study calculated average annual discharge at the international border 0.17  0.2 8
Annex IV
3
m3/s (Muñoz et al., 2017). Such groundwater systems are common throughout the
northern regions of Chile (see Rech et al., 2002, 2003).
Before crossing into Chile, the Silala becomes entrenched and carves a ravine into the
existing bedrock that is more than 10 m deep (Muñoz et al., 2017). During the 20th
Century, the river was impounded on the Chilean side along two stretches, one
approximately 40-50 m from the international border and the other near the Inacaliri
Police Station, located approximately 5 km from the border (Fig. 1). A major drainage
channel, Quebrada Negra reaches the Silala from the southeast some 500 m downstream
from the border. The study sites in this report all carry the abbreviation SIL16-pXX
where SIL16 signifies the year and locality of sampling (Silala 2016) where pXX
denotes the sequentially numbered stations or stops (paradas in Spanish).
1.3 Main objectives of this report
The first objective of this report is to locate, describe and date the unconsolidated
(Quaternary) surficial sedimentary deposits within the Río Silala ravine to establish the
erosional and depositional history of the Silala. The dating of these deposits allows for a
more precise estimation of past fluctuations in the local groundwater table and enables
the reconstruction of past major phases of deposition along with past episodes of
incision (see Rech et al., 2002; Quade et al., 2008 and Sáez et al., 2016 for published
examples of how these methods are applied in the Atacama Desert).
The second objective is to then use this past local history of the Silala River sediments
to correlate with regional palaeoclimate variations described for the central Andes and
the Atacama Desert during the Holocene (the last 11,500 years). Such a correlation
provides evidence that the depositional history of the Silala is linked to rainfall and
associated aquifer recharge.
1.4 Structure of the present report
This report is divided into five chapters. The first chapter introduces the nature of the
report and why it was commissioned, followed by the geographic context and the main
objectives of this study. Chapter 2 provides a summary of the major findings and
presents a sequence diagram of events relevant to the Holocene sedimentary history of
the Río Silala. Chapter 3 presents the methods used to attain the objectives, along with
site selection and location and explains the procedures used to describe the stratigraphic
columns, as well as the method used to date the samples collected. Chapter 4 presents
Annex IV
9
4 the major results of our study and Chapter 5 discusses these results in terms of the local environmental evolution of the Silala River and how this evolution relates to regional past climate change. 2 SUMMARY OF THE MAJOR FINDINGS 1)We show that after an initial period of deep incision to form the present ravine, theunconsolidated sediments present in the Río Silala have a depositional history thatdates to the early Holocene (more than 8430-8350 years ago or ca. 8.4 ka) (Fig. 2).2)These deposits formed during three aggradational (deposition of sediment) phasesdated to older than ca. 8.4 to 1.9 ka (Unit 1) (Fig. 3.1), older than 0.65 - 0.2 ka(Unit 2) (Fig. 3.2) and younger than 0.2 ka to the recent 20th century (Unit 4) (Fig.3.4.1). These phases of aggradation are for the most part, coeval with periods ofelevated groundwater tables that have been well-documented and dated throughoutthe Atacama Desert.3)Abrupt incision likely occurred due to lowering of the water table <1.9 ka (Fig. 3.2)and >0.2 ka. A drop in the water table causes mass mortality of plants on thesurface which in turn protect the riverbed from erosion. Major incision ensues,mostly due to the removal of unconsolidated deposits by rare high rainfall eventsthat produced episodic flooding of the river, which can quickly erode theunconsolidated sediments (Fig. 3.3). Such episodes are more frequent when largefalls in the water table are caused by regional and persistent phases of arid climateon decadal or centennial timescales.4)Very recent (21st century) incision of Unit 4 (>1.5 m in some areas) is currentlyvisible throughout parts of the Río Silala where dried-out standing vegetation canbe seen (Fig. 3.4.3). This incision may be due to multiple factors, the clarificationof which is beyond the present study.5)The evidence presented here shows that the Silala River is a fluvial system that hasfluctuated over several millennia. The depositional units present were deposited in aperennial river system that was (at very specific times in the past) subject toincreased erosion in concordance with regional climate change. This regime ofperennial flooding and incision has existed for at least the last 8500 years. Thecycle of erosion and deposition clearly continues to the present.
10
Annex IV
5
Figure 2. A summary schematic diagram showing the cross-cutting relationships of the
Holocene sedimentary units present in the Silala River. Unit U4 (deposited during the 20th
century) is inset into the older (early to mid-Holocene) U1, indicating a major phase of incision
after deposition of U3, after 500 years ago. All radiocarbon ages are in calibrated years before
1950 AD (cal yrs BP), except for those more recent ages marked AD.
Annex IV
11
6 Figure 3. Sequence of depositional and erosional phases in the Silala River. The sequence was elaborated from a composite of the stratigraphic columns described here and is labelled from the oldest event (1) to the youngest (4). All radiocarbon ages are in calibrated years BP. 12
Annex IV
7
Figure 3. (Continued)
Annex IV
13
FIRST APPROXIMATION TO THE GEOLOGIC HISTORY OF THE SILALA RIVER AND REGIONAL PALAEOCLIMATE DURING THE LATE QUATERNARYFigure 4. A view of the Silala ravine looking NE showing majorerosional (T2, T3) anddepositional(T1)terraces.TheSilala River is immediately to the right ofT1in this photograph and is marked by the dense stands of brightyellowwaiya grass (Calamagrostiseminens). Cerrito de Silala is in the background. Note recently dried outwaiyagrass stumps on the T1 surface.
14
Annex IV
9
Beginning in Bolivia just downstream of the Orientales wetland, the Silala ravine
continues into Chile across the international border. Here, the Silala ravine is incised
10-15 m into local bedrock composed chiefly of Silala Ignimbrite and below the
Inacaliri Police Station, of Pliocene Cabana Ignimbrite (SERNAGEOMIN, 2017). This
incision followed a phase of alluvial and glacial outwash sediment accumulation when
glaciers existed along the flanks of adjacent peaks during the late Pleistocene 40 to 12
thousand years ago (SERNAGEOMIN, 2017). The first phases of incision formed the
first three erosional terraces present in the Silala ravine (from oldest to youngest, T4-
T3-T2) (SERNAGEOMIN, 2017). The youngest of these erosional terraces (T2) is
approximately 4 m above the current river level and represents a bedrock valley floor
that formed before the deposition of the sedimentary units present within the current
ravine. The T2 terrace was then further deeply incised (4+ m) before the deposition of
Unit 1, at the onset of the Holocene (before ca. 8.4 ka) which then gave way to the
current inset sediment deposits present within the modern valley. The recent (modern)
and lowermost fluvial terrace is designated T1 and is a depositional feature (Fig. 4).
Extensive past research in the Atacama Desert has shown that the sedimentary and
erosional history of such systems is the result of a complex relationship between the
height of the emergent groundwater table and regional climate change (Rech et al.,
2002, 2003; Quade et al., 2008; Sáez et al., 2016). Although aggradation of fine
sediments in stream deposits is typically associated with decreased river flow (and
erosion with increased flow), when base-level is controlled by the height of the
groundwater table in a desert environment, such sedimentary regimes can be
considerably more complex (Webb et al., 2014).
Regarding the perennial and intermittent fluvial systems in the Atacama Desert, Rech et
al. (2002: 344) state “Increased water-table heights at these locations [in their study]
reflect enhanced groundwater discharge, which can be supported only by higher
recharge and precipitation in the High Andes. Furthermore, the general concordance of
records from different parts of the regional hydrologic system—including groundwater
fed Andean streams (Quebrada Puripica, Río Tulán), point-source springs (Tarajne), and
water tables in the Calama basin (Río Loa, Río Salado) — suggests a relatively
synchronous response to changes in regional recharge, notwithstanding major
differences in size, gradients, and location of the hydrologic systems with respect to the
recharge area [our emphasis]”.
In summary, the depositional history of groundwater fed rivers can be linked to changes
in the long term climatic regime. When a significant increase in aridity on decadal or
greater timescales causes regional groundwater levels to drop, vegetation on the surface
on unconsolidated deposits dies off and facilitates incision of unconsolidated finegrained
deposits, with erosion caused by occasional or extreme flash flooding. The
Annex IV
15
10resultant high flow events can rapidly erode previously existing non-consolidated sediments, which were deposited under much wetter climates of previous regimes when groundwater levels were higher (Pigati et al., 2014). Our previous research with rodent middens (hardened nest deposits of rock dwelling rodents that contain valuable palaeoecological information) and sedimentary deposits formed by extensive wetlands in the Salado River basin (22º S) show that these valleys have undergone significant variations in the groundwater table which were synchronous with large changes in past rainfall (Latorre et al., 2002, 2003, 2005, 2006; Rech et al., 2002, 2003; Maldonado et al., 2005; Díaz et al., 2012). For example, a large increase in precipitation occurred at the end of the Pleistocene (17,500 to 9,500 years ago), known as the Central Andes Pluvial Event (CAPE, see Latorre et al., 2006; Quade et al., 2008). This millennial-scale event formed a slew of spring and in-stream wetland deposits (palaeowetlands) throughout the western slope of the Andes (Rech et al., 2002, 2003; Latorre et al., 2007; Quade et al., 2008). Other studies have further shown increased groundwater levels and fluvial activity in the low-elevation hyperarid basins such as in the Pampa del Tamarugal basin during the CAPE (Nester et al., 2007; Gayó et al., 2012a). Increases in the height of groundwater tables also occurred during the mid-Holocene (8000 to 4000 years ago) and in the late Holocene from 1000 to 700 years ago (Betancourt et al., 2000; Rech et al., 2003; Gayó et al., 2012b). Likely driven by increased rainfall, these long term increases in recharge raised the local height of the groundwater table by several meters, effectively flooding valley floors with extensive wetlands in areas that today are canyons and ravines with dry river beds. The last major recharge event due to higher precipitation occurred as recently as 200 years ago, and since then the rains have declined throughout the Altiplano, especially during the 20thcentury (Morales et al., 2012; Mujica et al., 2015; Lima et al., 2016). 4MATERIALS AND METHODSField methods4.1Holocene surficial geologic deposits along the Río Silala were described, mapped, and dated with radiocarbon (from here on termed “14C”) and uncalibrated dates areexpressed in “14C yrs BP” where “BP” (Before Present) refers to before the year 1950AD (see section 4.2.2). Locations of in-stream unconsolidated deposits for detailed sedimentological descriptions were chosen based on the depositional facies present, the presence of wood or organic beds feasible for accurate radiocarbon dating, and the identification of clear stratigraphic unconformities. A ~6 km stretch of the Silala ravine 16
Annex IV
11
(Fig. 1) was investigated to understand potential changes in stream flow and to
reconstruct the aggradation (sediment accumulation) and incision history. A surficial
map and five valley cross-sections were constructed for this reach of the stream from
sedimentological descriptions, inset stratigraphic relations, and radiocarbon dating of
plant macrofossils included within these deposits. A large outcrop of brown to light
grey, coarse to fine-grained sediments with ~2 m of exposed section (SIL16-P03)
located along the southeast bank of the Silala was identified as being the most
promising for understanding the Holocene palaeoenvironmental setting of the site and is
a key focus of this study. Wherever possible, samples were collected from the top and
bottom of stratigraphic units to constrain the 14C ages of the deposits. The locations of
the outcrops used for the description of stratigraphic sections were established using a
GPS with a horizontal accuracy of ±10 meters. The ages of the stratigraphic units were
constrained by cross-cutting stratigraphic relationships and by 14C ages on 11 samples
of plant fragments and organic-rich sediments.
Initial description of 4.2 sites used in this study (see Fig. 1 for locations)
4.2.1 SIL16-P01: S 22.02395°; W 68.03601°; 4241 m.a.s.l. (meters above mean sea
level).
Surface outcrop of whitish fine-grained sediments with indications of being recently
vegetated by waiya (i.e. most plants are no longer living and remain only as clumps).
The whitish appearance is most likely calcium carbonate (which fizzes in the field when
reacted with dilute (10%) HCl).
4.2.2 SIL16-P02: S 22,02610°; W 68,03855°; 4223 m.a.s.l.
The site consists of a small terrace with fine sedimentation located on the southwest
bank of the river about 1 m above the level of the current channel. On this terrace, there
is an extensive mantle of fine sediments (silts, sands) and dried standing vegetation
composed chiefly of “huailla” or “waiya” grasses (Calamagrostis eminens) (Fig. 4).
Annex IV
17
12SIL16-P03: S 22,02690°; W 68,03942°; 4216 m.a.s.l.4.2.3An extensive outcrop of palaeowetland sediments located where the Silala ravine is joined by the Quebrada Negra (Figure 5). This deposit is approximately 4 m thick and contains fluvial gravels, laminated organic sediments (often termed “black mats”) and fine sediments (silt, fine sands) typical of the palaeowetland deposits present in the region (e.g. Rech et al., 2002). The deposit is capped by a massive c. 2 m thick debris flow unit. Laboratory methods4.3Radiocarbon dating4.3.1Radiocarbon dating is one of the most extensively and widely used dating methods in the environmental and earth sciences. It can be applied to almost all kinds of organic materials and spans dates from a few hundred years ago to about 50,000 years ago depending on the laboratory (Taylor, 2014). In general, it is always better to date properly identified single remains (such as a cereal grain, bone or identified terrestrial plant) rather than a mixture of unidentified (i.e., “bulk” organic remains). The isotope radiocarbon 14C is formed in the upper atmosphere by spallation (loss of proton byimpact with cosmic rays) of 14N and then oxidized to form carbon dioxide. This is takenup by plants through photosynthesis. Because all the carbon present in plants originally comes from the atmosphere, the ratio of 14C to stable carbon isotopes (12C mostly) in theplant is virtually the same as that in the atmosphere. Plant eating animals (herbivores and omnivores) get their carbon by eating plants. Thus, all animals in the food chain, including carnivores, get their carbon indirectly from plant material, even if it is by eating animals which themselves eat plants. The net effect of this is that all living organisms have the same 14C to 12C ratio as the atmosphere (for more detail see https://c14.arch.ox.ac.uk/). Once an organism dies, however, carbon is no longer replaced. Because 14C is radioactive, it will slowly decay away. The proportion of radiocarbon to stable carbon (which does not decay) will reduce according to the exponential decay law: R = A exp(-T/8033) where R is 14C/12C ratio in the sample, A is the original 14C/12C ratio of the livingorganism and T is the amount of time in years that has passed since the death of the organism (see https://c14.arch.ox.ac.uk/). 18
Annex IV
13
By measuring the ratio, R, in a sample we can then calculate the age of the sample in
years:
T = -8033 ln(R/A)
4.3.2 Calibration of radiocarbon ages
14C measurements are reported in terms of years “before present” (BP), on the
assumption that the atmospheric 14C concentration has always been the same as it was in
1950 and that the half-life of 14C is 5568 years. Hence, the term “before present” refers
to any age before 1950 AD (see https://c14.arch.ox.ac.uk/). If a sample is found to have
a radiocarbon concentration exactly half (i.e. percent Modern Carbon (pMC) = 50) of
that for material which was modern in 1950 (pMC= 100) the radiocarbon measurement
would be reported as 5568 BP. These ages cannot be directly converted to calendar
years, however, because the proportion of radiocarbon in the atmosphere has varied by a
few percent over time. To see what a radiocarbon determination means in terms of a
true age, the atmospheric concentration over time must be established. This information
is garnered from thousands of 14C measurements on tree rings and other samples of
known age (e.g. speleothems, marine corals or samples from sedimentary records with
independent dating) which are then compiled into calibration curves by the IntCal group
(see Reimer et al., 2013). These curves are the basis for calibrations performed by
programs such as CALIB and OxCal, among others (see https://c14.arch.ox.ac.uk/).To
calibrate (convert) the radiocarbon ages in this report to calendar years, we have
employed the SHCAL13 calibration curve which is specific for terrestrial samples from
the southern hemisphere and spans the last 45,000 years (Hogg et al., 2013).
Annex IV
19
14Radiocarbon analysis 4.3.3Figure 5. Photographs of selected samples submitted for AMS 14C dating (taken with a Nikon D50 SLR camera coupled to a binocular microscope). Samples A, C, E and F are plant macrofossil remains ofthe grass Calamagrostiseminens. Samples B and D areorganic rich peat fragments.Wood, plant macrofossils and organic-rich sediments were selected (Fig. 5) and submitted to the Center for Applied Isotope Studies (CAIS) at The University of Georgia (Athens, Georgia, USA) in December, 2016. There they were converted into graphite targets for Accelerator Mass Spectrometry (AMS) 14C dating and aliquots ofcarbon dioxide for δ13C analysis (which is used tocorrect for any fractionation –i.e.differential mass selection-that may have occurred in 14C/12C). Samples were pretreatedwith 1N HCl at 80° Celsius for 1 hour, then washed with deionized water on fiberglass filter and rinsed with diluted NaOH to remove possible contamination by humic acids. After that the sample was treated with diluted HCl again, washed with deionized water and dried at 60°C. For AMS analyses the cleaned samples were combusted at 900°C in 20
Annex IV
15
evacuated/sealed ampoules in the presence of CuO. The resulting carbon dioxide was
cryogenically purified from the other reaction products and catalytically converted to
graphite using the method of Vogel et al. (1984). Graphite 14C /12C ratios were
measured using the CAIS 0.5 MeV accelerator mass spectrometer. The sample ratios
were compared to the ratio measured from the Oxalic Acid II (specifically NBS SRM
4990C) (a 14C standard- see http://www.c14dating.com/agecalc.html). The sample
13C/12C ratios were measured separately using an isotope ratio mass spectrometer
(IRMS) and expressed as δ13C with respect to the Pee Dee Belemnite Standard, with an
error of less than 0.1‰. The quoted uncalibrated dates have been given in radiocarbon
years before 1950 (years BP), using the 14C half-life of 5568 years. The error is quoted
as one standard deviation and reflects both statistical and experimental errors. The dates
for all samples have been corrected for natural isotope fractionation except the standards
and background targets, which have been reported as δ13C= -25‰.
We used the Oxcal radiocarbon calibration and analysis software (version 4.2.4) to
generate a median (or midpoint) and a 2-sigma probability (95%) range for each 14C
date in calendar years (Table 1). To calibrate radiocarbon ages into calendar years, we
used the calibration curve specifically created for the Southern Hemisphere, termed
“SHCal13” (Hogg et al. 2013). Nevertheless, those 14C dates with more than 100%
modern carbon (i.e. pMC >100) are considered “post-bomb” dates (i.e. the organic
tissues preserved formed after the proliferation of atmospheric nuclear detonations,
which greatly increased the amount of 14C in the atmosphere during the 1950s). These
were calibrated using the Bomb13 SH12 curve (Hua et al., 2013) by converting pMC
into F14C (or fraction modern carbon), as recommended by Reimer et al. (2004).
5 RESULTS
5.1 Chronology
The chronological framework was elaborated from a total of 12 samples of either wood,
plant leaf remains (macros), or organic-rich (peat) sediments selected for AMS 14C
dating at The University of Georgia (Table 1). The samples collected in the
palaeowetland deposits (P03-173 cm) contained thin mats of broken to complete plant
fragments, most readily identifiable as Calamagrostis eminens grasses (Fig. 5) that are
not affected by 14C reservoir effects (i.e. all of their carbon is atmospheric in origin) and
thus suitable for precise dating. No identifiable plant macrofossils were available for
dating in the peat samples we collected and dated. In such cases, we cannot preclude the
possibility that some non-plant material is subject to 14C hard-water effects, since
groundwater discharge (which in Chile often contains CO2 of volcanic origin that
Annex IV
21
16contains no 14C- see Geyh et al., 1999) feeds these depositional systems. Calibrationshows that median ages span from 8390 to recent (post-bomb) for the sediments described here (Figs. 6, 7; Table 1). Calibration of radiocarbon dates from sediments forming the T1 terrace indicate that these are “post-bomb” (i.e. contain an excess of 14Ccompared to 1950 AD) and thus this terrace was deposited in the late 20th century (seeFigs. 2, 3, 4 and 7). 22
Annex IV
17
UGAMS Sample Id 14C age Std. δ13C, Median Max. Age Min. Age Material
No. (SIL16) (yrs BP) Dev ‰ (cal yrs BP) (cal yrs BP) (cal yrs BP) dated
27789 P01-80cm 5960 25 -26.7 6740 6850 6660 Leaves
27790 P01-50cm 2040 25 -27.9 1960 2010 1900 Plant
macros
27791 P02-95cm 4200 25 -26.0 4700 4830 4580 Leaves
27792 P02-30cm 99.95 pMC 20 -26.6 -5(1955 -4(1954 AD) -6(1956 Plant
AD) AD) macros
27793 P03-5cm 7630 25 -25.8 8390 8430 8350 Peat
27794 P03-20cm 5010 25 -26.7 5690 5850 5610 Plant
macros
27795 P03-65cm 4010 25 -26.7 4440 4520 4300 Plant
macros
27796 P03- 149.03 - -23.4 -21(1971 -13(1963AD) -22(1972 Wood
302cm pMC AD) AD)
27797 P03- 560 20 -27.2 530 550 510 Plant
173cm macros
27798 P04-40cm 124.95 - -28.5 -32(1982 -12(1962 AD) -33(1983 Plant
pMC AD) AD) macros
27799 P05-15cm 110.56 - -23.4 -48(1998 -8(1958 AD) -50(2000 Plant
pMC AD) AD) macros
27800 P02-95cm 440 20 -26.5 480 510 340 Peat
Table 1. Radiocarbon dates used in this report. Calibration was done using OxCal4.2.4 with the
SHCAL13 curve (Hogg et al., 2013) (pMC = percent Modern Carbon).
Annex IV
23
18Figure 6. Left panel: SHCAL13 calibration curve (blue) plotted against all seven 14C dates older than 200 14C yrs BP. Vertical and horizontal lines are the associated error bars at two standard deviations for each sample. Right panel: Individual distributions for each 14C date (red) plotted against the SHCAL13 curve (blue) and the resulting calibrated distribution (gray).
24
Annex IV
19
Figure 7. Post-bomb 14C dates (in red), calibration curve SH1-2 (blue) and resulting intercept
intervals of calibrated dates (grey distributions). Two dates are possible but sample numbers
with the highest (>90%) probability are in Table 1. P02-30cm- 14C date 99.95 pMC (post 1954)
is calibrated to 1954-1956 (95,4 % probability); P03-302cm- 1970-1972 (91,4 % probability);
P04-40cm- 1981-1983 (79,4 % probability) and 1962-1963 (16.0 % probability); P05-15cm-
1996-2000 (89,7 % probability).
Stratigraphy, S 5.2 edimentary Units and Depositional Environments
Four main units (from oldest to youngest termed U1, U2, U3 and U4) have been defined
from five sedimentary sequences described here from the Silala River ravine (Figs. 2,
8). Table 2 summarizes the main characteristics of these sedimentary Units. The general
stratigraphy for the upper two units of section SIL16-P04 is similar to those present at
SIL16-P01, SIL16-P02 and SIL16-P05 (Fig. 8A-C) suggesting that the stratigraphy of
these sequences is representative of the sedimentary infill in all areas of the ravine.
Annex IV
25
Figure 8. Examples of stratigraphic sections discussed with associated 14C ages. A) Section SIL16-P04 showing exposed sands and gravels of Unit 4 lying unconformably on top of peat from Unit 1, B) Trenching of section SIL16-P01 down to a depth of 80 cm exposes both Unit 4 sands and gravels and Unit 1 peats, C) Section SIL16-P02 showing the surface of terrace T1 (notice dried vegetation) and good exposure of Units 4 and 1, D) section SIL16-P03 showing good exposure of Unit 3 (a massive, pedogenically altered mudflow) and part of Unit 2. The oldest and thickest section described here (site SIL16-P03, Fig. 9D) indicates that sediments accumulated to approximately 4 m above the current river level before being incised to their current level. After this incision, accumulation resumed with the deposition of ~1.8 m of sediment, which corresponds to Unit 4. These Units represent widespread Holocene wetland deposition of organic peat, black mats and silty-sand channels. Unit 2 was only found at site SIL16-P03 (Fig. 9D). Unit 3 corresponds to the abrupt deposition of a mudflow (possibly originating laterally from Quebrada Negra) and lies unconformably on top of Unit 2 (Fig. 9D). 26
Annex IV
21
Sedimentological des 5.2.1 cription of stratigraphic units
Unit 1 (ca. 8,400 to 1,900 yrs BP) is a dark brown peat (high organic matter content)
intermixed with massive, medium to coarse sand and abundant plant macrofossils, such
as leaves, terrestrial roots and wetland vegetation (Figs. 5, 9). This unit was observed in
several of our sections, including SIL16-P01 (from 80 to 20 cm depth), SIL16-P02
(from 120 to 50 cm depth) and SIL16-P03 (from 310 to 240 cm depth) (Fig. 10). We did
not reach the base of this unit (i.e. which would be at the contact with a yet undescribed
unit or with bedrock), so its total thickness and oldest age for onset of deposition is
unknown, but would presumably postdate terrace T2. The age of this unit is based on
six AMS 14C dates, three samples from SIL16-P03, one from SIL16-P02 and two from
SIL16-P01 section (Table 1). Coarse sediments include gravel-size clasts in a sandy silt
matrix with some reworked fine sediments indicative of a high to moderate transport
energy environment. The most probable explanation for such deposits is that they were
formed due to an elevated water table, which would have promoted the fine sediment
infill of the Silala channel with organic peat, black mats and silty-sand channels. Similar
modern analogues for this unit can be found throughout the central Andes and Altiplano
and consist of in-stream or flooded wetlands (known locally as bofedales) with cushion
plants (e.g. Oxychloe andina) and grasses such as Calamagrostis eminens.
Unit 2 (younger than 1900 to 530 yrs BP) starts with the deposition of median to coarse
light brown sand intercalated by coarse brown to light grey sand with sub-angular
gravels. The unit lies unconformably on top of Unit 1, indicating some degree of
incision before the deposition of Unit 2 initiated with the deposition of sand interbedded
with fine gravels. Ripple cross-bedding was observed in the sand deposits, indicative of
aggradation and river flooding (Boggs, 2006). The exact age for the onset of Unit 2
deposition is unknown, but must have occurred sometime after 1,900 years ago. The
uppermost 20 cm of Unit 2 are composed of interbedded massive silts and up to nine
black mats (thin, 1-5 cm organic rich layers) with laterally non-continuous sand lenses
(Fig. 9C). This interval shows some horizontal variability along the exposed profile,
with convoluted grading to parallel lamination (Fig. 8C). These sediments are
characteristic of in-stream wetland deposits (see Pigati et al., 2014) that likely formed
during a phase of elevated (+3 m above the current river level) groundwater table. 14C
dating show ages of 530 yrs ago for these wetland deposits. The very top of Unit 2 is
eroded and overlain by Unit 3 (Fig. 9B). This implies that the youngest ages are
maximums and Unit 2 could be slightly younger than that indicated by our ages.
Unit 3 (younger than 530 yrs BP) has been found only at site SIL16-P03 and consists of
a 120-cm thick massive light grey, fine to medium matrix-supported and poorly sorted
gravels. This unit contains abundant modern roots and has been lightly altered by
Annex IV
27
22 pedogenesis, including the formation of rhizoliths and Stage I coatings of carbonates on individual clasts (see Burgener et al., 2016). The lowermost 55 cm of Unit 3 has accumulated some calcium carbonate but is yet too incipient to form a calcrete soil horizon (i.e. a petrocalcic or “Bkm” soil horizon, see Shaetzl and Anderson, 2005) (Fig. 9B). We interpret Unit 3 as having formed by massive deposition of a mudflow sometime after 530 yrs BP. The extensive pedogenic alteration implies that Unit 3 cannot be very recent. Unit 4 (ca. 200 yrs BP to recent) consists of fine sub-rounded gravels interbedded with loamy sand. The unit has abundant plant remains of the waiya grass C. eminens and clumps of dried grasses can often be seen on the surface of this unit. Section SIL16-P04 shows gravels and sands from 60 to 40 cm depth followed by organic-rich sand and clay with plant remains and modern roots between 40-20 cm depth. The uppermost 20 cm is composed of dried remains of grasses with brown to grey organic rich loam and can be capped by a thin layer of calcium carbonate crust. For sections SIL16-P01 and SIL16-P02, Unit 4 is composed mostly of massive light brown fine sand with interspersed angular clasts and abundant modern roots. Where described, Unit 4 sits unconformably atop Unit 1 in all our sections (Figure 10). This implies that an episode of massive incision (several metres) occurred in the Silala valley after the deposition of Unit 3. The incision is capped by a final phase of aggradation that possibly began in the early 1800s. High energy fluvial environments (Unit 4 contains rip-up clasts of peaty clay, likely from Unit 1), were followed by a final phase of finer fluvial aggradation forming the historical riverbed (T1 terrace). Four 14C dates on plant remains from this unit show that it formed until very recently (as recent as 2000 AD), although the onset of deposition is not well-constrained (but cannot be older than Unit 3). The dried-out surface of Unit 4 is equivalent to the T1 terrace (Fig. 4) and lies ~2 m above the current river level. 28
Annex IV
23
Figure 9. Details from section SIL16-P03. A) Exposed section along the south bank of the Silala
River (in blue). View to the northeast, B) A view of the dugout section, showing Unit 1 (highly
Annex IV
29
24 indurated at the base), and Unit 2. Shovel is 90 cm in length, C) detail of laminated silts, sands and black mats with abundant plant macrofossils, D) detail of base of the section, showing the contact between Unit 1 peats and Unit 2 gravels and sands, E) detail of clayey peat samples from the base of Unit 1. Table 2. Summary of Stratigraphic Units and ages described for in-stream sedimentary infill of the Silala River. Unit Sedimentological properties Age (calibrated 14C yrs BP) Depositional environment Unit 4 Sandy loam with large angular clasts. Forms the recent T1 terrace ca. 200 cal yr BP to recent Modern fluvial environment surrounded by waiya grasses Unit 3 Massive matrix-supported gravels, heavily altered by pedogenesis <530 cal yr BP Mudflow Unit 2 Massive coarse to medium sands, interbedded gravels, uppermost 20 cm with interbedded laminated silts and black mats <1900-530 cal yr BP Extensive river floods that evolved into a short-lived wetland Unit 1 Dark brown clayey peat, interbedded sand lenses, abundant plant remains > ca. 8400-1900 cal yr BP Bofedal, a high altitude wetland with elevated groundwater table 30
Annex IV
25
Figure 10. Summary diagram showing the location of described and dated sections along the Silala River from the
international border (at right) down to Inacaliri (at left). Inferred and apparent unconformities are shown by pink
wavy line. All ages are in calibrated years before 1950 AD (cal yrs BP).
Annex IV
31
26 6 DISCUSSION Holocene sedimentary infill of the Silala River 6.1Figure 3 shows a complete sequence of events indicating how each sedimentary unit is associated with the sedimentary history of the Silala River. The evidence presented here shows that after a phase of initial incision (below the T2 terrace), fine sediments and peat began to infill the Silala ravine by at least 8350-8430 years ago. These peatlands are indicative of elevated groundwater tables and a permanently inundated valley floor (Squeo et al., 2006; Schittek, 2014). The depositional environment is most akin to a high altitude waiya wetland (bofedal) with multiple slow flowing channels. This environment persisted until sometime after 1900 years ago (the onset of erosion is not well constrained here), when the upper part of Unit 1 was eroded, most likely caused by a large and abrupt drop in the groundwater table and linked to similar drops recorded at Río San Salvador and Quebrada Puripica between 1,100 to 1,300 years ago (Tully, 2010; see next section). After this phase of incision, deposition of sands and gravels resumed (Unit 2) indicative of increased energy in the environment coupled with a rising groundwater table, which reached a maximum of c. 3 m above the modern river around 500 years ago. Based on the evidence, we infer that during this period of infilling, an environment with sandy channels and high energy streams gradually infilled and evolved into a wetland. Both the higher base level and stronger seasonality in precipitation would have modified the hydrological dynamics of the river, and generated the alternating silts and blackmats present in the wetland facies of Unit 2. This was followed by a mass wasting event (most likely a mudflow) that eroded part of Unit 2 and deposited Unit 3. The deposition of this unit was then followed by a major phase of incision sometime after the deposition of Unit 3. This incision removed most of the Unit 2 sediments from the valley as well as a portion of the Unit 1 sediments and would have been caused by a major drop in the water table of >3 m. Deposition resumed as the height of the water table increased some 200 years ago based on regional climate variations (see next section). Local water tables rose to c. 1.8 m above the current level of the river throughout the 20th century and formed the T1 terrace surface. Very recent incision (within the last 20 years) of more than 1 m is visible at several sites in the valley and there are multiple areas with dried out stands of wetland grasses. Understanding the multiple possible causes for this very recent incision go beyond this report. For example, current groundwater levels in the alluvial aquifer have been found to be about 6 m below the current bed level in places (Arcadis, 2017), which is consistent with this current incision. As flows in the river have remained constant over the last 20 years, and the river has been found to have a stable armoured bed and be actively transporting
32
Annex IV
27
sediment (Mao, 2017), the current groundwater levels in the alluvial aquifer are likely to
be associated with longer term climatic changes, perhaps on centennial timescales.
Overall it can be concluded that the river continues to be geomorphologically active; the
processes that formed the recent terraces are active today.
Regional palaeoclimate 6.2 and correlation with the Silala record
Regional climate variations during the Holocene have been reconstructed from
palaeowetlands (Rech et al., 2002, 2003; Tully, 2010; Sáez et al., 2016), rodent middens
(Latorre et al., 2002; Mujica et al., 2015), lake records (Geyh et al., 1999; Grosjean et
al., 2001) and tree-rings (Morales et al., 2012). Figure 11 shows some of these
reconstructions as well as the phases of aggradation and sedimentary infill in the Silala
River. The ages reported here for Unit 1 agree well with a major wet phase during the
mid-Holocene documented from many well-dated sites throughout the region. Because
we could not date the onset of Unit 1 or Unit 2, we speculate that these must have
occurred shortly before the dated portions of these sections based on regional climate
change.
Annex IV
33
Figure 11. Records of regional hydroclimate change reconstructed from different proxies across the Antofagasta Region. All ages are in calibrated years before 1950 AD (cal yrs BP). Modified from Tully (2010). The incisions of Unit 2 and Unit 4 correlate well with tree-ring records from the Altiplano, which show intense drought in the 16th and 18th centuries (Morales et al., 2012). This could explain the large falls in the water table necessary for the large incision of Unit 2 and Unit 3. A large increase in rainfall is documented at the turn of the 19th century which lasted until the 1850s (Morales et al., 2012; Mujica et al., 2015). This wetter phase correlates well with what could be the onset of deposition of Unit 4 but more dates will be needed from the base of these sections to confirm this hypothesis. 34
Annex IV
29
7 REFERENCES
Arcadis, 2017. Detailed Hydrogeological Study of the Silala River. (Vol. 4, Annex II).
Betancourt, J. L., Latorre, C., Rech, J., Quade, J. and Rylander, K. A., 2000. A 22,000-
yr record of monsoonal precipitation from northern Chile's Atacama Desert. Science,
289, 1542-1550.
Boggs, S., 2006. Principles of Sedimentology and Stratigraphy, Fourth Edition,
Prentice-Hall, Englewood Cliffs, NJ.
Burgener, L., Huntington, K. W., Hoke, G. D., Schauer, A., Ringham, M. C., Latorre,
C., and Díaz, F. P., 2016. Variations in soil carbonate formation and seasonal bias over
>4 Km of relief in the western Andes (30° S) revealed by clumped isotope thermometry.
Earth and Planetary Science Letters, 441, 188-199.
Díaz, F.P., Latorre, C., Maldonado, A., Quade, J. and Betancourt, J.L., 2012. Rodent
middens reveal episodic, long-distance plant colonizations across the hyperarid
Atacama Desert over the last 34,000 years. Journal of Biogeography, 39, 510-525,
doi:10.1111/j.1365-2699.2011.02617.x.
Gayó, E.M., Latorre, C., Jordan, T.E., Nester, P.L., Estay, S.E., Ojeda, K.F. and
Santoro, C.M., 2012a. Late quaternary hydrological and ecological changes in the
hyperarid core of the Northern Atacama Desert (~21˚S). Earth-Science Reviews, 113,
120-140.
Gayó, E.M., Latorre, C., Santoro, C.M., Maldonado, A. and De Pol-Holz, R., 2012b.
Hydroclimate variability in the low-elevation Atacama Desert over the last 2500 years.
Climate of the Past, 8, 287-306, doi:10.5194/cpd-8-287-2012.
Geyh, M., Grosjean, M., Núnez, L.A., Schotterer, U., 1999. Radiocarbon reservoir
effect and the timing of the Late-Glacial Early Holocene humid phase in the Atacama
Desert, Northern Chile. Quaternary Research, 52, 143-153.
Grosjean, M., van Leeuwen, J.F.N., van der Knaap, W.O., Geyh, M.A., Ammann, B.,
Tanner, W., Messerli, B., Núñez, L., Valero-Garcés, B.L., Veit, H., 2001. A 22 000 14C
year BP sediment and pollen record of climate change from Laguna Miscanti (238 S),
Northern Chile. Global and Planetary Change, 28, 35–51.
Hogg, A. G., Hua, Q., Blackwell, P. G., Niu, M., Buck, C. E., Guilderson, T. P.,
Thomas, P., Heaton, T. J., Palmer, J. G., Reimer, P. J., Reimer, R. W., Turney, C. S. and
Zimmerman R. H., 2013. SHCal13 southern hemisphere calibration, 0–50,000 years Cal
BP. Radiocarbon, 55 (4), 1889-1903.
Hua, Q., Barbetti, M., and Rakowski, A. Z., 2013. Atmospheric Radiocarbon for the
Period 1950–2010. Radiocarbon, 55(4), 2059-2072.
Annex IV
35
Klappa, Colin F, 1980. Rhizoliths in terrestrial carbonates: classification, recognition, genesis and significance. Sedimentology, 27, 613–629. Latorre, C., Betancourt, J. L., Rylander, K. A. and Quade, J., 2002. Vegetation invasions into absolute desert: A 45,000-yr rodent midden record from the Calama-Salar de Atacama basins, northern Chile (22-24° S). Geological Society of America Bulletin, 114, 349-366. Latorre, C., Betancourt, J., Rylander, K. A., Quade, J. and Matthei, O., 2003. A vegetation history from the arid prepuna of northern Chile (22-23° S) over the last 13,500 years. Paleogeography, Paleoclimatology, Paleoecology, 194, 223-246. Latorre, C., Betancourt, J. L., Rech, J. A., Quade, J., Holmgren, C., Placzek, C., Maldonado, A., Vuille, M. and Rylander, K., 2005. Late quaternary history of the Atacama Desert. 23°S: The Archaeology and Environmental History of the Southern Deserts, 73-90. Latorre, C., Betancourt, J.L., and Arroyo, M.T.K., 2006. Late quaternary vegetation and climate history of a perennial river canyon in the Río Salado basin (22ºS) of northern Chile. Quaternary Research, 65, 450-466. Lima, M., Christie, D. A., Santoro, M. C., and Latorre, C., 2016. Coupled socio-environmental changes triggered indigenous Aymara depopulation of the semiarid Andes of Tarapacá-Chile during the late 19th-20th centuries. PloS one, 11(8), e0160580. Maldonado, A., Betancourt, J.L., Latorre, C. and Villagrán, C., 2005. Pollen analyses from a 50,000-yr rodent midden series in the southern Atacama Desert (25º30'S). Journal of Quaternary Science, 20(5), 493-507. Mao, L., 2017. Fluvial Geomorphology of the Silala River, Second Region, Chile. (Vol. 5, Annex V). Morales, M. S., Christie, D. A., Villalba, R., Argollo, J., Pacajes, J., Silva, J. S., Alvarez, C. A. and Solíz, C., 2012. Precipitation changes in the South American Altiplano since 1300AD reconstructed by tree-rings. Climate of the Past Discussions, 8, 653-666. Mujica, M.I., Latorre, C., Maldonado, A., González-Silvestre, L., Pinto, R., De Pol-Holz, R. and Santoro, C.M., 2015. Late quaternary climate change, relict populations and present-day refugia in the northern Atacama Desert: A case study from Quebrada La Higuera (18°S). Journal of Biogeography, 42, 76–88, DOI:10.1111/jbi.12383. Muñoz, J. F., Suárez, F., Fernández, B., Maass, T. Hydrology of the Silala River Basin. (Vol. 5, Annex VII). 36
Annex IV
31
Nester, P.L., Gayó, E.M., Latorre, C., Jordan, T.E. and Blanco, N., 2007. Perennial
stream discharge in the hyperarid Atacama Desert of northern Chile during the latest
Pleistocene. Proceedings of the National Academy of Sciences USA, 104(50), 19724–
19729.
Pigati, J.S., Rech, J., Quade, J., Bright, J., 2014. Desert wetlands in the geologic record,
Earth-Science Reviews, 132, 67–81, doi:10.1016/j.earscirev.2014.02.001.
Quade, J., Rech, J.A., Betancourt, J.L., Latorre, C., Quade, B., Rylander, K.A. and
Fisher, T., 2008. Paleowetlands and regional climate change in the central Atacama
Desert, northern Chile. Quaternary Research, 69, 343–360.
Rech, J.A., Quade, J., Betancourt, J.L., 2002. Late quaternary paleohydrology of the
central Atacama Desert (22º-24°S). Chile. Geol. Soc. Am., 114, 334-348.
Rech, J. A., Pigati, J. S., Quade, J., and Betancourt, J. L., 2003. Re-evaluation of mid-
Holocene deposits at Quebrada Puripica, northern Chile. Paleogeography,
Palaeoclimatology, Palaeoecology, 194(1), 207-222.
Reimer, P. J., Baillie, M. GL., Bard, E., Bayliss, A., Beck, J. W., Bertrand, C. JH.,
Blackwell, P. G., Buck, C. E., Burr, G. S., Cutler, K. B., and Damon, P. E., 2004.
IntCal04 terrestrial radiocarbon age calibration, 0–26 cal kyr BP, Radiocarbon, 46(3),
1029-1058.
Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G., Bronk Ramsey, C.,
Grootes, P. M., Guilderson, T. P., Haflidason, H., Hajdas, I., HattŽ, C., Heaton, T. J.,
Hoffmann, D. L., Hogg, A. G., Hughen, K. A., Kaiser, K. F., Kromer, B., Manning, S.
W., Niu, M., Reimer, R. W., Richards, D. A., Scott, E. M., Southon, J. R., Staff, R. A.,
Turney, C. S. M., and van der Plicht, J., 2013. IntCal13 and marine13 radiocarbon age
calibration curves 0-50,000 years cal BP. Radiocarbon, 55(4).
Sáez, A., Godfrey, L. V., Herrera, C., Chong, G., and Pueyo, J. J., 2016. Timing of wet
episodes in Atacama Desert over the last 15 ka. The groundwater discharge deposits
(GWD) from Domeyko range at 25° S. Quaternary Science Reviews, 145, 82-93.
SERNAGEOMIN, 2017. Geology of the Silala River Basin. (Vol. 5, Annex VIII).
Schittek, K., 2014. Cushion peatlands in the high Andes of northwestern Argentina as
archives for paleoenvironmental research. Dissertationes Botanicae, 412, 176p.
Shaetzl, R. and Anderson, S., 2005. Soils: Genesis and Geomorphology, Cambridge
University Press, 821 p.
Squeo, F. A., Warner, B. G., Aravena, R. and Espinoza, D., 2006. Bofedales: high
altitude peatlands of the central Andes. Revista Chilena de Historia Natural, 79 (2),
245-255.
Annex IV
37
32 Taylor, R. E., 2014. Radiocarbon dating in archaeology. In: Encyclopedia of Global Archaeology, Springer, New York, pp. 6226-6235. Tully, C., 2010. Holocene mega-droughts in the central Atacama Desert, Chile. M.Sc. Thesis, Miami University, Oxford, Ohio, USA. 53 pp. Available at: http://rave.ohiolink.edu/etdc/view?acc_num=miami1272312883. Vogel, J. S., Southon, J. R., Nelson, D. E., and Brown, T. A., 1984. Performance of catalytically condensed carbon for use in accelerator mass spectrometry. Nuclear Instruments and Methods in Physics Research Section B5: Beam Interactions with Materials and Atoms, 5(2), 289-293. 38
Annex IV
Annex V
Mao, L., 2017. Fluvial Geomorphology of the Silala River, Second Region, Chile
39
40
Annex V
FLUVIAL GEOMORPHOLOGY OF THE SILALA RIVER,
SECOND REGION, CHILE
Luca Mao (PhD)
Associate Professor, Pontificia Universidad Católica de Chile
Research Assistants: Ricardo Carrillo, Joaquín Lobato, Silvia Arcandi
May, 2017
Annex V
41
GLOSSARY Partially based on the glossary of terms of Charlton (2008). Aggradation: (as a fluvial process) A net accumulation of sediment, which leads to an increase in the elevation of channel beds and floodplain surfaces. Alluvial Reach: A river flowing on a channel formed in alluvium, i.e. flowing on sediments deposited by fluvial processes. Aquifer: Geological formation capable of storing, transmitting and yielding exploitable quantities of water. Armour Layer: A layer of coarse sediment that forms in mixed gravel-bed rivers and which covers the finer underlying subsurface bed material. Bed Load: The coarser fraction of the sediments transported downstream by a river, which is moved by traction along the bed of the channel by rolling, sliding and saltation. Bed Shear Stress: A shearing force exerted by flowing water on the bed of the channel. Cascade Morphology: Reach characterized by a disorganized disposition of coarse material. Found in steep, upland channel reaches. Channel Bed: The shape and material that forms the river lower boundary of a river, stream or artificial channel. Colluvial Reach: A channel whose sediments have not been deposited by fluvial processes, but rather by material that has been weathered and transported to the base of slopes by processes of mass movement and surface wash. Confinement: (in the context of a river channel) The degree to which a river channel is laterally free to migrate within the alluvial plan or is in contact with hillslopes or ancient terraces. Degradation: (in the context of a river channel) The reduction in the elevation of channel beds and floodplain surfaces due to net erosion. Discharge: Volume of water flowing per unit time, for example through a river cross-section or from a spring or a well. Drainage Network: The branching network of rivers and tributaries within a drainage basin. Ephemeral Reach: A portion of the river network in which liquid discharges occur only part of the time and which may be dry for long periods. 42
Annex V
Erosion: The action of surface processes that remove sediments from one location and
transport them it away to another location.
Geomorphology: The scientific discipline studying of the physical features of the
surface of the earth.
Groundwater: Subsurface water occupying the saturated zone (i.e. where the pore
spaces (or open fractures) of a porous medium are full of water) water held underground
in the soil or in pores and crevices in rock.
Gully: A steep-sided, trench-like feature ranging in size from tens of centimeters to
several meters deep, which is commonly associated with dryland environments.
Headwater Streams: A tributary stream of a river, close to or forming part of its
source. The upper-most smallest parts of the drainage network.
Hollow: An unchanneled depression in the basin.
Hydraulic Radius: A measure of channel hydraulic efficiency defined by the ratio
between cross-sectional area and the wetted perimeter.
Incision: The downward cutting of a channel into its substrate.
Intake: An artificial hydraulic structure aimed at diverging the flowing water from a
natural river to a side channel or pipe.
Perennial Reach: A portion of the river network in which liquid discharges flow all
year-round.
Reach: A length of river along which the main channel controls are sufficiently uniform
to allow a fairly consistent morphological structure to be maintained and thus evaluated
(its length is usually considered being 5-10 times the channel width).
Sediment: Material transported by water either in suspension or as bed load from the
place of origin to the place of deposition loose material deposited by water that makes
the channel bed.
Step/Pool Morphology: Sequences of steps formed from coarse material separated by
relatively still pools. Found in steep, upland channel reaches.
Water Stage: The height of the water level in a river.
Wetted Perimeter: The cross-sectional length of a channel boundary that is in contact
with the flow.
Annex V
43
TABLE OF CONTENTS 1 INTRODUCTION ..................................................................................................... 1 2 SUMMARY AND CONCLUSIONS ........................................................................ 4 3 BACKGROUND ....................................................................................................... 5 3.1 Fluvial processes ................................................................................................ 5 3.2 Sediment transport, incipient motion of sediments, and armouring................... 6 4 MATERIALS AND METHODS .............................................................................. 9 4.1 Longitudinal profiles and cross sections ............................................................ 9 4.2 Water stage and liquid discharge ...................................................................... 10 4.3 Surface and subsurface grain size distributions ............................................... 10 4.4 Tracers (coloured particles) and Passive Integrated Transponder (PIT) tags .. 11 4.5 Experiment with natural flow discharge .......................................................... 13 4.6 Bedload sampling ............................................................................................. 13 4.7 Morphological assessment of the Silala River ................................................. 13 4.8 Fish survey ....................................................................................................... 16 5 RESULTS ................................................................................................................ 16 5.1 Longitudinal profiles and cross sections .......................................................... 16 5.2 Water stage and liquid discharge ...................................................................... 18 5.3 Surface and subsurface grain size distributions ............................................... 19 5.4 Tracers (coloured particles) and PIT tags ......................................................... 21 5.5 Experiment with natural flow discharge .......................................................... 23 5.6 Bedload sampling ............................................................................................. 26 5.7 Morphological classification and Morphological Quality Index (MQI) of the Silala River .................................................................................................................. 27 5.8 Fish population in the Silala River ................................................................... 30 6 DISCUSSION .......................................................................................................... 32 6.1 Presence of the armour layer ............................................................................ 32 6.2 Tracers and PIT tags ......................................................................................... 34 6.3 Fish population ................................................................................................. 36 44
Annex V
6.4 River morphology............................................................................................. 36
7 FINAL REMARKS ................................................................................................. 37
8 REFERENCES ........................................................................................................ 38
Annex V
45
1 1 INTRODUCTION This report on the fluvial geomorphological forms and processes of the Silala River was requested by the Dirección Nacional de Fronteras y Límites del Estado (DIFROL) of the Republic of Chile, and was undertaken by Prof. Luca Mao, supported in the field by a team composed by Ricardo Carrillo, Joaquín Lobato, and Silvia Arcandi. Luca Mao (PhD in Environmental Watershed Management, University of Padova, Italy) is Associate Professor at the Pontificia Universidad Católica de Chile, and his field of expertise is fluvial geomorphology and ecology. This report was written by Luca Mao, under the supervision and instruction of Professors Howard Wheater and Denis Peach. The area under investigation corresponds to the Silala River basin. Figure 1 shows the general geographical location of the basin, and Figure 2 shows the Silala basin and river network. 46
Annex V
2
Figure 1. Location of the Silala basin.
Annex V
47
3 Figure 2. The Silala basin to the Codelco Intake, Inacaliri. This report on the fluvial geomorphological nature of the Silala River has three main objectives: To characterize the fluvial nature of the forms featured in the main channel ofthe Silala River;To characterize the fluvial nature of the processes ongoing in the main channelof the Silala River, namely sediment mobility and transport;To analyze the ecological characteristics of the Silala River, and particularly thefish population.48
Annex V
4
Overall, the aim of this investigation was to assess whether the main channel of the
Silala River features forms and exhibits processes that are what one would expect from
a geomorphologically active river system.
To achieve the objectives of the investigation, a series of field measurements, including
the survey of cross-sections, profiles and size of sediments, were carried out, along with
measurements of the displacement of marked sediments, and the survey of fish in the
main channel of the Silala River.
This report is structured as follows: in Chapter 2 the main conclusions are summarized;
in Chapter 3 a general background to the relationships between fluvial forms and
processes is provided, along with details of the current understanding of sediment
transport and dynamics in gravel-bed rivers; Chapter 4 is a materials and methods
chapter, in which full details of the field surveys of fluvial forms, morphology, sediment
transport and fish population are given; Chapter 5 is the results chapter, in which the
outcomes of the field surveys and analysis are provided, with support of photos, figures
and tables; in Chapter 6 the main evidence gathered from the fluvial geomorphological
and ecological activities is discussed, supported by references to state-of-the-art
literature on the topics; Chapter 7 is a short final remarks chapter, with the main
outcomes of the investigation; Chapter 8 contains the list of references cited in the
report.
2 SUMMARY AND CONCLUSIONS
This report shows that the Silala River features the main morphological characteristics
typical of a permanent river with size-selective transport of sediments, such as the
presence of a well-developed static armour layer. Field surveys of sediment transport
using traps and marked sediments (coloured and tagged clasts) revealed that finer
sediments are moved in higher percentages and for longer distances than coarser
sediment fractions, which corroborates the size-selective nature of sediment dynamics
in the river. The stream features the typical cascade-step/pool morphology that is to be
expected in an alluvial river with the given boundary conditions of slope, grain-size of
sediments, and lateral confinement. Furthermore, the morphological quality index
classifies the river as featuring overall good morphological conditions. The river hosts
an abundant population of rainbow trout in good condition, showing that the general
conditions of the river (availability of food and presence of aquatic habitats) are
perfectly able to sustain a healthy population of resident fish.
Overall, this report provides evidence that the forms and the processes ongoing in the
Silala River are typical of an alluvial stream with a permanent flow regime.
Annex V
49
5 3 BACKGROUND 3.1 Fluvial processes A river basin is the area of land drained by a river and its tributaries, which can be called the drainage network. A river basin is generally defined in terms of the surface topography, but in cases where groundwater provides important contributions to river flow, the area contributing groundwater to a river may be different from the surface topographic catchment, depending on the local hydrogeology (Leopold et al., 1964). In the upper part of the drainage network, the tributaries are generally steeper and narrower than in lowland reaches, as they receive less water, and they are usually confined by the geometry of the valleys, which are steeper in the upper part of the basin. In the very upper portion of the basins, when incised into the hillslope, headwater channels are referred as gullies, hollows, or more generally colluvial reaches (Leopold et al., 1964; Wohl, 2014). These reaches are often ephemeral or intermittent, i.e. lacking continuous flowing water. From a geomorphological point of view, colluvial reaches are usually shaped by mass wasting phenomena (i.e. debris or hyper-concentrated flows). These events, along with the lack of running water throughout the year, shape channels which have unstructured morphological units, longitudinal levees, and other features that can be easily related to mass wasting processes. In contrast, downstream reaches, where slope is shallower and discharges are higher and more regular (i.e. perennial) due to the larger basin area, are shaped by fluvial processes. Reaches of this kind can be regarded as part of a geomorphologically-active fluvial system, and are known as alluvial rivers, i.e. sediments in their channels have been transported and deposited from reaches upstream due to the action of flowing water (Church, 2006). Because of that, the geometry of these channels is self-formed and dynamic, evolving with time, and it is substantially different to the geometry of colluvial reaches. All river systems may interact with groundwater – either gaining or losing water from river channels. However, large-scale groundwater discharges are most commonly associated with alluvial river reaches. Fluvial processes in alluvial reaches are mainly associated with sediment transport dynamics, which involve the movement of coarse sediments (i.e. > 2 mm) as bedload (rolling, sliding, or bouncing over the channel bed). Sediments in the bed and on the banks of alluvial reaches are important because they determine the morphology of the channel. Overall, the morphology of an alluvial river is determined by a trade-off between the forces of erosion, i.e. the discharge that a channel conveys and the slope that defines the force that the flow exercises on the bed, and on the other hand the sediment load, determined by the size of bed sediments and the rate of sediments 50
Annex V
6
transported from upstream (Church, 2006). This complex balance determines the
morphology and dynamics of an alluvial reach. Lane (1955) conceptualized this idea as
a balance, in which a river in a state of “dynamic equilibrium” is the complex result of a
series of adjustments of its geometry (especially the slope), size of sediments in the
channel, liquid discharge (as a result of the hydrologic regime) and sediment transport
(as a result of sediment supply conditions and energy available to transport sediments)
(Figure 3). Any change in one of these control factors would result in a morphological
adjustment of the river (either as aggradation or degradation). The forms of a river are
important to understand as they control the physical evolution of the system and
constitute the physical structures that determine the presence and diversity of potential
habitats for fish and macroinvertebrates.
Figure 3. Lane’s balance, representing the dynamic equilibrium concept for fluvial
geomorphology. The balance illustrates that changes in a fluvial system depend on the
relationship between resisting forces (Qs = bed material load and D50 = median grain size of
bed sediments) and driving forces (Qw = dominant discharge and S = channel slope). Lane’s
balance of resisting vs. driving forces shows that a change in one factor causes a change in
another, resulting in degradation (i.e. vertical incision) or aggradation (i.e. sediment
deposition).
3.2 Sediment transport, incipient motion of sediments, and armouring
The evaluation and prediction of coarse sediment transport is essential for
understanding fluvial system morphology and morphodynamics. However, bed load
(sediment transported along the bed of the river, rather than in suspension) is
notoriously challenging to measure in the field (Vericat et al., 2006) and difficult to
predict within one to two orders of magnitude using available formulas (e.g. Barry et al.,
Annex V
51
7 2004). Indeed, bedload equations are usually empirical and derived from data acquired through flume experiments (e.g. Wilcock and Crowe, 2003), when several simplifications are made, especially regarding the incipient motion of sediments, the availability of sediment to be transported, and the presence and stability of armour layers (Bathurst, 2007). The process of sediment transport is challenging to study because sediment motion depends not only on the magnitude of local flow conditions, but also on the characteristics of the sediments. The process is not linear but combines factors which include bed morphology, turbulence in the water, interaction of sediment particles, availability of sediment, hydrological forces, and presence of vegetation. Bedload transport generally occurs when a certain threshold, called critical, is passed. The critical threshold is usually expressed in terms of dimensionless shear stress τ* needed to entrain a grain of a certain size. The force of flow acting upon the grain is defined as: 𝜏𝜏∗=𝜏𝜏(𝜌𝜌𝑠𝑠−𝜌𝜌)𝑔𝑔𝐷𝐷50 where 𝜌𝜌𝑠𝑠 and 𝜌𝜌 are the densities of sediment and water, respectively, 𝐷𝐷50 is the median grain size of surface sediments, and τ is the shear stress acting on the channel bed, defined as: 𝜏𝜏=𝜌𝜌𝜌 𝜌𝜌𝜌𝜌 where g is the acceleration of gravity, 𝑅 is the hydraulic radius and 𝑆 is the slope of the channel. Shields (1936) established that for fully turbulent flows the dimensionless shear stress needed to move a particle, (i.e., the so-called critical shear stress τ*c) is approximately 0.06. Shields would thus imply that a higher shear stress is needed to entrain and transport a coarser grain. In other words, a finer particle is easier to transport than a coarser one, a condition called size-selective transport (Shields, 1936). Subsequent studies highlighted certain limitations of this approach, and Parker et al. (1982) introduced the equal-mobility theory, which states that the sediment entrainment conditions are independent of grain size. Measurements in rivers at near-equilibrium conditions (Parker et al., 1982; Andrews, 1983; Marion and Weirich, 2003) show that the transport of unimodal mixed sediments is only weakly size selective at low stresses and approaches equal mobility at high stresses. Examining a wider range of flows, several studies (e.g. Ashworth and Ferguson, 1989; Kuhnle, 1992; Wathen et al., 1995) emphasized the size-selective nature of the gravel transport, which approaches equal-mobility only during the highest flows. 52
Annex V
8
Gravel bed rivers typically feature an armoured surface over an underlying, finer
subsurface mix of sediments. The coarser surface layers can be differentiated into static
and mobile armours. A static armour layer is created by a flow that selectively entrains
only the finer elements (i.e. Shields-type entrainment condition) when there is a lack of
upstream sediment supply (Proffitt and Sutherland, 1983; Chin et al., 1994; Church et
al., 1998). This is typical of river reaches located downstream of lakes or dams, or
perennial rivers with lack of abundant sediment supply. The long periods of low flow
allow sediments to re-arrange and create the armour layer of coarse grains, as finer
sediments are selectively eroded and can travel downstream (Figure 4). Conversely, in a
river with mobile armour, a lower difference in mobility between large and small grains
results in a higher percentage of larger grains exposed to the flow (Parker and
Klingeman, 1982; Parker et al., 1982). A mobile armour layer can coexist with the
transport of all available grain sizes, and its development has been explained using the
concept of equal-mobility of all grain sizes of a heterogeneous gravel bed under
unlimited sediment supply conditions (Parker and Toro Escobar, 2002).
Figure 4. An armoured bed.
Hassan et al. (2006) showed that the presence and degree of armour index (i.e. the ratio
between the surface and subsurface median size of sediments: D50/D50ss) depend on the
hydrological regime and sediment supply conditions. In ephemeral or intermittent
rivers, the presence of flash floods and abundant and continuous sediment supply
permits the formation of a mobile armour layer. Conversely, a river dominated by a
perennial regime with relatively constant discharges and low sediment supply allows the
formation of a well-developed static armour layer.
Annex V
53
9 4 MATERIALS AND METHODS The main channel of the Silala River is confined, as riparian vegetation (mainly Calamagrostis Eminens) grows on both right and left bank, stabilizing the banks and leading to a low width/water stage (i.e. water level) ratio. The survey has been carried out over two reaches: reach A and reach B, which are 175 and 140 m long, respectively, and approximately 2 km apart (Figure 5). Figure 5. Location of the study reaches A and B. 4.1 Longitudinal profiles and cross sections Longitudinal profiles and cross sections of each reach were surveyed using a laser distance meter with clinometer and a prism pole (Laser Technology Impulse). A laser rangefinder is a topographical instrument which uses a laser beam to determine the 54
Annex V
10
distance to an object by measuring the time taken by a laser pulse to be reflected off a
target (a prisma in this case) and return to the sender. The longitudinal profile was
measured taking points at every change of slope, and at least every meter, to calculate
the channel slope. Cross sections were measured in order to calculate the channel width
and depth. Four cross sections were measured on each reach. The precision of the
topographical survey is +/- 5 mm.
4.2 Water stage and liquid discharge
Water stage monitoring was conducted in reach A with a battery-powered pressure
transducer installed at the upstream end of the reach. The instrument was installed in a
plastic pipe fixed to the left bank on 2nd September 2016. Water level was recorded
every 10 minutes until 11th January 2017. Because the instrument measures the absolute
pressure (water pressure and atmospheric pressure), another device was installed nearby
in order to measure atmospheric pressure. Water stage changes are obtained after
adjusting for barometric pressure fluctuations.
A stage-discharge relationship was derived in the field by measuring on several
occasions the liquid discharge using the salt dilution method (see Moore, 2005, for
details of the method). Measurements were taken using a portable conductivity meter
(model WTW Cond304i with TetraCon 325 probes, storing data every second) and, as a
tracer, a variable quantity (0.2-0.3 kg) of salt (NaCl). The salt, dissolved in a 10 litre
bucket, was injected into the main stream approximately 15 m upstream from the probe,
to allow for adequate lateral mixing. Salt-tracer discharge measurements were repeated
in each reach over a range of discharges, with at least three repetitions of salt injections
performed in every reach for a single discharge.
4.3 Surface and subsurface grain size distributions
Sampling of bed surface material was undertaken at the two reaches using the grid-bynumber
method. The sampling procedure involved 150 grid-by-number pebble count
samples per reach (grid spacing 0.3 m) (Bunte and Abt, 2001). Grid-by-number derived
curves were truncated at 2 mm. The operation consists of walking along the channel
following an imaginary grid in the channel bed. At every step the stone at the end of the
operator’s boot was taken and measured with a calliper. Surface grain-size curves were
then derived for both sites by integrating all 150 samples.
The subsurface bed material grain-size distribution was estimated using the volume-byweight
methods (truncated at 2 mm) applied to samples taken in the two study reaches.
To sample the subsurface, the overlying surface sediment was removed, taking an areal
sample that exposes subsurface particles. The subsurface grain size does not need to be
Annex V
55
11 converted, as the volume-by-weight is comparable to the grid-by-number analysis used for the surface sampling (Bunte and Abt, 2001). The median grain size (D50) was derived from each grain-size curve, so that the armouring ratio could be calculated as the ratio between the surface (D50) and subsurface (D50ss) median grain size (i.e. D50/D50ss) of each reach. If this ratio approaches a value close to 1, the size of surface and subsurface sediments is comparable and the lack of armour is typical for watercourses with high sediment supply, whereas if the ratio has a higher value the transport capacity exceeds sediment supply and the subsurface sediment size is smaller than the surface sediment size. In some cases (high energy mountain streams or absence of sediment supply), the ratio can reach a value of 3 or higher (see Hassan et al., 2006). 4.4 Tracers (coloured particles) and Passive Integrated Transponder (PIT) tags Coloured tracers and PIT tags were used to investigate sediment transport. PIT tags use the radio-frequency identification (RFID) technique. RFID is a wireless, automatic identification system. It uses electromagnetic fields to automatically identify and track tags attached to an object. PIT tags are transmitters without batteries that emit an identification code through radio frequencies, which is detected and registered by an antenna. The antenna used to identify the PITs works up to a range of 30 cm, enough for the conditions of the site (Oregon RFID, and were available from the Pontificia Universidad Católica de Chile). The relative position of each PIT is measured with a laser distance metre with clinometer (Laser Technology Impulse, available from the Pontificia Universidad Católica de Chile) with reference to several georeferenced points along each reach. PIT tags can be detected even if they have been buried in the channel bed. Natural clasts from the Silala channel were equipped with standard, 23 mm long, PIT tags (Texas Instruments) in the laboratory. The procedure consists of drilling holes 5 mm in diameter in the clasts, inserting a PIT into the hole, applying epoxy glue and leaving the glue to dry for about one day. The three principal dimensions (a – long, b – medium, c – short) of the clasts were measured using a calliper and then the clasts were weighed to determine the relationship between b-axis diameter and weight. Overall, 136 PIT-tagged clasts were inserted in the studied reaches, with clasts ranging in size (b-axis diameter) from 28 mm to 69 mm (Figure 6). Other natural clasts were collected from the main channel, sieved in four grain size classes: 4 to 8 mm, 8 to 16 mm, 16 to 32 mm and 32 to 45 mm, and coloured using different spray paint colours. The following number of coloured tracers were inserted in the channel bed on each reach (Figure 6) at each field survey, to be searched for at the following field visit: 56
Annex V
12
 625 particles between 4 and 8 mm, spray-painted in red;
 100 particles between 8 and 16 mm, spray-painted in yellow;
 60 particles between 16 and 32 mm, spray-painted in blue;
 30 particles between 32 and 45 mm, spray-painted in green.
Tracers and PIT tags were placed on the channel bed at certain cross-sections, marked
in the field using metal or wooden sticks and flags (Figure 7). These sections were
precisely located in the measured longitudinal profiles, such that displacement lengths
of each tracers and PITs were straightforward to calculate after field surveys of
searching.
Tracers and PIT tags were placed in each reach at the first field visit (4 September
2016), and then searched for at three times, namely on the 15th October 2016, 18th
December 2016, and 10th January 2017. A further survey was conducted on the 11th
January 2017 after a field experiment with the natural discharge (see later). After each
search, new tracers were added to the cross-sections if needed, and all transported
tracers were removed in order to avoid double-counting at the following field visit.
Figure 6. Grain size distribution of the PIT tags and the coloured particles (tracers) placed in
the study reaches.
0
25
50
75
100
1.0 10.0 100.0 1000.0
% Finer than
D (mm)
Tracers
PITs
Annex V
57
13 Figure 7. Cross-sections where coloured particles of the 32 - 45 mm grain size class (on the left) and PIT tags (on the right) were placed in the channel bed. All transects with tracers and PIT tags were marked in the field using coloured flags. 4.5 Experiment with natural flow discharge Downstream of the international border, The Antofagasta (Chili) and Bolivia Railway Company Ltd. (FCAB) collects and constantly diverts to a pipe a certain amount of the liquid discharge. On the 11th January 2017, the company allowed all discharge to flow downstream, by closing the intake pipe, thus reproducing the natural flow discharge. This operation lasted for 3 hours, from 12:00 to 15:00. 4.6 Bedload sampling We collected direct measurements of bedload transport rates using a Bunte-type trap, with 1mm mesh size (Bunte et al., 2007). The bedload trap was installed at the centre of the river. During the four field visits with low flow, the bedload sampling lasted 2 to 3 days because the liquid discharge and sediment transport rate were very low, whereas during the natural flow discharge experiments, three samples were collected, and each sampling lasted 10 minutes. Bedload samples were then taken to the laboratory, where they were dried, sieved and weighed. The grain size distribution and the bedload transport rate (in g m-1 s-1) from each sample were then calculated. Overall, four and three samples were taken during the low and the high flow conditions, respectively. 4.7 Morphological assessment of the Silala River A systematic classification of river systems is useful for the prediction of their behaviour (processes) from their appearance (form), for the development of specific hydraulic and sediment relations for a given morphological channel type and state, and 58
Annex V
14
to provide a mechanism to refer one river to another of similar character. There is a
plethora of classifications for forms and morphological quality of river systems. In this
case we used the Rosgen (1994) classification, commonly applied in North America,
and the Morphological Quality Index (MQI) (Rinaldi et al., 2013), which was developed
in Italy and now modified and extensively used in Europe. Although this index has been
developed for the Italian context, it has been applied successfully to other rivers in
Europe, and in absence of an appropriate approach proposed for the Andean region, it is
considered to provide good insight on the hydromorphological state of the Silala River.
Rosgen (1994) proposed a classification for natural rivers evolved from field
observation of hundreds of rivers of various sizes in all the climatic regions of North
America. The Rosgen stream classification was basically developed to predict a river’s
behaviour from its appearance, to develop specific hydraulic and sediment relationships
for a given stream type, and to provide a consistent frame of reference for
communicating stream morphology and condition among a variety of disciplines. The
classification involves three stages and is based on discrete combinations:
 The Level I classification is based on the stream characteristics that result from
relief, landform, and valley morphology (Figure 8);
 The Valley type, a primary determinant of stream form;
 The Level II classification, which provides more detailed morphological
description of stream type from field measurements of channel form and bed
composition.
Level I combines the influences of climate, depositional history and life zones (desert
shrub, alpine, etc.) on channel morphology (Rosgen, 1994). Nine stream types are
defined in this first level (Figure 8). Then, the identification of valley types and related
landforms provides the basis for an initial indication of river morphology. In Level II
classification, stream types are divided into discrete slope ranges and dominant channelmaterial
particle sizes. The parameters used in this section are ranges in slope,
width/depth ratio, entrenchment ratio and sinuosity.
Annex V
59
15 Figure 8. The first level of classification of the Rosgen-type of rivers, which depends on the longitudinal profile, shape of cross-sections, and plan view (reproduced with permission from Rosgen, 1994). The MQI (Rinaldi et al., 2013) is used, through a process of evaluation and monitoring of the fluvial system, to evaluate the deviation of present conditions of a river from a given reference state, considered as the potential conditions of the river in the absence of human disturbance. The analysis for the evaluation of the MQI includes the evaluation of functionality and artificiality of the channel, and past channel changes. There are questions related to each indicator, and the result for every question varies from A (score 0) to C (maximum score for that indicator). The final score represents the total deviation STOT from non-altered conditions and is calculated as the sum of the scores assigned to all indicators. Then a Morphological Alteration Index (MAI) is defined as complementary to the previous result: 𝑀𝑀𝑀𝑀𝑀𝑀=𝑆𝑆𝑇𝑇𝑇𝑇𝑇𝑇/𝑆𝑆𝑀𝑀𝑀𝑀𝑀𝑀 where SMAX is the maximum possible deviation for the given stream typology (sum of all the C answers). The Morphology Quality Index is eventually defined as: 𝑀𝑀𝑀𝑀𝑀𝑀=1−𝑀𝑀𝑀𝑀𝑀𝑀 60
Annex V
16
MQI assumes a minimum value of 0 and a maximum value of 1. The five classes are:
1. Very good or high. MQI>0.85
2. Good: MQI=0.7/0.85
3. Moderate: MQI=0.5/0.7
4. Poor: MQI=0.3/0.5
5. Very poor or bad: MQI=0/0.3
4.8 Fish survey
Fish communities were sampled on reaches A and B using upstream, double-pass
electrofishing (HT-2000 Battery Backpack Electrofisher, Halltech Aquatic Research,
Inc.) on two occasions: 2 September and 19 December 2016. All fish were identified to
species level, measured for total length (mm), and weighed (g) before being returned to
the river.
5 RESULTS
5.1 Longitudinal profiles and cross sections
Figure 9 shows the longitudinal profiles of reaches A and B. Channel slope is calculated
as the ratio between the difference in elevations and the horizontal distance. The
channel slope is 0.057 m m-1 for reach A and 0.047 m m-1 for reach B.
Figure 9. Longitudinal profiles of Reach A (left) and Reach B (right).
0
2
4
6
8
10
12
0 50 100 150 200
Elevation (m)
Distance (m)
0
1
2
3
4
5
6
7
0 50 100 150
Elevation (m)
Distance (m)
Annex V
61
17 Four cross sections were measured in each reach. In Figure 10 are shown two cross sections representative of the channel geometry of the two reaches. Figure 10. Representative cross-sections for Reach A (left) and Reach B (right). The blue lines represent the water surface at the time of the survey (approximately 21 l s-1). Hydraulic parameters were derived for each cross-section during mean flow. In cross-section A, wetted area is 0.078 m2, channel width is 0.89 m, wetted perimeter (length of the wetted portion of cross-section) is 0.988, and mean channel depth is 0.088 m. Hydraulic radius, calculated as ratio between wetted area and wetted perimeter, is 0.079 m and confinement ratio, calculated as ratio between channel width and mean channel depth is 0.1. The shear stress acting on the channel bed at low discharge can be thus calculated as: 𝜏𝜏=𝜌𝜌𝜌𝜌𝜌𝜌𝑅𝑅ℎ=44 𝑁𝑁 𝑚𝑚−2 where  is water density, g is the gravitational force, S is the channel slope (0.056 m m-1 for reach A) and Rh is the hydraulic radius. In cross section B, wetted area is 0.25 m2, channel width is 1.48 m, wetted perimeter is 1.722 and mean channel depth is 0.17 m. Hydraulic radius, calculated as above for reach A, is 0.148 m. The confinement ratio (channel width/mean channel depth) is 0.12. With this channel geometry, and considering a slope of 0.047 m m-1, the shear stress acting on the channel bed at low discharge is about: 𝜏𝜏=𝜌𝜌𝜌𝜌𝜌𝜌𝑅𝑅ℎ=68 𝑁𝑁 𝑚𝑚−2 Although reach B features a much higher hydraulic radius due to the fact that the mean discharge is three times higher than for reach A (see next chapter), the shear stress is comparable to that calculated for reach A, because the mean channel slope is higher in 0.00.20.40.601234Elevation (
m)Distance (m)0.00.20.40.60.801234Elevation (
m)Distance (m)62
Annex V
18
reach A. The fact that reaches A and B feature a similar shear stress shows that the two
reaches are comparable in this study.
5.2 Water stage and liquid discharge
Water stage and water temperature were monitored continuously in reach A for four
months (Figure 11). Results show that the river has a permanent flow regime, as there is
always at least 0.15 m of water stage in the monitored cross-section, even if daily
fluctuations of discharge are evident. Indeed, the water stage nearly doubled during the
day for a few hours in September, but the difference between minimum and maximum
daily water stage was more reduced from October (Figure 11). In the water stage
profile, anomalous oscillations and peaks (especially in November) are likely to be due
to artificial hydraulic operations upstream (cleansing of water tank and experimental
aquifer pumping test operations). The very high water stage measured on the 11th
January 2017 corresponds to the experiment with the natural flow discharge (see later),
where the water stage at reach A reached 0.4 m.
Figure 11. Record of water stage and water temperature from the sensor installed in reach A.
Discharge was measured in Reach A on four occasions during low flow (with 3
repetitions each), with a range of discharge ranging from 18.3 l s-1 to 22.6 l s-1.
Furthermore, during the experiment with the natural flow discharge, a further 11
measurements were taken, with discharge up to 100 l s-1. With these measurements, and
the water stage profile, a stage-discharge relationship can be calculated as a regression
from the measured points (Figure 12).
0.15
0.2
0.25
0.3
0.35
0.4
0.45
6
8
10
12
14
16
18
02-09-2016 22-09-2016 12-10-2016 01-11-2016 21-11-2016 11-12-2016 31-12-2016
Water stage (m)
Temperature (°C)
Annex V
63
19 Figure 12. Stage-discharge relationship for reach A. A continuous discharge profile can be thus calculated from the water stage-liquid discharge relationship converting water stage measured in A to discharge (Figure 13). Figure 13. Record of liquid discharge for reach A.Discharge was also measured for Reach B on six occasions during low flow (with 3 repetitions each), for a range of discharges from 54 l s-1 to 58 l s-1, so liquid discharge in reach B is roughly triple the discharge measured at reach A. 5.3 Surface and subsurface grain size distributions Results for grain size analysis of surface and subsurface are shown in Figure 14 (for reach A) and Figure 15 (for reach B). y = 499.01x -60.206R² = 0.97220204060801001200.150.200.250.300.35Liquid discharge (
l
s-1)Water stage on reach A (m)02040608010012014002-09-201622-09-201612-10-201601-11-201621-11-201611-12-201631-12-2016Liquid discharge (
l
s-1)64
Annex V
20
Figure 14. Surface (Surf) and subsurface (Sub) grain size distribution obtained for reach A.
Figure 15. Surface (Surf) and subsurface (Sub) grain size distribution obtained for reach B.
Percentiles and standard deviation of the grain size curve are calculated for reach A and
B for surface and subsurface samples (Table 1). D16, D50, D84 identify the size of the
sediments (in mm) corresponding to percentiles 16, 50, and 84% of the cumulative
distribution-curves, i.e. the sediment size for which a certain percentage (for example
50) of the sediment sample is finer is the “50th percentile”.
Surface A Subsurface A Surface B Subsurface B
D16 (mm) 6 4 6 4
D50 (mm) 30 12 36 18
D84 (mm) 102 31 99 42
σ (-) 4.0 2.9 4.0 3.1
Table 1. Percentiles (D16, D50, and D84) and standard deviation (σ) of the grain size curve.
0
25
50
75
100
1 10 100 1000
% Finer than
D (mm)
SubA
SurfA
0
25
50
75
100
1 10 100 1000
% Finer than
D (mm)
SubB
SurfB
Annex V
65
21 Surface grain size distributions of reach A and B are similar, whereas the subsurface grain size distribution of reach B is slightly coarser than A. The armour ratio is calculated as the ratio of surface median to subsurface median size, or D50/D50SS, where D50 identifies the median grain size of the surface material and D50SS denotes the median grain size of the subsurface material. An essentially unarmoured stream is one for which the armour ratio is near unity, whereas armouring can be considered well-developed at ratios higher than 2. The armour ratio in the Silala River is 2.5 for reach A and 2 for reach B. 5.4 Tracers (coloured particles) and PIT tags The displacement length of painted particles and PITs was surveyed on the 15th October 2016, 18th December 2016, and 10th January 2017. Figure 16 shows the transects where the tracers were placed and the location of all transported clasts. Results of recovery rate (i.e. percentage of sediments that were found at the following search) and movement rate (i.e. the percentage of sediments that moved) are presented for reach A in Figures 17 and 18 (figures for reach B would be similar and are not shown). As expected, the recovery rate increases with the size of tracers, because finer particles are more likely to be transported far downstream of the surveyed reach, and because transported particles are likely to be transported and buried within the same reach. Coarser particles are easier to recover because they are unlikely to be buried completely and are transported for shorter distances. The percentage of particles found downstream of the exact sites where they were placed in the channel bed increases for the coarsest particles. This is because finer particles are easier to be entrained and transported downstream than coarser particles. Since PIT tags are detected using a portable antenna with detection range of approximately 30 cm, their recovery rate, which reaches 95%, is always higher than for coloured particles. These results infer that the sediment moves under sediment-selectivity conditions (i.e. Shields, 1936), which is perfectly compatible with the presence of a strong static armour layer (Figures 14 and 15). 66
Annex V
22
Figure 16. Transects where coloured tracers were placed and locations where tracers were
recovered in reaches B (on the left) and A (on the right).
Figure 17. Recovery rate of coloured tracers in Reach A.
Figure 18. Percentage of coloured tracers (of those that were recovered) displaced in Reach A.
0
25
50
75
100
4 to 8 mm 8 to 16 mm 16 to 32 mm 32 to 45 mm
Recovery rate (%)
Reach A, first survey
Reach A, second survey
Reach A, third survey
0
25
50
75
100
4 to 8 mm 8 to 16 mm 16 to 32 mm 32 to 45 mm
% of tracers that moved
Reach A, first survey
Reach A, second survey
Reach A, third survey
Annex V
67
23 Regarding the displacement length of coloured particles and PIT tags, Figure 19 shows that the transport length decreases for coarser particles. Finer clasts (4 to 8 mm) were found up to 80 m downstream from the point in which they were placed, with an average displacement length of approximately 25 m. The coarsest class of coloured particles (32 to 45 mm) and PIT tags was generally transported for less than 5 m. Figure 19. Displacement length for coloured tracers and PIT tags. 5.5 Experiment with natural flow discharge On the 10th January 2017, the full discharge of the river was left flowing downstream from the FCAB station, for three hours (12:00 to 15:00) allowing a sediment transport experiment to be carried out under the natural flow discharge. The resulting hydrograph peaked at approximately 110 l s-1, and featured a smooth rising limb up to the maximum discharge, and a much more abrupt falling limb (Figure 20). Figure 21 shows the aspect of the river at reach A during low flow and high flow conditions. Reach A Outliers Extremes Reach B Outliers Extremes4 a 88 a 1616 a 3232 a 45PITGrain size (mm)0102030405060708090Displacement length (
m)68
Annex V
24
Figure 20. Hydrograph of 10th January 2017, measured in reach A, showing the experiment
with the natural flow discharge in the Silala River.
Figure 21. The Silala River at reach A during flow low flow conditions (around 20 l s-1, on the
left) before the experiment of natural flow discharge and during the high flow conditions at
approximately 110 l s-1 (on the right).
Taking as reference the same cross-section representative of reach A as presented in
Figure 10, while during the low flow conditions the shear stress was about 44 N m-2, at
high flow conditions (natural discharge), the wetted area was 0.544 m2, channel width
was 1.85 m, wetted perimeter was 2.212 m, and mean channel depth was 0.295 m. This
resulted in a hydraulic radius of 0.246 m and the shear stress acting on the channel bed
of around 137 N m-2, i.e. more than triple the shear stress at low flow conditions (Figure
22).
16
36
56
76
96
116
136
12:00:00 13:12:00 14:24:00 15:36:00 16:48:00
Liquid discharge (l s-1)
Annex V
69
25 Figure 22. Representative cross-sections for Reach A, with the water surface for approximately 20 l s-1) and during the experiment with the natural flow discharge (approximately 110 l s-1). As the entrainment and transport of sediments is a function of the shear stress or stream power available to exert this action on the channel bed, the entrainment rate and displacement lengths of sediments is expected to increase with the discharge. Indeed, Figure 23 shows that the percentage of moved tracers was higher for the experiment with the higher discharge, even if its duration was relatively short (only 3 hours) when compared with the very long duration of low flow conditions between field surveys. As expected, the percentage of moved tracers also decreased for coarser sediments. The same trend was identified using the PIT tags (Figure 23). Indeed, the percentage of moved PITs increased for the experiment with higher discharge, and the coarsest PITs placed in the channel bed (62 mm) moved only during high flow although they remained immobile during 4 months of low flow conditions. Figure 23. Percentage of moved tracers (on the left) and PIT tags (on the right) during low and high flow conditions on reach A. Figures for reach B are similar and are not shown. 0.00.10.20.30.40.50.60.701234Elevation (
m)Distance (m)Low flow (20 l s-1)High flow (110 l s-1)02550751004 to 8 mm8 to 16 mm16 to 32 mm32 to 45 mm% of tracers that movedReach A, low flowReach A, high flow025507510030-40 mm40-50 mm50-60 mm60-70 mm% of PITs that movedLow flowHigh flow26As 70
Annex V
26
As to the displacement length of sediments of different size under low and high flow
conditions, Figure 24 shows that coarser sediments are transported less than finer
fractions, and more importantly that sediments are transported for longer distances
under the experiment with the natural flow conditions. This is true for both tracers and
PIT tags.
Figure 24. Displacement lengths of tracers (on the left) and PIT tags (on the right) during low
and high flow conditions. Data for reach A and B are merged on these graphs.
5.6 Bedload sampling
Results from bedload sampling during the low and the high flow conditions
(approximately 20 and 110 l s-1, respectively) indicate that transport rate was three
orders of magnitude higher during the natural flow conditions experiments (Figure 25).
Low flow
Outliers
Extremes
High flow
Outliers
Extremes
4 a 8 8 a 16 16 a 32 32 a 45
Grain size class (mm)
0
10
20
30
40
50
60
70
80
90
Displacement length (m)
Low flow
Outliers
Extremes
High flow
Outliers
Extremes
30-40 40-50 50-60
PIT size (mm)
0
5
10
15
20
25
30
35
40
Displacement length (m)
Annex V
71
27 Figure 25. Bedload transport rate assessed during low and high flow conditions. Figure on the right shows a sampling operation during high flow conditions. 5.7 Morphological classification and Morphological Quality Index (MQI) of the Silala River At the first Level of the Rosgen’s stream classification, the Silala River can be readily classified as an A-type reach. Type A is generally a well-entrenched river, with a low width/depth ratio (0.1 for the Silala River), and is totally confined (laterally contained). The channel slopes in the type A streams range from 4 to 10 percent (being 5.7 and 4.7 in reaches A and B of the Silala River, respectively), and streamflows at the bankfull stage are typically described as step/pool-cascade morphology (see e.g. Figure 26). Normally, type A streams are found within valley types that, due to their inherent channel steepness, exhibit a high sediment transport potential and a relatively low in-channel sediment storage capacity (Rosgen, 1994). At the level of Rosgen’s valley morphology classification, the Silala River can be classified as a valley type I, characterized by high elevational relief, highly dissected fluvial slopes, which typically hosts step/pool and cascade bed features. Regarding the Level II stream classification, given the very low entrenchment ratio (< 1.4), the very low width to depth ratio at bankfull discharge (< 12), the low sinuosity (< 1.2), the high slope (> 0.1) and the gravel nature-cobble nature of the sediments on bed surface, the Silala River can be classified as an A4a+ stream type. Median 25%-75% Non-Outlier Range Outliers Extremes20 l s-1110 l s-10.0000050.0000500.0005000.0050000.0500000.500000g m-1
s-128Figure26. 72
Annex V
28
Figure 26. Step-pool unit in the reach B of the Silala River.
As to the MQI, results of the application of the index to the Silala River are shown in
Table 2. The Morphological Quality Index indicates that both reaches of the Silala River
are overall in good quality conditions. The only aspects that penalized the score of the
MQI in the Silala are related with the impacts on the hydrological regime due to water
extraction downstream of the border, and the partial disruption of longitudinal
connectivity due to the presence of small weirs.
Annex V
73
29 Question Topic Reach A Reach B Continuity F1 Longitudinal continuity in sediment and wood flux 0 0 F2 Presence of a modern floodplain (not to be assessed in confined channels) / / F3 Hillslopes - river corridor connectivity 0 0 F4 Processes of bank retreat (not to be assessed in confined channels) / / F5 Presence of a potentially erodible corridor (not to be assessed in confined channels) / / Morphology F6 Bed configuration - valley slope 0 0 F7 Forms and processes typical of the channel pattern (not in confined channels) / / F8 Presence of typical fluvial forms in the alluvial plain (not to be assessed in confined channels) / / F9 Variability of the cross-section 0 0 F10 Structure of the channel bed 0 0 F11 Presence of in-channel large wood (not to be assessed above the tree-line) / / Vegetation in the fluvial corridor F12 Width of functional vegetation in the fluvial corridor 0 0 F13 Linear extension of functional vegetation along the banks 0 0 Upstream alteration of longitudinal continuity A1 Upstream alteration of discharges 6 6 A2 Upstream alteration of sediment transport 3 3 Alteration of longitudinal continuity in the reach A3 Alteration of discharges in the reach 6 6 A4 Alteration of sediment transport in the reach 0 0 A5 Crossing structures 0 0 Alteration of lateral continuity A6 Bank protections 0 3 A7 Artificial levees (not to be assessed in confined channels) / / A8 Artificial changes of river course (not in confined channels) / / Alteration of channel morphology and/or substrate A9 Other bed stabilization structures 0 0 Intervention of maintenance and removal A10 Sediment removal 0 0 74
Annex V
30
A11 Wood removal
(not to be assessed above the tree-line)
/ /
A12 Vegetation management 0 0
Reach
A
Reach
B
Total deviation STOT 15 18
Maximum deviation SMAX=119-Sna 92 92
Morphological alteration
index
MAI=STOT/SMAX 0.16 0.20
Morphological Quality
Index
MQI=1-MAI 0.84 0.80
Quality class of the reach Good Good
Table 2. Questionnaire of the Morphological Quality Index (MQI) applied to reaches A and B
of the Silala River.
5.8 Fish population in the Silala River
123 fish were captured, weighted and measured on the Silala River (Figure 27). All
captured fishes were rainbow trout (Oncorhynchus mykiss), and no native specimens
were found. The captured rainbow trout ranged from 5 to 21 cm, and from 2 to 72 g
(Figure 28). In Chile, trout were introduced for sport fishing beginning in 1890
(Arismendi et al., 2014), and salmonids managed to be spread throughout much of the
country’s highland and piedmont areas in the following decades (Vila et al., 1999),
creating resident populations. Presently, rainbow trout can be found in the mountainous
reaches of Chile’s perennially flowing rivers, from the highlands bordering Bolivia
(18°S) to the subantarctic rivers of Patagonia (55°S). In the Silala River, it is most likely
that the rainbow trout arrived in the Chilean reach from upstream reaches, where we
understand that several attempts to install aquaculture sites were made in the past (e.g.
https://boliviasol.wordpress.com/2012/04/19/proyecto-piscicola-aprovech…).
However, it is interesting to note on Figure 28 that the length-to-weight ratio of
the rainbow trout in the Silala is remarkably similar to the length-to-weight ratio of
rainbow trout derived using several datasets from central Chile (Mao et al., 2016) and
several Environmental Assessment Impact reports (gathered from http://sea.gob.cl).
Indeed the regression curve obtained using data from Central Chile is the following:
𝑦𝑦 = 0.0106𝑥𝑥3.02
where y is fish weight (g) and x is fish length (cm).
Annex V
75
31 The regression curve obtained using data from Chilean Environmental Assessment Impact reports is: 𝑦𝑦=0.0169𝑥𝑥2.79 whereas the best-fit regression for the rainbow trout in the Silala is: 𝑦𝑦=0.0331𝑥𝑥2.50 Figure 27. Images of fish collected in a bucket during one survey (on the left), and one fish ready to be measured. 76
Annex V
32 Figure 28. Length-to-weight relationship for rainbow trout captured in the Silala River and other rivers of Central Chile. 6 DISCUSSION 6.1 Presence of the armour layer Results show that the channel of the Silala River is well armoured, as the armour ratio is 2.5 for reach A and 2 for reach B. As briefly presented in the introduction, the armour layer could be static (sensu Chin et al., 1994; when flows are strong enough to move mainly the smaller sizes, and there are limited sediment supply conditions) or mobile (sensu Parker et al., 1982 under equal-mobility conditions and unlimited sediment supply of the moved grain sizes). Because sediment supply in the Silala River is relatively small due to the lack of major sediment sources along the main channel and the dense mat of Calamagrostis Eminens that grows on the banks, protecting them from being eroded, the armour layer in the Silala is considered static. It is worth comparing the armour ratio of the Silala River with the armour ratios of other rivers. Hassan et al. (2006) collected values of armour ratios of other rivers worldwide, and showed that the degree of armouring is rather weak (average armour ratio around 1) in ephemeral streams, and is between 1 and 3 on rivers in humid areas, being higher on y = 0.0169x2.7986R² = 0.9739y = 0.0331x2.5067R² = 0.948y = 0.0106x3.0257R² = 0.99530.11101001000110100Fish weight (
g)Fish length (cm)central ChileSilalaEIA data
Annex V
77
33 rivers with high sediment supply conditions (from abundant bank erosions and upstream tributaries). Indeed, Figure 29 shows that humid/snowmelt streams are likely to develop a higher degree of armouring than desert streams. It is particularly interesting to note that the armour rate of the Silala River (red points in Figure 29) is comparable with values found in rivers with perennial regimes and a humid environment. This is due to the fact that, although located in an arid environment, the regime of the Silala is not ephemeral but rather a groundwater-dominated permanent river, with very high base flow index. Indeed, the record of water discharge in the study reaches (e.g. Figure 13) shows that the discharge ranges around the same average value with slight daily fluctuations, although more significant pulsations of discharge are likely to occur due to rainfall events, especially in summer. Figure 29. Figure reproduced from Hassan et al. (2006) showing the armour ratio of rivers with different regimes. Data collected in the reaches A and B of the Silala River is plotted in red. The caption of the Hassan et al. (2006) paper is the following: Median size of surface and subsurface of (a) ephemeral streams and (b) snowmelt, humid, and arid streams. Lines of equal armour ratio are shown. Bars are errors around the mean value. Figures 1a and 1b are based on Table 1 and the following published data from humid and snowmelt streams: Allt Dubhaig (Wathen et al., 1995); Tom McDonald (Hassan and Woodsmith, 2004); Carnation Creek (Hassan et al., 2006); Harris Creek (Church and Hassan, 2002); Fraser River (Church et al., 1987); Boise River, Little Slate Creek, Lochsa River, Lolo Creek, MF Red River, North Fork Clearwater River, Rapid River, Selway River, South Fork Payette River, South Fork Salmon River, and Valley Creek (Whiting and King, 2003); Jacoby Creek, Prairie, and North Caspar (Lisle, 1989); Redwood Creek (Lisle and Madej, 1992); Willamette River (Klingeman and Emmett, 1982), Oak Creek (Parker and Klingeman, 1982); Segehen Creek (Andrews and Erman, 1986); Haut Glacier d’Arolla (Cudden and Hoey, 2003); Bas Glacier d’Arolla 112244661101001000110100D50(mm)D50S(mm)Armour ratioAridHumidSnow meltSilala
78
Annex V
34 (Warburton, 1992); Goodwin Creek (Kuhnle, 1992); Virginio Creek (Tacconi and Billi, 1987); Drau River (Habersack and Laronne, 2001); Fool Creek, East St. Louis, and Little Granite (Ryan, 2001; Ryan and Emmett, 2002); and Mamquam and East Creek, both in British Columbia (Hassan et al., 2006).6.2 Tracers and PIT tags Results from the field experiment on entrainment ratio and displacement length of marked sediments (coloured particle sand PIT tags) clearly demonstrate that the channel bed is alluvial, as sediments are actively and continuously transported in the river. Also, sediments are entrained and transported following the size-selective modality. Although equal-mobility conditions have also been measured in the field (e.g. Marion and Weirich, 2003), the size-selective nature of sediment transport has been widely reported in the literature (e.g. Ashworth and Ferguson, 1989; Kuhnle, 1992; Wathen et al., 1995), with sediment transport approaching equal-mobility only during the highest flows. It is thus not surprising that the Silala River exhibits size-selective transport during the relatively low flow conditions in which the river was studied, and this reinforces the evidence that fluvial processes are ongoing in the channel, and are shaping the channel geometry, thus demonstrating that it is not only a river system that transmits perennial flows, but one that continues to be geomorphologically active, i.e. reshaping the channel due to fluvial processes. In order to compare the displacement length measured in the Silala with other data available in the literature, it is worth expressing the grain size and the travel distance in dimensionless ways, to allowing comparison. If dimensionless terms are used for scaling both grain size (using the median diameter of subsurface sediments D50SS) and the displacement lengths (using the mean transport distance of the size equal to D50) of tracers and PIT tags, data fall quite well on previously reported trend (e.g. Church and Hassan, 1992; Figure 30). The relationship proposed by Church and Hassan (1992) reads: 𝐿𝐿𝐿𝐿𝐷𝐷50𝑆𝑆𝑆𝑆=1.77[1−𝑙𝑙𝑙𝑙𝑙𝑙10(𝐷𝐷𝐷𝐷50𝑆𝑆𝑆𝑆)]1.35While for the Silala River we obtained the following empirical relationship (R2 = 0.82): 𝐿𝐿𝐿𝐿𝐷𝐷50𝑆𝑆𝑆𝑆=1.152[1−𝑙𝑙𝑙𝑙𝑙𝑙10(𝐷𝐷𝐷𝐷50𝑆𝑆𝑆𝑆)]2.180
Annex V
79
35 The convex trend shown in Figure 30 confirms that the mean displacement length reduces rapidly for fractions coarser than the median size (i.e. size-selective transport), whereas the displacement length of fractions finer than D50 is more insensitive to grain size, as expected in perennial fluvial systems. Figure 30. Relationship between scaled travel distance (Li/L50SS) and scaled particle size (Di/D50SS). The dashed grey line represents the Church and Hassan (1992) relationship. The black line represents the regression curve for the data of Silala. Regarding the displacement length of sediments of different sizes during low and high flow conditions, results are clear in showing that tracers and PITs travelled for longer distances during the experiment with higher discharge. Similar trends are commonly found in alluvial rivers with size-selective transport processes. For instance, Ashworth and Ferguson (1989) and Marion and Weirich (2003) previously stated that, in a condition of size-selective transport, bedload displacement distance decreases with increasing grain size. Also, empirical relationships are available between the mean distance of movement and the excess stream power or excess of shear stress above the critical threshold (e.g. Church and Hassan, 1992; Gintz et al., 1996; Hassan et al., 1992; Lenzi, 2004). As expected in alluvial rivers with size-selectivity bedload process, the bedload transport rate increases with the discharge, and the grain size of transported material increases with increasing shear stress (Komar and Shih, 1992; Marion and Weirich, 2003). This has been verified in the Silala River, analyzing both the transport rate and the grain size of transported sediments during low flow and the high flow experiment. 0.010.11100.1110Li/L50 SSDi/D50ssTracers Reach APITs Reach ATracers Reach BPITs Reach BChurch & Hassan(1992)Silala
80
Annex V
36 6.3 Fish population Results show that the length-to-weight relationship derived for the rainbow trout found in the Silala River is remarkably similar to the relationship derived from a range of other rivers in Central Chile. This suggests that, although probably introduced unintentionally from upstream during or after aquaculture attempts, the population of trout found optimal conditions for spawning, reproducing, and growing in the Silala River. As with other invasive salmonids in Chile, since their introduction, rainbow trout have successfully outcompeted native fishes through habitat exclusion (Penaluna et al., 2009) and predation (Soto et al., 2006), decimating a freshwater fish fauna characterized by low diversity, high endemism (Vila et al., 1999), and high levels of endangerment (OECD and ECLAC, 2005). It is thus not possible to say if prior to rainbow trout introduction there were native species in the Silala, but no specimens of native fish could be found during the study period. 6.4 River morphology The morphological assessment of the river confirms that the main channel features a morphology that fits the range of grain size, width, slope, and confinement of the site. The morphology is typical of gravel/cobbled-bed, high-gradient streams, featuring step-pools and cascade morphology. The lack of lateral bars is typical in these environments, and in the case of Silala is due to the dense vegetation reinforcing the banks and the lack of abundant sediment supply. The morphological quality index (MQI) is an index aimed at assessing the current geomorphological conditions of a river reach, and especially the deviation of such conditions from reference conditions. In the literature there is still a lively discussion as to what should be considered as a reference condition in fluvial systems (e.g. Brierley and Fryirs, 2005; Palmer et al., 2005; Dufour and Piégay, 2009; Burchsted et al., 2010), especially because unimpacted or “pristine” conditions are virtually impossible to find anywhere (e.g. Comiti, 2012; Wyżga et al., 2012), and because natural climate variability makes rivers change constantly. Indeed, rivers follow an evolutionary trajectory, adjusting their morphology to changes in boundary conditions (Brierley et al., 2008; Dufour and Piégay, 2009). Because of this, reference conditions are framed in the MQI and other recent indexes in terms of an expected state defined as the best conditions that can be attained by a river, given the prevailing catchment boundary conditions (Brierley and Fryirs, 2005). The reference conditions for a river reach are defined considering channel forms and processes, artificiality and channel adjustments. In the application to the Silala River, the MQI indicates that channel forms and processes are close to the potential state, and that the overall quality is slightly reduced Annex V
81
37 by the presence of hydrological impacts (water intake upstream of the studied reaches) and transversal weirs that reduce longitudinal connectivity in the river. 7 FINAL REMARKS The presence of a static armour layer in the Silala River infers that there are fluvialprocesses ongoing in the main channel, that size-selective transport is able towinnow the finer grains from the channel, leaving on the surface the coarsestfractions, which are intrinsically more difficult to entrain and move. The fact thatthe channel is armoured could be considered counterintuitive in a river located in anarid environment, but this is because the regime is one of permanent flow. Indeed,although located in an arid environment, the regime of the Silala is not ephemeralbut rather a groundwater-dominated permanent river.In the Silala River, finer sediments are moved in higher percentages and for longerdistances than coarser sediment fractions. Also, sediments are transported in higherpercentages and for longer displacement lengths during higher discharges. Asexpected, bedload transport rate was also higher during the natural dischargeexperiment. These characteristics are clear evidence of size-selective transportdynamics in natural rivers, and are to be expected during regular flows in limitedsediment supply conditions, which characterize the Silala River. Evidence of size-selective transport justifies the presence of a static armour layer and reinforces theindication that fluvial processes are ongoing in the Silala River.The abundant presence of invasive rainbow trout in good condition and in a widespan of size and weight demonstrate that the general conditions of the river (aquatichabitats, water quantity and quality, availability of food) are good enough to sustaina healthy population of resident fish.The stream features a cascade-step/pool morphology, which would be expectedgiven the local range of slope and grain size. This reinforces the consideration thatthe Silala is an alluvial river with fully-developed fluvial processes, which shape thechannel morphology through active coarse sediment dynamics. The morphologicalquality index classifies the river as featuring overall good morphological conditions,thus exhibiting forms and processes that are close to the full potential of the river,given the existing boundary conditions.
82
Annex V
38 8 REFERENCES Andrews, E. D., 1983. Entrainment of gravel from naturally sorted river bed material. Geological Society of America Bulletin, 94, 1225–1231. Andrews, E. D., Erman, D. C., 1986. Persistence in the size distribution of surficial bed material during an extreme snowmelt flood. Water Resources Research, 22, 191-197. Arismendi, I., Penaluna, B. E., Dunham, J. B., García de Leaniz, C., Soto, D., Fleming, I. A., Gomez-Uchida, D., Gajardo, G., Vargas, P. V., León-Muñoz, J., 2014. Differential invasion success of salmonids in southern Chile: patterns and hypotheses. Reviews in Fish Biology and Fisheries, 24(3), 919–941. Ashworth, P. J. and Ferguson, R. I., 1989. Size selective entrainment of bedload in gravel bed streams. Water Resources Research, 25, 627–634. Barry, J. J., Buffington, J.M., King, J.G., 2004. A general power equation for predicting bed load transport rates in gravel bed rivers. Water Resources Research, 40: W10401. Bathurst, J. C., 2007. Effect of coarse surface layer on bedload transport. Journal of Hydraulic Engineering, 133(11), 1192–1205. Brierley, G. J., Fryirs, K. A., 2005. Geomorphology and River Management: Applications of the River Style Framework. Wiley-Blackwell, Oxford, UK. Brierley, G. J., Fryirs, K. A., Boulton, A., Cullum, C., 2008. Working with change: the importance of evolutionary perspectives in framing the trajectory of river adjustment. In: Brierley, G., Fryirs, K. A. (Eds.), River Futures: An Integrative Scientific Approach to River Repair. Society for Ecological Restoration International, Island Press, Washington, DC, USA, pp. 65–84. Bunte, K., Abt, S. R., 2001. Sampling surface and subsurface particle-size distributions in wadable gravel-and cobble-bed streams for analyses in sediment transport, hydraulics, and streambed monitoring. General Technical Report RMRSS-GTR-74, Fort Collins, CO: US Department of Agriculture, Forest Service, Rocky Mountain Research Station. Bunte, K., Swingle, K.W., Abt, S.R., 2007. Guidelines for Using Bedload Traps in Coarse Bedded Mountain Streams: Construction, Installation, Operation and Sample Processing. US Department of Agriculture, Forest Service, Rocky Mountain Research Station. Burchsted, D., Daniels, M., Thorson, R., Vokouin, J., 2010. The river discontinuum: applying beaver modifications to baseline conditions for restoration of forested headwaters. Bioscience, 60(11), 908–922. Annex V
83
39 Chin, C. O., Melville, B. W., Raudkivi, A. J., 1994. Streambed armoring. Journal of Hydraulic Engineering, 120, 899–918. Church, M., 2006. Bed material transport and the morphology of alluvial river channels. Annual Review of Earth and Planetary Sciences, 34(1), 325-354, DOI:10.1146/annurev.earth.33.092203.122721 Church, M., Hassan, M. A., 1992. Size and distance of travel of unconstrained clasts on a streambed. Water Resources Research, 28, 299–303. Church, M., Hassan, M. A., 2002. Mobility of bed material in Harris Creek. Water Resources Research, 38(11), 1237, doi:10.1029/2001WR000753. Church, M., Hassan, M. A., Wolcott, J. F., 1998. Stabilizing self-organized structures in gravel-bed stream channels: field and experimental observations. Water Resources Research, 34(11), 3169– 3179. Church, M., McLean D. G., Wolcott J. F., 1987. River bed gravels: sampling and analysis. In: Thornes, C. R., Bathurst, J. C., Hey R. D. (Eds.), Sediment Transport in Gravel Bed Rivers, pp. 43-88, John Wiley, Hoboken, N. J. Comiti, F., 2012. How natural are Alpine mountain rivers? Evidence from the Italian Alps. Earth Surface Processes and Landforms, 37, 693–707 DOI:10.1002/esp.2267. Cudden, J. R., Hoey, T. B., 2003. The causes of bedload pulses in a gravel channel: the implications of bedload grain-size distributions. Earth Surface Processes and Landforms, 28, 1411-1428. Dufour, S., Piégay, H., 2009. From the myth of a lost paradise to targeted river restoration: forget natural references and focus on human benefits. River Research and Applications, 24, 1–14. Gintz, D., Hassan, M., Schmidt, K., 1996. Frequency and magnitude of bedload transport in a mountain river. Earth Surface Processes and Landforms, 21, 433– 445. Habersack, H. M., and Laronne, J. B., 2001. Bed load texture in an Alpine gravel bed river. Water Resources Research, 37, 3359–3370. Hassan, M. A., Church, M., Ashworth, P. J., 1992. Virtual rate and mean distance of travel of individual clasts in gravel-bed channels. Earth Surface Processes and Landforms, 17, 617–627. Hassan, M. A., Egozi, R., Parker, G., 2006. Experiments on the effect of hydrograph characteristics on vertical grain sorting in gravel bed rivers. Water Resources Research, 42, 1–15. DOI:10.1029/2005WR004707. 84
Annex V
40 Hassan, M. A., Woodsmith, R., 2004. Bed load transport in an obstruction-formed pool in a forested gravelbed stream. Geomorphology, 58, 203–221. Klingeman, P. C., Emmett W. W., 1982. Gravel bedload transport processes. Hey, R. B., Bathurst J. C., Thorne C. R. (Eds), Gravel-Bed Rivers: Fluvial Processes, Engineering and Management, pp. 141– 179, John Wiley, Hoboken, N. J. Komar, P.D. and Shih, S., 1992. Equal mobility versus changing bed load grain sizes in gravel-bed stream. In: Billi, P., Hey, R.D., Thorne, C.R., Tacconi, P. (Eds.), Dynamics of Gravel-bed Rivers. Wiley, Chichester, UK, pp. 73– 93. Kuhnle, R. A., 1992. Fractional transport rates of bedload on Goodwin Creek. In: Billi, P., Hey, R. D., Thorne, C. R., Tacconi, P. (Eds.), Dynamics of Gravel-Bed Rivers. Wiley: Chichester, 141–155. Lane, E.W., 1955. Design of stable channels. Transactions of the American Society of Civil Engineerss, 120(1), 1234-1260. Leopold, L.B., Wolman, M.G., Miller, J.P., 1964. Fluvial Processes in Geomorphology. W.H. Freeman and Co., San Francisco, CA. 522 pp. Lenzi M. A., 2004. Displacement and transport of marked pebbles, cobbles and boulders during floods in a steep mountain stream. Hydrological Processes, 18, 1899–1914. Lisle, T. E., 1989. Sediment transport and resulting deposition in spawning gravels, north coastal California. Water Resources Research, 25, 1303–1319. Lisle, T. E., Madej M. A., 1992. Spatial variation in armoring in a channel with high sediment supply. In: Billi, P., Hey, R.D., Thorne, C.R., Tacconi, P. (Eds.), Dynamics of Gravel-bed Rivers. Wiley, Chichester, UK, pp. 277-293. Mao, L., Ugalde, A., Iroume, A., Lacy, S.N., 2016. The effects of replacing native forest on the quantity and impacts of in-channel pieces of large wood in Chilean streams. River Research and Applications. doi: 10.1002/rra.3063. Marion, D.A., Weirich, F., 2003. Equal-mobility bed load transport in a small, step–pool channel in the Ouachita Mountains. Geomorphology, 55, 139–154. Moore, R.D., 2005. Introduction to salt dilution gauging for streamflow measurement Part 3: slug injection using salt in solution. Streamline Watershed Management Bulletin, 8(2), 1–6. OECD, ECLAC, 2005. OECD environmental performance reviews: Chile 2005. OECD Environmental Performance Reviews. OECD Publishing. Annex V
85
41 Palmer, M. A., Bernhardt, E. S., Allan, J. D., Lake, P. S., Alexander, G., Brooks, S., Carr, J., Clayton, S., Dahm, C. N., Follstad, S. J., Galat, D. L., Loss, S. G., Goodwin, P., Hart, D. D., Hassett, B., Jenkinson, R., Kondolf, G. M., Lave, R., Meyer, J. L., O'Donnell, T. K., Pagano, L., Sudduth, E., 2005. Standards for ecologically successful river restoration. Journal of Applied Ecology, 42, 208–217. Parker, G. and Klingeman, P. C., 1982. On why gravel bed streams are paved. Water Resources Research, 18, 1409–1423. Parker, G., Klingeman, P.C., McLean, D. G., 1982. Bedload and size distribution in paved gravel-bed streams. Journal of the Hydraulic Division, 108(4), 544–571. Parker, G. and Toro-Escobar, C. M., 2002. Equal mobility of gravel in streams: the remains of the day. Water Resources Research, 38(11), 1264. Penaluna, B. E., Arismendi, I., Soto, D., 2009. Evidence of interactive segregation between introduced trout and native fishes in Northern Patagonian Rivers, Chile. Transactions of the American Fisheries Society, 138(4), 839–845. Proffitt, G. T. and Sutherland, A. J., 1983. Transport of non-uniform sediments. Journal of Hydraulic Research, 21, 33–43. Rinaldi, M., Surian, N., Comiti, F., Busettini, M., 2013. A method for the assessment and analysis of the hydromorphological condition of Italian streams: the Morphological Quality Index (MQI). Geomorphology, 180(1), 96–108. Rosgen, D. L., 1994. A classification of natural rivers. Catena, 22, 169–199. Ryan, S. E., 2001. The influence of sediment supply on rates of bedload transport: a case study of three streams on the San Juan national forest. In: Proceedings of the Seventh Federal Interagency Sedimentation Conference, vol. 1, III-48 – III-54. Ryan, S. E. and Emmett W. W., 2002. The Nature of Flow and Sediment Movement in Little Granite Creek near Bondurant, Wyoming. General Technical Report RMRS-GTR-90. Shields, A., 1936. Application of Similarity Principles and Turbulence Research to Bed-load Movement. California Institute of Technology, Pasadena, CA. English translation by Ott, W.P., van Uchelen, J.C., pp. 36. Soto, D., Arismendi, I., González, J., Sanzana, J., Jara, F., Jara, C., Guzmán, E., Lara, A., 2006. Southern Chile, trout and salmon country: invasion patterns and threats for native species. Revista Chilena de Historia Natural, 79(1), 97-117. 86
Annex V
42 Tacconi, P. and Billi P., 1987. Bed load transport measurements by the Vortex-tube trap on Virginio Creek, Italy. In: Thorne, C. R., Bathurst, J. C., Hey R. D. (Eds.), Sediment Transport in Gravel-bed Rivers, John Wiley, Hoboken, N. J., pp. 583–616. Vericat, D., Church, M., Batalla, R. J., 2006. Bed load bias: comparison of measurements obtained using two (76 and 152 mm) Helley–Smith Samplers in a gravel bed river. Water Resources Research, 42, 1-13, W01402. DOI:10.1029/2005WR004025. Vila, I., Fuentes, L., Contreras, M., 1999. Peces límnicos de Chile. Boletín Museo Nacional de Historia Natural de Chile, 48, 61–75. Warburton, J., 1992. Observations of bed load transport and channel bed changes in a proglacial mountain stream. Arctic and Alpine Research, 24, 195-203. Wathen, S. J., Ferguson, R. I., Hoey, T. B., Werritty, A., 1995. Unequal mobility of gravel and sand in weakly bimodal river sediments. Water Resources Research, 31, 2087–2096. Whiting, P. J. and King, J. G., 2003. Surface particle sizes on armoured gravel streambeds: effects of supply and hydraulics. Earth Surface Processes and Landforms, 28, 1459-1471. Wilcock, P. R. and Crowe, J. C., 2003. Surface-based transport model for mixed-size sediment. Journal of Hydraulic Engineering, 129, 120–128. DOI:10.1061//(ASCE)0733-9429(2003)129:2(120). Wohl, E. 2014. Rivers in the Landscape: Science and Management. Wiley-Blackwell, 330 pp. Wyżga, B., Zawiejska, J., Radecki-Pawlik, A., Hajdukiewicz, H., 2012. Environmental change, hydromorphological reference conditions and the restoration of Polish Carpathian rivers. Earth Surface Processes and Landforms, 37(11), 1213–1226. DOI:10.1002/esp.3273. Annex V
87
88
Annex VI
McRostie, V., 2017. Archaeological First Baseline Study for the Silala River, Chile
89
90
Annex VI
ARCHAEOLOGICAL FIRST BASELINE STUDY
FOR THE SILALA RIVER, CHILE
Virginia McRostie B. (PhD)
Assistant Professor, Pontificia Universidad Católica de Chile
May, 2017
Annex VI
91
TABLE OF CONTENTS 1.INTRODUCTION ..................................................................................................... 11.1 Presentation ............................................................................................................. 1 1.2. Location .................................................................................................................. 1 1.3. Objective of the report ............................................................................................ 3 1.4 Summary of the methodology ................................................................................. 4 1.5 Structure of the report.............................................................................................. 4 2.SUMMARY OF CONCLUSION .............................................................................. 43.METHODOLOGY .................................................................................................... 54.PRESENTATION OF FINDINGS ............................................................................ 75.DISCUSSION OF FINDINGS ................................................................................ 126.CONCLUSIONS ..................................................................................................... 187.REFERENCES ........................................................................................................ 19APPENDIX A APPENDIX B 92
Annex VI
1
1. INTRODUCTION
1.1 Presentation
This report was requested by DIFROL (Dirección Nacional de Fronteras y Límites del
Estado) within the context of the Silala River dispute between Chile and Bolivia, to be
heard before the International Court of Justice (ICJ) at The Hague. As an archaeologist
from the Pontificia Universidad Católica I was asked to advise on the archaeological
context of the Silala River basin. This report was elaborated under the supervision and
instruction of Dr. Howard Wheater and Dr. Denis Peach.
1.2. Location
The Silala basin is shared by Bolivia (upstream) and Chile (downstream), and lies in the
Atacama Desert approximately 300 km northeast of Antofagasta, Chile. The river starts
in the foothills of Cerro Inacaliri o del Cajón (5689 m.a.s.l.) and Cerro Silala (5703
m.a.s.l.), at the northern end of the National Reserve of Andean fauna Eduardo Abaroa,
in the province of Sud Lípez, department of Potosí, Bolivia. The spring sources
originate about 4 km from the border with Chile, from where it flows about 7 kilometers
to discharge into the river San Pedro, which is a tributary of the Loa, the longest river in
Chile and the main river that crosses the Atacama Desert, which flows into the Pacific
Ocean (Figure 1).
Annex VI
93
2 Figure 1. Silala River location. At the international border, the Silala ravine is incised 10-15 m into local bedrock composed of the late Pliocene Cabana Ignimbrite (4.12 million years old) and the late Pliocene or early Pleistocene (ca. 2.6-1.5 million years old) Silala Ignimbrite. This incision likely followed a phase of alluvial and glacial outwash sediment accumulation when glaciers still existed along the flanks of adjacent peaks during the late Pleistocene 94
Annex VI
3
(40 to 12 thousand years ago) (Latorre and Frugone, 2017). This cutting of the ravine
has formed the four erosional terraces present in the Silala ravine (from oldest to
youngest, T4-T3-T2-T1) (Figure 2).
Figure 2. Schematic cross-section showing the location of the four terraces of the Silala River.
Heights might vary within the ravine.
Although there are many ephemeral tributaries that flow to the river, there are two
major tributary channels that join the Silala in the reaches immediately downstream of
the border. These are the Quebrada Negra, which converges with the river from the
southeast some 500 meters below the frontier with Bolivia, and further downstream the
Quebrada Inacaliri, which joins the Silala River near the Codelco intake (See Figure 3).
1.3. Objective of the report
The objective of this report is to evaluate the relationship between the human settlement
pattern and the Silala River primarily within pre-Columbian periods (ca. 9000 - 500
BP1), and to a lesser extent within historical periods (XVIth - XXth century). Desert
societies are indissolubly linked to water bodies, as has been documented since the
1 Before present.
Annex VI
95
4 beginning of pre-Columbian history in the area. Most early Holocene settlements (ca. 11500-8000 BP) are found near water courses within an epoch of high humidity (Núñez et al., 2005). During the middle Holocene (ca. 8000-5000 BP) drier conditions caused the abandonment of previous occupations and a few human sites have been found in eco-refuges, where water and biotic elements were still available (Núñez et al., 1999; Grosjean et al., 2007; Núñez and Santoro, 2011). Finally, wetter conditions during the late Holocene (ca. 3000 BP onwards) promoted demographic expansion as well as sedentary and agricultural villages within oases and adjacent to major fluvial courses (Santoro et al., 2016). Hence, how man inhabited the space and geography of the Silala River can be linked to an understanding of its hydrological history. 1.4 Summary of the methodology The Silala basin was prospected from the international border with Bolivia to a few meters below the Inacaliri Police Station. For each archaeological finding or site, a file card with standard information was created (Appendix A, Archaeological filecards Silala Report). Sites and materials were classified using regional archaeological typologies (Appendix B, Summary of the sites). 1.5 Structure of the report In Chapter II, the conclusions of the report are summarized. In Chapter III, the methodology used to develop this report is defined. In Chapter IV, the findings are presented. These are also described, in more technical detail, on the card files included in Appendix A (Archaeological filecards Silala Report) and on the tables in Appendix B (Summary of the sites). The findings are discussed in Chapter V, within the context of the regional archaeology and history. 2.SUMMARY OF CONCLUSIONThis first archaeological baseline study for the Silala River gave evidence of 46 sites that can be ascribed to pre-Columbian (9000 - 500 BP), Historical (XVI-XXth centuries) or Undefined periods, depending on the surface materials and structures found. Their location within the local geography and their relative chronologies suggests that the Silala River was active and dynamic at least 1500 years before present. 96
Annex VI
5
3. METHODOLOGY
The Silala basin was prospected (walked following footpaths) over three days and by
one person (Figure 3). Special attention was paid to the tributary channels that join the
main course of the Silala River, either as active or dry courses, and to springs and/or
wetlands, as they could be promising terrain for pre-Columbian communities to settle.
As can be seen on Figure 3, the entire length of the river from the International border to
the Inacaliri Police Station was examined, as well as the terraces immediately above the
river on both the south and the north side of the ravine.
Figure 3. Geographical area prospected and sites encountered.
Annex VI
97
6 Sites with evidence of human occupation were recorded using the UTM system and a standard card file (see Appendix A), on which main variables were registered as follows: VARIABLE DESCRIPTION SITE NAME and NUMBER Siarq following by correlative numbers TYPE OF SITE and DESCRIPTION Stone structures, caves, isolated artefacts UTM N - E – m.a.s.l. LOCATION WITHIN THE RIVER AREA On the riverbed or within/above the terraces CONSERVATION OF THE SITE Good-regular-bad condition CHRONOLOGY PRE-COLUMBIAN2 to 450 BP) – HISTORICAL3 (450 BP to present) – PRE-COLUMBIAN and HISTORICAL4- UNDEFINED5 PICTURES Photos from the site and materials Table 1. Variables recorded on the card files. To achieve the aims of this study the division between pre-Columbian and historical periods is a useful starting point, though based on regional archaeology (Aldunate et al., 1986; Sinclaire, 2004) a more accurate timeline is provided to discuss the pre-Columbian sites (Table 2). In the future more studies regarding historical archaeology of the area will consolidate a better chronology for this latter period. 2 Site with indigenous artifacts (e.g. pottery, lithics, archaeobotanical remains). 3 Site with Historical elements (e.g. cans, tables, iron, clothes, batteries, plastics, etc.). 4 If there was evidence of both periods (e.g. pre-Columbian: pottery, and historical: plastic). 5 Sites built by humans like for example stone structures, but without any diagnostic feature or artefact to classify them as pre-Columbian or historical. 98
Annex VI
7
9000 BP 3000 BP 1000 BP 550 BP 410 BP
Archaic Formative Late
Intermediate
Period (LIP)
Late Hispanic
Early/Middle/Late Early/Late Mallku Inka Historical
mobile hunter
gatherers
agropastoral
economies,
sedentary
villages,
pottery, textiles,
metallurgy
agrohydraulical
systems, wide
interchange,
Highlands
influences
empire
expansion
….
Table 2. Pre-columbian periods in the Upper Loa and general characteristics.
Due to legal provisions of archaeological heritage (Law 17.288 of National Monuments)
materials could not be removed from the site, collected nor sampled for a complete
analysis (e.g. 14C dating, pottery analyses). Hence, the understanding of the settlement
pattern is preliminary and the chronology is relative, corresponding to judgements based
on accepted typological categories of the history and prehistory of the Upper Loa area
and the region of Antofagasta. Archaeologists with expertise on pottery and lithic
analyses were consulted about the findings encountered (Varinia Varela, Carole
Sinclaire, Itací Correa, Victoria Castro and Patricio de Souza, pers. com.).
4. PRESENTATION OF FINDINGS
At the Silala River, a total of 46 sites were recorded. Sixteen of these sites (35%) are
undefined as they could not be classified chronologically due to the lack of diagnostic or
characteristic materials on the surface. A further sixteen sites (35%) were classified as
pre-Columbian (PR), 11 as historical (H) (24%) and three as both pre-Columbian and
historical (PR/H) (6%).
Twenty-nine of the sites were found on the terraces (T2, T3, T4) of the Silala River
(63%), versus 17 (37%) on terrace (T1). Pre-Columbian and undefined sites had a
greater presence on terraces T2, T3 and T4, historical sites had a greater presence on
terrace T1 and pre-Columbian / historical sites were located almost evenly on both the
lowest and higher terraces (Figures 4 and 5).
Annex VI
99
8 Figure 4. Distribution of sites according to chronology and geography (T-1-T2-T3-T4). 6 3 6 2 10 13 5 1 024681012141618PRUndefinedHistoricalPR/HT2-T3-T4T19Figure5. 100
Annex VI
9
Figure 5. Archaeological sites within terraces (T1, T2, T3, T4).
Annex VI
101
10 When possible, pre-Columbian sites were classified into sub-periods based on their materials (Table 3). See also Appendix B for a summary of all the sites. Site Chronology Location 1 Sub-period Isolate finding Siarq 13 PR T1 Formative Siarq 14 PR T1 Undefined X Siarq 15 PR T1 Undefined Siarq 16 PR T1 Undefined X Siarq 17 PR T3 Formative Siarq 18 PR T3 Formative/LIP Siarq 19 PR T3 Formative/LIP Siarq 26 PR T3 LIP Siarq 27 PR T2 LIP Siarq 31 PR T1 LIP X Siarq 32 PR T1 LIP X Siarq 34 PR T3 Undefined Siarq 35 PR T3 LIP Siarq 41 PR T4 LIP Siarq 43 PR T3 Undefined Siarq 45 PR T4 Formative Table 3. Pre-Columbian sites encountered in the Silala River, their location within the basin and their sub-period. The most characteristic pre-Columbian features are fragments of pottery, lithic flakes, fragments of bones and Geoffroea decorticans6 husks, within stone structures or caves (Figure 6). Sometimes these features are isolated on the surface (Table 3). 6 Chañar is a tree of the family Leguminoseae that provides wood suitable for construction and crafts, but also has a fruit whose pulp is sweet and rich in carbohydrates. 102
Annex VI
11
Figure 6. Pre-Columbian findings a) stone structures b) grinding stone c) pottery d) lithic flakes
and bones.
At at least six of the historical sites, plastic items or batteries were found and these can
be attributed to the later decades of the XXth century. The other historical sites might
have earlier chronologies as cans, glasses and nails might represent earlier centuries.
Historical archaeology on routes from the coast to Calama, provides evidence of cans,
glasses and nails within the XIXth century (Borie, 2014). Also some of these sites
present pottery that could represent earlier historical periods (Figure 7).
A B
C D
Annex VI
103
12 Figure 7. Broken bottle associated with a pottery fragment. The presence of stone structures without archaeological material on the surface, classified as Undefined, might represent shelters for short stays, though without archaeological excavations we cannot establish the timing of occupation and/or construction. 5.DISCUSSION OF FINDINGSThis is the first baseline or systematic archaeological survey that has been made at the Silala River. It forms an important database to complement previous studies carried out in the Upper Loa (Aldunate et al., 1986; Berenguer et al., 2011; Castro et al., 1984; De Souza, 2014). The Silala River within the local geography has been defined as an important route for peoples and products. Martínez (1985: 111) points out that “to get from Lípez to Aiquina there are several mountain passes. Two of them have traditionally been the most used ones, due to the advantages that they present: Portezuelo (narrow pass) de Linzor and Siloli [Silala]. Both, located above of Toconce and Inacaliri respectively.” In fact, during the Mallku phase within the Late Intermediate Period (ca. 1000 BP) there is strong evidence of an altiplanic wave of migration as well as intense relationships and traffic between the Lípez area (department of Potosí, southwest of Bolivia) (Martínez, 1985) and the Upper Loa Region. These relationships are also well recorded in the colonial documents (encomiendas and doctrines) Lípez and Atacamas appeared always intermixed, confused (Martínez, 2011). Thus, ethnography carried out with San Pedro station inhabitants recognize this past connection with Lípez, both in time and in space 104
Annex VI
13
(Martínez, 1986; Mora, 2015). Also within historical periods, there are several roads
reported within the area, which had an intense traffic of couriers, muleteers, and
caravans of llamas until they fell into disuse after the inauguration of the railroad from
Antofagasta to Bolivia (1892) (Berenguer et al., 2011; Sanhueza, 1992). Until recently,
Castro (pers. com.) and police officers of the area, report Bolivian herders that from
time to time are found on the Chilean side looking for their animals.
Hence, the presence and distribution of several pre-Columbian sites on the Silala River
might reflect the use of this path for at least the last 1500 years. This suggests that this
route sustained water and associated biotic resources making it feasible for people to
transit the area as well as settling sporadically. The settlement patterns and
archaeological evidence encountered suggest the presence of water through a millennial
scale period, and the changes in the watershed through time could have influenced the
preservation and/or the differentiated use of the space by human populations (T1 versus
T2, T3 or T4).
Chronologically, pre-Columbian materials found establish a Late Formative period
occupation (ca. 1500 BP) on site Siarq 17, based on the only complete lithic artefact
found. This is a white silica arrowhead with narrow peduncle, triangular limb and sharp
fins (De Souza, pers. com.). Also, pottery fragments found at different sites might
correspond to the Late Formative period (ca. 1500 BP), Late Intermediate Period in its
Mallku phase (ca. 1000 BP) and Historical period (V. Varela, C. Sinclaire, I. Correa and
V. Castro, pers. com.).
Other evidence that can give us clues to the chronology and movements of products
from Atacama to Lípez is the presence of chañar or Geoffroea decorticans husks on pre-
Columbian sites. This fruit allows us to infer occupations later than 900 BP, according
to recent studies carried out in the Loa and Salar de Atacama area (McRostie ms). It was
a food highly prized by the Atacama and Altiplanic populations as it is cataloged as
“sweet” (Martínez, 1998).
With respect to a possible Inka road in the area, we found remains of a well-defined
path that goes from the house of the “The Antofagasta (Chili) and Bolivia Railway
Company Ltd.” (FCAB), to the southern upper terrace, which has a stone retaining wall,
associated stone structures and a kind of sanctuary or “animita” i.e. tiny house structure
with remains of charcoal (Siarq 8). However, no surface material was found and
therefore it could not be ascribed to any given period. It is likely to have formed a part
of ancestral footpaths created by San Pedro and Lípez herders.
Future research regarding settlement patterns, architecture, routes as well several
archaeological materials (e.g. archaeobotanical, zooarchaeological, lithics) should refine
this pre-Columbian sequence.
Annex VI
105
14 The presence of historical records may be evidence of ethnographic herding people that commonly circulated between the Upper Loa and Bolivia as referred to above. Most of these sites have eroded glasses, cans and nails. These could reflect an occupation until the first half of the XXth century, when FCAB replaced these routes (1892), the conditions downstream of this study area changed and the proletarianization of the communities led to an abandonment of the area and traditional ways of life (Berenguer et al., 2011; Carrasco, 2014; Mora, 2015; Villagrán and Castro, 1997). Plastics in form of paintbrushes, brooms and clothes might be part of modern activities related to frontier delimitation and water management. Ultimately the frequency and density of the sites found are consistent with the archaeological and ethnographical settlement pattern described for the area. Stone structures generally attached to rocks or caves would be part of semi-permanent or transient settlements that could serve as “beds” (allow stays of up to two weeks), “balconies” (used for one day, occasionally two) and / or mounds to protect themselves from the sun and wind or to hunt, while herding (Villagrán and Castro, 1997). The site Siarq 35 could reflect a “bed” completely closed and with large recessed spans. In addition, the presence of pottery fragments, chañar and abundant coprolites (dissecated feces) of camelids reflect its pre-Columbian herding character (Figure 8). 106
Annex VI
15
Figure 8. Siarq 35. Stone structure, pottery fragments and chañar husk.
Annex VI
107
16 With regard to the locations of the sites, as shown in Figure 5, it could be that the limited number of pre-Columbian sites and materials (four of six are isolated findings) found on terrace T1, is due to the destruction of the sites and/or the removal of archaeological during the construction of the road and channeling and intake works on the Silala River. However, it could also be that the course of the Silala River has had successive floods for millennia, which may have erased much of the evidence corresponding to any settlements on terrace T1 and/or that the flow of the Silala used to be higher than today and that eventually no sites could be maintained there. In fact, the sedimentary section SL16 examined by Latorre and Frugone (2017), taken from T2 at the conjunction of the Silala River and Quebrada Negra confirms that the river has been flowing more or less on its present route and cutting the ravine for over 8500 years. During that time water levels rose on several occasions and a number of wetlands developed and existed alongside the river. From the unit analyzed they established that within ca. 8500-1900 cal yr BP a bofedal, or high altitude wetland was present in that area. Then from ca. 1900 to 530 cal yr BP there were extensive river floods that evolved into short lived wetlands. Close to the edge of T2 where the sedimentological section SIL16 was located, there are several archaeological sites on T3 (Siarq 17, 18, 19), which present stone structures, pottery (Late Intermediate Period Mallku ca. 1000 BP), Geoffroea decorticans (chañar), as well as the only lithic arrowhead encountered (ca. 1500 BP) (Figure 9). Hence T3, would have been an ideal location nearby these wetlands or bofedales, to exploit the forage resources by groups that move with caravans of camelids, or for birdlife hunting and collection of vegetal materials. 108
Annex VI
17
Figure 9. Site Siarq 17. White silica arrowhead and stone structures.
Annex VI
109
18 Also, the >8.5 ka date reported by Latorre and Frugone (2017) is significant and might indicate potential Archaic settlements (ca. 9000 - 3000 BP). Archaic hunter-gatherers of these highlands are usually found near palaeolakes and/or palaeowetlands as occur in the nearby area of Ojos de San Pedro (De Souza, 2014). Within this survey, some sites with just lithic flakes and nearby Quebrada Inacaliri could represent evidence of these early hunter gatherers (e.g. Siarq 41, 43, 44, 45). 6.CONCLUSIONSThe objective of this report was to evaluate archaeological occupation in the environs of the Silala River and its hydrological history. Forty-six archaeological sites and isolated findings, identified to different chronological periods were encountered. At least 16 were classified as pre-Columbian, 11 of them have an almost secure adscription to the Formative and Late Intermediate Period, that is to say they most likely reflect an occupation within 1500 and 550 years before present. The other five pre-Columbian sites do not have diagnostic elements; hence radiocarbon dates and typological analyses could be carried out to refine this relative sequence in the future. Another 16 sites were classified as Undefined because they do not show any diagnostic feature in the surface but just stone structures. Three sites may represent pre-Columbian and Historical occupations whilst the other sixteen are mainly Historical based on their surface materials. The low frequency of pre-Columbian sites on Terrace 1 of the Silala could be due to changes in the river flow through past millennia. The tight relationship between archaeological sites and palaeo-wetlands demonstrates a link between humans and water-biotic resources since at least 1500 years before present. However, sedimentology and hydrological studies reveal that wetlands were available since at least 8500 years before present. The archaeological record supports the historical and ethnographic evidence related to movements and close relationship of the Atacama and Bolivia Highlands. Future research might refine the chronologies described in this study as well reveal new sites within this area. 110
Annex VI
19
7. REFERENCES
Aldunate, C., Berenguer, J., Castro, V., Cornejo, L., Martínez, J.L. and Sinclaire, C.,
1986. Sobre la cronología del Loa superior. Revista Chungará, 16/17, 333-346.
Berenguer, J., 2011. Qhapaq Ñan. Las Rutas del Inca en el Norte de Chile. Consejo de
Monumentos Nacionales, 2° Edición, Santiago, Chile.
Borie, C., 2014. De la Pampa a la Costa y de la Costa a la Pampa. Estudio de un
Espacio Clave de la Ruta entre Cobija y Calama. Memoria para optar al Título de
Arqueólogo, Facultad de Ciencias Sociales, Departamento de Antropología,
Universidad de Chile, Santiago, Chile.
Castro, V., Aldunate, C. and Berenguer, J., 1984. Orígenes altiplánicos de la Fase
Toconce. Estudios Atacameños, 7, 159-178.
Carrasco, A., 2014. Entre dos aguas: Identidad moral en la relación entre corporaciones
mineras y la comunidad indígena de Toconce en el Desierto de Atacama. Revista
Chungará, 46(2), 247-258.
De Souza, P., 2014. Tecnología Lítica, Uso del Espacio y Estrategias Adaptativas de
los Cazadores-Recolectores del Arcaico Medio en la Cuenca Superior del río Loa
(7000-5000 AP). Nuevos Aportes para la Comprensión de los Procesos Culturales de
las Poblaciones Arcaicas de los Andes Centro-Sur. Tesis para optar al grado de Doctor
en Antropología, mención Arqueología. Universidad Católica del Norte y Tarapacá. San
Pedro de Atacama, Chile.
Grosjean, M., Santoro, C., Thompson, L., Núñez, L. and Standen, V., 2007. Mid-
Holocene climate and culture change in the South Central Andes. In: Anderson, D.,
Maasch, K. and Sandweiss, D. (Eds.), Climate Change and Cultural Dynamics: A
Global Perspective on Mid-Holocene Transitions, pp. 51-115. San Diego, Academic
Press.
Latorre, C. and Frugone, M., 2017. Holocene Sedimentary History of the Río Silala
(Antofagasta Region, Chile). (Vol. 5, Annex IV).
Martínez, J.L., 1985. La formación del actual pueblo de Toconce (Siglo XIX). Revista
Chungará, 15, 99-124.
Martínez, J.L., 1986. Los grupos indígenas del Altiplano de Lípez en la sub-región del
Río Salado. Revista Chungará, 16/17, 199-201.
Martínez, J.L., 1998. Pueblos del Chañar y el Algarrobo: los Atacamas en el Siglo
XVII. Santiago: DIBAM.
Annex VI
111
20 Martínez, J.L., 2011. Gente de la Tierra de Guerra. Los Lipes en las Tradiciones Andinas y el Imaginario Colonial. Lima: DIBAM y PUCP. Mora, G., 2015. Hacer Familia con la Cuenca de San Pedro – Inacalari. Programa de Magíster de Asentamientos Humanos y Medio Ambiente, Pontificia Universidad Católica de Chile. Núñez, L., Grosjean, M., and Cartajena, I., 1999. Un ecorefugio oportunístico en la Puna de Atacama durante eventos áridos del Holoceno medio. Estudios Atacameños, 17, 125-174. Núñez, L., Grosjean, M. and Cartajena, I., 2005. Ocupaciones Humanas y Paleoambientes en la Puna de Atacama. Universidad Católica del Norte-Taraxacum, San Pedro de Atacama. Núñez, L. and Santoro, C., 2011. El tránsito arcaico-formativo en la circumpuna y valles occidentales del centro sur andino: hacia los cambios “neolíticos”. Revista Chungará, 43, 487-530. Sanhueza, C., 1992. Tráfico caravanero y arriería colonial en el siglo XVI. Estudios Atacameños, 10, 173-187. Santoro, C., Capriles, J. M., Gayo, E. M., de Porras, M. E., Maldonado, A., Standen, V. G., Latorre, C., Castro, V., Angelo, D., McRostie, V., Uribe, M., Valenzuela, D., Ugalde, P., Marquet, P. A., 2016. Continuities and discontinuities in the socio-environmental systems of the Atacama Desert during the last 13,000 years. Journal of Anthropological Archaeology, in press. Sinclaire, C., 2004. Prehistoria del período Formativo en la cuenca alta del río Salado (Región del Loa Superior). Revista Chungará, 36 (Supl. esp. 2), 619-639. Villagrán, C. and Castro, V., 1997. Etnobotánica y manejo ganadero de las vegas, bofedales y quebradas en el Loa superior, Andes de Antofagasta, Segunda Región, Chile. Revista Chungará, 29(2), 275-304. 112
Annex VI
APPENDIX A
ARCHAEOLOGICAL FILECARDS SILALA REPORT
Annex VI Appendix A
113
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 1Small cave to the south of the Silala riverbed. Within the cave there are some cans and soot in the roof. On the west of the cave there is a small stone structure in which two pottery fragments were found 60018875625503845South, within the riverbed of the SilalaDisturbed by historical interventions related to the river canalization and the roadpre Columbian-HistoricalName 114
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Disturbed by historical interventions related to the river
canalization and the road
pre Columbian-Historical
Stone structures. There is a lot of intervention and
garbage such as cans, brooms and clothes. One piece of
pottery might reflect pre Columbian occupation
600214
75625322
3853
South, within the river bed of the Silala. They are located
almost at the border with Bolivia in front of the FCAB
intake
Siarq 2
Annex VI Appendix A
115
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologyGoodSiarq 3Cave with stone enclosure. Without any surface materialsUndefined60025875652913881South of the Silala, within the middle of the terraceName 116
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
600076
7564946
3872
South of the Silala River, above the terrace
Good
Historical
Siarq 4
Cave with stone enclosure. With cans and wood tables on
the surface
Annex VI Appendix A
117
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronology75649413867South of Silala River. This structure is within what local people have mentioned as the Inca roadGoodUndefinedSiarq 5Circular stone structure without surface materials600038Name 118
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 6
Circular stone structure without surface materials
600012
7564875
3864
South of the Silala River, above the terrace
Good
Undefined
Annex VI Appendix A
119
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river area Conservation of the siteRelative chronologyNorth of the road that goes through the Silala riverbed GoodHistoricalSiarq 7Corral. Structures within a cave. Some cans, glass and horse coprolites60010975651333859Name 120
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
South of the Silala, going up to the terrace there is a road
with a wall made of stones. Also there are other stone
structures without any surface materials
600036
7564965
3866
South of the Silala, going up to the terrace
Siarq 8
Good
Undefined
Annex VI Appendix A
121
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronology60002375649773857South of the Silala, within the rocks of the cliffGoodSiarq 9Cave closed with a tall stone wall. Inside there were cans and animal bones with straight cuttings, diagnostic of a historical periodHistoricalName 122
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
599998
7564931
3858
South of the Silala riverbed
Good
Siarq 10
Rectangular stone structures with plenty of historical
materials such as iron, wooden tables, nails etc.
Immediatly southeast of the FCAB intake, so probably
related to FCAB intake activities
Historical
Annex VI Appendix A
123
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronology59982575645913838South of the Silala riverbedGoodHistoricalSiarq 11Small stone structure attached to the rock south of the Silala riverbed. There is glass and charcoal on the surfaceName 124
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
7564565
3836
South of the Silala riverbed
Good
Undefined
Siarq 12
Small stone structure attached to the rock south of the
Silala riverbed. No material on the surface
599784
Annex VI Appendix A
125
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologyStone wall that goes from the car road within the Silala riverbed up to the rock of the canyon, where there is a cave in which lithic flakes and pottery were found3829South of the Silala riverbedGoodpre ColumbianSiarq 135996867564451Name 126
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
South of the Silala riverbed
Bad
pre Columbian
Siarq 14
Isolated finding. Lithic flake (part of the lithic debris
when making an artefact)
599661
7564418
4234
Annex VI Appendix A
127
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 15Goodpre ColumbianOn the south bank of the river there are some terraces built with stones. These could be agricultural fields. There is no surface material, though nearby are siarq 13 and 1459964775644203828South of the Silala riverbedName 128
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 16
Isolated finding. Lithic flake (part of the lithic debris
when making an artefact)
pre Columbian
599283
7563978
4197
South of the Silala riverbed
Bad
Annex VI Appendix A
129
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 17Stone structure enclosed with natural rock. At the other side of the rock a white silica arrowhead with narrow peduncle, triangular limb and sharp fins was found, and nearby there was a Geoffroea decorticans husk, from a native tree that grows in the lower oases of Chile and a basic staple for pre Columbian communities59925875638984212South of the Silala, in the terrace near the confluence with Quebrada NegraGoodpre Columbian. The white arrowhead might correspond from Late Formative Period to Late Period (ca. 1500-500BP), whilst the chronology of the Geoffroea is not clear, it seems to appear in the area by ca.1000 BPName 130
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Good
pre Columbian. Pottery might correspond to Late
Formative or Intermediate Period (ca.1500-1000BP)
Siarq 18
Stone structure some meters southeast of Siarq 17. It has
pottery and animal (camelid?) coprolites inside
599280
7563905
4218
South of the Silala, in the terrace near the confluence with
Quebrada Negra
Annex VI Appendix A
131
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronology4233South of the Silala, in the terrace near the confluence with Quebrada NegraGoodpre ColumbianSiarq 19Two circular stone structures. There are some lithic debris as well pottery in the surroundings5993127563829Name 132
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
South of the Silala, in the terrace near the confluence with
Quebrada Negra
Good
Undefined
Siarq 20
Cave with surrounding stone wall. Without surface
material
599220
7563807
4227
Annex VI Appendix A
133
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologyGoodHistoricalSmall cave with stone wall. Some glass and cans. Related to Siarq 2059921075637994228Within Quebrada NegraSiarq 21Name 134
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 22-23
Southeast of Quebrada Negra and within the Silala
riverbed there are several stone structures. Just one
fragment of pottery was found among other historical
materials
pre Columbian-Historical
598962
7563792
4203
South of the Silala riverbed
Regular, affected by the road
Annex VI Appendix A
135
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 24Stone wall. No material on the surface59838975638664202South of the Silala, going up to the terraceGoodUndefinedName 136
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
7563878
3900
South of the Silala, going up to the terrace
Good
Undefined
Siarq 25
Stone wall. No material on the surface
596784
Annex VI Appendix A
137
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronology4097South of the Silala, going up to the terrace, within the police area Goodpre ColumbianSiarq 26Small cave with pottery and handstones and fragmented bones5967837563879Name 138
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
South of the Silala River
Good
pre Columbian
Siarq 27
Stone structures attached to a cave. Some pottery on the
surface
5976731
7563877
4087
Annex VI Appendix A
139
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 28GoodHistoricalStone wall. No material on the surface. Seems to be part of a modern fence59668875638114108South of the SilalaName 140
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 29
Cave with stone structure enclosure. No surface material
Undefined
596654
7563833
4080
South of the Silala, in front of the Inacaliri Police Station
Good
Annex VI Appendix A
141
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 30Stone structure59661975640834002North of the Silala, going up to the terraceGoodUndefinedName 142
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Bad
Undefined
Siarq 31
Isolated fragment of pottery. Area highly disturbed by
modern construction
596396
7563838
4058
In front of the Inacaliri Police Station
Annex VI Appendix A
143
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronology4055In front of the Inacaliri Police StationBadUndefinedSiarq 32Isolated fragment of pottery. Area highly disturbed by modern construction5964017563838Name 144
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
Relative chronology
West of Quebrada Negra, near the road
Good
Undefined
Siarq 33
Rock cave with stone structure
599259
7563715
3853
Annex VI Appendix A
145
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologyGoodpre ColumbianThree rock structures and some lithic flakes on the surface59916075638253850Southwest of Quebrada Negra to the southeast of the Silala RiverSiarq 34Name 146
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 35
Cave enclosed with stone structures and large recessed
spans with a frame as doorway. Inside and outside of this
structure there are pottery fragments, coprolites
(camelids), charcoal and husks of Geoffroea decorticans
pre Columbian. Pottery and Geoffroea decorticans
suggest a maximum age of 1500 BP and minimum of 550
BP
598245
7563778
3825
Southeast of the Silala above the terrace
Good
Annex VI Appendix A
147
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 36Semicircular stone structure. Without surface material59824375637753833Southeast of Silala above the terrace GoodUndefinedName 148
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 37
7563493
3791
Southeast of Silala above the terrace, in front of
recently installed Cell-phone Tower
Good
Undefined
Semicircular stone structure. Without surface material
596634
Annex VI Appendix A
149
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronology3678South of Silala RiverBadUndefinedSiarq 38Rectangular stone structure. Without surface material5962917563654Name 150
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
South of Silala River. Downstream of the Inacaliri Police
Station
Good
Undefined
Siarq 39
Stone structure attached to rock. No surface material
596150
7563358
3668
Annex VI Appendix A
151
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 40Goodpre Columbian-HistoricalBig cave with structures in its interior. On the surface a husk of peach, Geoffroea decorticans and a grinding stone was found59750175643443820Northwest of the Silala River. It is located in a ravine that flows into the Silala. It has a meadow in front of itName 152
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 41
Semicircular stone structure. With plenty of pre
Columbian material on the surface. Pottery, lithic flakes,
bones
pre Columbian. Pottery might represent Late Intermediate
period (ca. 1000 BP)
597018
7564387
3811
Northwest Silala River
Good
Annex VI Appendix A
153
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 42Semicircular stone structure59661775642473785 Northwest of the Silala River. Above rocks of a ravine that flows into the SilalaGoodUndefinedName 154
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Siarq 43
Cave with stone structures in the interior. Some lithic
blades on the surface
596620
7562450
3781
Northwest of the Silala River, within an inactive ravine
that flowed into the river
Good
pre Columbian
Annex VI Appendix A
155
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologySiarq 44Cave with stone structures. Cans and soot on the roof59663475634203790Northwest of the Silala River, within an inactive ravine that flowed into the riverGoodHistoricalName 156
Annex VI Appendix A
Name of the site
Type and
description
UTM Datum WGS
East
North
m.a.s.l.
Location within the
river area
Conservation of the
site
Relative chronology
Northwest of the Silala River, within an inactive ravine
that flowed into the river
Good
pre Columbian
Siarq 45
Stone structures with many fragments of pottery inside
596576
7564256
3782
Annex VI Appendix A
157
Name of the siteType and description UTM Datum WGSEastNorthm.a.s.l.Location within the river areaConservation of the siteRelative chronologyGoodHistoricalStructures within areas of big rocks59616475636283694Northwest of the Silala River, downstream of the Inacaliri Police StationSiarq 46 158
Annex VI Appendix A
1
APPENDIX B
SUMMARY OF THE SITES
Site Chronology Location Sub-period Finding
Siarq 1
Pre
Columbian
(PR)/Historic
al (H) T1
Und.-XXth
century
pottery, cans, batteries, stone
structure
Siarq 2 PR/H T1
Und.-XXth
century
clothes, plastics, pottery, stone
structure
Siarq 3 Undefined T2 Undefined stone structures
Siarq 4 Historical T3 XXth century tables, cans, stone structure
Siarq 5 Undefined T3 Undefined stone structures
Siarq 6 Undefined T3 Undefined stone structures
Siarq 7 Historical T1 Undefined
glasses, cans, horse or mule
feces, stone structures
Siarq 8 Undefined T2 Undefined stone structures
Siarq 9 Historical
T1
Undefined
animal bones with straight
historical cuttings, cans, stone
structure
Siarq 10 Historical T1 XXth century iron, nails, tables
Siarq 11 Historical T1 Undefined glass, chaorcal, stone structure
Siarq 12 Undefined T1 Undefined stone structures
Siarq 13 PR
T1
Formative
pottery, lithic flakes, stone
structure
Siarq 14 PR T1 Undefined one isolated andesite flake
Siarq 15 PR T1 Undefined agricultural terraces
Siarq 16 PR T1 Undefined one isolated andesite flake
Siarq 17 PR T3 Formative
pottery, lithic arrowhead, chanar
husk, stone structure
Annex VI Appendix B
159
2 Site Chronology Location Sub-period Finding Siarq 18 PR T3 Formative/ Late Intermediate Period (LIP) pottery, chanar husk, dissecated camelid coprolites, stone structure Siarq 19 PR T3 Formative/LIP lithic flake, stone structures Siarq 20 Undefined T4 Undefined stone structures Siarq 21 Historical T4 Undefined glasses, cans, stone structure Siarq 22 Historical T1 Undefined glass, pottery, stone structure Siarq 23 Historical T1 Undefined glass, pottery, stone structure Siarq 24 Undefined T2 Undefined stone structures Siarq 25 Undefined T1 Undefined stone structures Siarq 26 PR T3 LIP pottery, bones, grinding stones, stone structures Siarq 27 PR T2 LIP pottery, stone structure Siarq 28 Historical T3 XXth century stone structures attach to police fences Siarq 29 Undefined T2 Undefined stone structures Siarq 30 Undefined T2 Undefined stone structures Siarq 31 PR T1 LIP one isolate pottery fragment Siarq 32 PR T1 LIP one isolate pottery fragment Siarq 33 Undefined T4 Undefined stone structures Siarq 34 PR T3 Undefined lithic flake, stone structures Siarq 35 PR T3 LIP pottery, chanar husk, dissecated camelid coprolites, enclosed stone structure Siarq 36 Undefined T3 Undefined stone structures Siarq 37 Undefined T4 Undefined stone structures Siarq 38 Undefined T1 Undefined stone structures Siarq 39 Undefined T1 Undefined stone structures Siarq 40 PR/H T3 Undefined peach husk 160
Annex VI Appendix B
3
Site Chronology Location Sub-period Finding
Siarq 41 PR T4 LIP
pottery, bones, charcoal, stone
structures
Siarq 42 Undefined T4 Undefined stone structures
Siarq 43 PR T4 Undefined
lithic flakes, bones, stone
structures
Siarq 44 Historical T4 Undefined cans, stone structure
Siarq 45 PR T4 Formative pottery, stone structure
Siarq 46 Historical T1 XXth century
cans, glasses, plastics, stone
structures
Annex VI Appendix B
161
162
Annex VII
Muñoz, J.F., Suárez, F., Fernández, B., Maass, T., 2017. Hydrology of the Silala River Basin
163
164
Annex VII
HYDROLOGY OF THE SILALA RIVER BASIN
José Francisco Muñoz (PhD)
Professor, Pontificia Universidad Católica de Chile
Francisco Suárez (PhD)
Associate Professor, Pontificia Universidad Católica de Chile
Bonifacio Fernández (PhD)
Professor, Pontificia Universidad Católica de Chile
Tamara Maass, Civil Engineer
May, 2017
Annex VII
165
GLOSSARY This glossary of hydrological terms is based on the following: http://www.wmo.int/pages/prog/hwrp/publications/international_glossary/…http://www.nws.noaa.gov/om/hod/SHManual/SHMan014_glossary.htmhttp://www.geo.utexas.edu/faculty/jmsharp/sharp-glossary.pdfAir Moisture: Water vapour content of the air. Alluvial Deposits: Clay, silt, sand, gravel, pebbles or other detrital material deposited by flowing water. Alluvial River: River in which the channel is made up of mobile sedimentary materials deposited by flowing water. Alluvial rivers are self-formed; their channels are shaped by the flows that they experience, and the ability of these flows to erode, deposit, and transport sediment. Aquifer: Geological formation capable of storing, transmitting and yielding exploitable quantities of water. Artesian Well: Well tapping a confined aquifer whose piezometric surface lies above the ground surface. Atmospheric Circulation: Large-scale movement of air. Atmospheric Moisture: Water vapour content of the air in the atmosphere. Bedload: The coarser fraction of the sediments transported downstream by a river, which is moved by traction along the bed of the channel by rolling, sliding and saltation. Cascade Units: Exist where steep channel gradients occur, where the channel is dominated by boulders and cobbles and channel-spanning pools do not exist. Cascade-Step/Pool Morphology: Combination of Cascade and step/pool channel types. Conductivity: Hydraulic Conductivity is a property of a porous medium which, according to Darcy’s law, relates the specific discharge to the hydraulic gradient. Confined Aquifer: A confined aquifer is one containing groundwater that is under pressure exceeding atmospheric pressure. The recharge area to a confined aquifer is at some distance and is unconfined but at higher elevation than the confined aquifer. Discharge: Volume of water flowing per unit time, for example through a river cross-166
Annex VII
or from a spring or a well.
El Niño South Oscillation: Periodic fluctuation in sea surface temperature and the air
pressure of the overlying atmosphere, associated with the equatorial Pacific Ocean.
Evaporation: Process by which water changes from liquid to vapour.
Evapotranspiration: Combination of evaporation from free water surfaces and
transpiration of water from plant surfaces to the atmosphere.
Fluvial Deposits: Sediments deposited by fluvial process.
Gauge: (verb) To estimate an amount by using a measuring device.
Groundwater: Subsurface water occupying the saturated zone (i.e. where the pore
spaces (or open fractures) of a porous medium are full of water).
Headwater: A tributary stream of a river, close to or forming part of its source.
Infiltration: The movement of water from the surface of the land into the subsurface.
Isohyet: Line joining points of equal precipitation for a given period of time.
Nickpoint: Part of the river where there is a sharp change in channel slope.
Pan Evaporation: Evaporation measured as a change of water level in a standardized
pan, which measures the integrated effect of radiation, wind, temperature and humidity
on the evaporation from an open water surface.
Perennial River: River that flows continuously throughout the year.
Pyroclastic Fall Deposits: Deposits of material that has been ejected from a volcanic
eruption into the air, such as an ash fall or tuff.
Recharge Zone: Area which contributes water to an aquifer either by direct infiltration
or by runoff and subsequent infiltration.
Resistivity: Electrical resistivity is a measure of the resisting power of a specified
material to the flow of an electric current.
River Basin: Area having a common outlet for its surface runoff.
River Stage: Water level of the river at a specific location.
Saturation: (in the context of the atmosphere) The state in which air contains the
maximum amount of water vapor that it can hold at a specific temperature and air
pressure.
Saturation: (in the context of soils or rocks) The state in which the pore spaces or open
fractures of a porous medium are full of water – the definition of ‘groundwater’.
Annex VII
167
Sediments: Material transported by water either in suspension or as bedload from the place of origin to the place of deposition. Spatial Variation: When a quantity that is measured at different spatial locations exhibits values that differ across the locations. Spring: Place where groundwater emerges naturally from the rock or soil. Step/Pools: Are composed of channel-spanning pools and boulder/cobble steps that typically cause subcritical flow in the pool and supercritical flow over the steps. Streamflow: General term for water flowing in a watercourse. Transmissivity: (in the context of groundwater) Rate at which water is transmitted through a unit width of an aquifer under a unit hydraulic gradient. Transpiration: Process by which water from vegetation is transferred into the atmosphere in the form of vapour. Tributary River (Tributaries): Watercourse that flows into a larger watercourse or into a lake. Unconfined Aquifer: Saturated water-bearing formation, which has a water table open to the atmosphere through permeable rock. Watershed: Basin syn. drainage basin, catchment, river basin, Area having a common outlet for its surface runoff. Weir: Overflow structure that may be used for controlling upstream water level, and/or for measuring discharge. Wells: Any artificial excavation or borehole constructed with the aim of either exploring for or producing groundwater, or injection, monitoring or dewatering purposes. Wetland: Areas under or contiguous to open water or with a shallow water table, including swamps, marshes, bogs, wet meadows, river overflows, mud flats, and natural ponds. Wetlands are typically characterized by water-loving vegetation (phreatophytes or, in areas with brackish water, halophytes). 168
Annex VII
TABLE OF CONTENTS
1 INTRODUCTION ..................................................................................................... 1
1.1 Study zone .......................................................................................................... 1
1.2 Objective ............................................................................................................ 3
1.3 Summary of the methodology ............................................................................ 3
1.4 Structure of the report......................................................................................... 3
2 SUMMARY AND CONCLUSIONS ........................................................................ 4
3 STUDY AREA .......................................................................................................... 8
3.1 Catchment ........................................................................................................... 8
3.2 History of water development concessions in the basin................................... 16
4 CLIMATE AND METEOROLOGY ...................................................................... 24
4.1 Geographical aspects ........................................................................................ 24
4.2 Large-scale atmospheric circulation ................................................................. 25
4.3 Moisture and local precipitation conditions on the Altiplano .......................... 29
4.4 Precipitation and sources of moisture .............................................................. 30
4.5 Inter-annual variability ..................................................................................... 33
5 EVOLUTION OF THE BASIN AND THE RAVINE ............................................ 35
5.1 Geological history and evolution of the Silala River basin geology and
geomorphology ............................................................................................................ 35
5.2 Geology and geological processes ................................................................... 38
5.3 Fluvial geomorphology .................................................................................... 39
5.4 Fluvial processes of the Silala River ................................................................ 44
6 HYDROLOGICAL CHARACTERISTICS ............................................................ 46
6.1 Hydrological processes in the basin ................................................................. 46
6.2 Instrumentation in the basin to measure hydrometeorological and hydrological
processes ...................................................................................................................... 47
6.3 Precipitation...................................................................................................... 50
6.3.1 Spatial variation of precipitation ............................................................... 51
Annex VII
169
6.3.2 Temporal variation of precipitation .......................................................... 54 6.4 Temperature...................................................................................................... 56 6.4.1 Spatial variation of temperature with altitude ........................................... 56 6.4.2 Temporal variation of temperature ........................................................... 57 6.4.3 Hourly temperature ................................................................................... 59 6.5 Evapotranspiration............................................................................................ 60 6.5.1 Pan evaporation regime ............................................................................. 62 6.5.2 Potential ET calculated from meteorological measurements in the study area ................................................................................................................... 65 6.5.3 Actual evapotranspiration from the Cajones and Orientales wetlands ..... 69 6.6 Infiltration ......................................................................................................... 74 6.7 River flow ......................................................................................................... 76 6.7.1 Annual and Monthly flow discharge ......................................................... 77 6.7.2 Daily flow discharge ................................................................................. 78 6.7.3 Base Flow Index........................................................................................ 79 6.8 Groundwater-surface water interactions .......................................................... 81 7 CONCLUSIONS ..................................................................................................... 85 8 REFERENCES ........................................................................................................ 90 APPENDIX A ................................................................................................................. 95 APPENDIX B ............................................................................................................... 101 APPENDIX C ............................................................................................................... 105 170
Annex VII
1
1 INTRODUCTION
The National Director of the Dirección Nacional de Fronteras y Límites del Estado
(DIFROL) of the Ministry of Foreign Affairs, Mrs. Ximena Fuentes, requested that the
Pontificia Universidad Católica de Chile (UC) undertake an interdisciplinary study
aimed at deepening the hydrological and hydrogeological knowledge of the
transboundary basin of the Silala River, located in the north of Chile.
This report presents an up-to-date physiographic, hydrological and hydrogeological
characterization of the Silala River basin. This study was led by Drs. José Francisco
Muñoz and Francisco Suárez and the report was written under the supervision and
instruction of Drs. Howard Wheater and Denis Peach.
1.1 Study zone
The area of the study is the Silala River basin, a transboundary watershed shared by
Bolivia (upstream) and Chile (downstream). The Silala River basin is located in the
Andean Plateau of the Atacama Desert, approximately 300 km northeast of
Antofogasta. The Silala River originates in Bolivian territory and flows towards the
Antofagasta Region in Chilean territory (Figure 1-1). The Silala River is one of the
main tributaries of the San Pedro River, which in turn is a tributary of the Loa River.
The Loa River is the longest Chilean river (440 km long) and the main watercourse in
the Atacama Desert. It drains to the Pacific Ocean where its outlet is located at latitude
21°26’ S.
Annex VII
171
2 Figure 1-1. The Loa River and its main tributaries. 172
Annex VII
3
1.2 Objective
The main objective of this study is to present an up-to-date physiographic, hydrological,
and hydrogeological characterization of the study area. The specific objectives are:
 To describe the climate of the Altiplano area where the Silala River basin is
located.
 To characterize the regional and local geology and geomorphology of the Silala
River basin.
 To characterize the hydrology of the region and of the Silala River basin,
including the temporal dynamics of precipitation, temperature, evapotranspiration and
river flow.
 To describe the hydrogeology of the Silala River basin.
1.3 Summary of the methodology
The characterization described in this report is based on a compilation of different
studies of the Silala River basin that were performed by Pontificia Universidad Católica
de Chile, the Chilean National Geology and Mining Service (“SERNAGEOMIN”) and
Arcadis between September 2016 and March 2017. In addition, the results from other
studies performed by the Chilean General Directorate of Water (Dirección General de
Aguas, abbreviated as DGA) were also compiled. The methodology of this report
consisted firstly of a review of the previous studies, summarizing the most important
results. Secondly, records from the meteorological and fluviometric stations located in,
and adjacent to, the Silala River basin were collected and key variables were analyzed,
such as precipitation, temperature and streamflow.
1.4 Structure of the report
The structure of the remainder of this report is as follows: Chapter 2 provides a
summary and the conclusions of the study; Chapter 3 provides backgound information
on the study area, i.e., it characterizes the Silala River basin and the history of the water
development concessions in the basin; Chapter 4 presents the climate and meteorology
of the study area; Chapter 5 presents the evolution of the basin and the ravines, where
the geological history and fluvial geomorphology is described; Chapter 6 presents the
hydrological characterization of the Silala River basin, in which different hydrological
processes such as precipitation, temperature, infiltration and river flow are explained
and quantified; Chapter 7 presents the main conclusions of this study. Finally, the
references cited in the report are shown in Chapter 8 and three appendices are hereto
enclosed.
Annex VII
173
4 2 SUMMARY AND CONCLUSIONS The main objective of this study is to present an up-to-date physiographic, hydrological, and hydrogeological characterization of the Silala River basin. This study presents an up-to-date characterization of the study area, and describes the climatic, geological, geomorphological and hydrological processes present in the basin. The Silala River basin is located on the western edge of the Andean Altiplano, and drains from Bolivian territory to the Antofagasta Region of Chile. Elevations in the basin headwaters generally exceed 4,000 m.a.s.l. The Silala River becomes one of the main tributaries of the San Pedro River, which flows into the Loa River. The Loa River is the longest Chilean river (440 km long) and the main watercourse in the Atacama Desert, draining from the Andes mountains to the Pacific Ocean. The waters of the Silala River originate from a series of springs located in the Orientales and Cajones wetlands in Bolivia, at more than 4,323 m.a.s.l. The waters from the Orientales wetland flow into a ravine which converges with the Cajones ravine to form a common ravine and associated stream called the Silala River ~750 m east of the Chile-Bolivia international border. The ravine of the Silala River crosses the international border at an elevation of ~4,277 m.a.s.l. The Silala River basin has a drainage area of 95.5 km2, of which 72.2% is located in Bolivia and the remainder is located in Chile (Alcayaga, 2017). The basin is defined as draining to a location known as Inacaliri, located ~5 km downstream of the Chile-Bolivia international border (596,453 E; 7,563,039 N datum WGS84-19S). In addition to the Bolivian springs that form the source of the perennial flows, a series of natural springs in Chilean territory also discharge their waters into the Silala River (Suárez et al., 2017). In addition, a well previously drilled on the Chilean territory for exploration purposes also discharges groundwater to the river, under artesian flow conditions (SPW-DQN well). The waters of the Silala River flow naturally towards Chile from the Bolivian highlands following the natural topographic gradient of the basin; the channel slope is relatively steep (5%), calculated from a nickpoint (7,000 m upstream of the Inacaliri Police Station) to the Inacaliri Police Station along the river (Alcayaga, 2017). The Silala River basin formed as the result of a series of geological events, including volcanic processes, which took place episodically over the past 12 Million years before present (Ma BP) (SERNAGEOMIN, 2017). The Silala River ravine was carved out by the action of the river. The first evidence of the existence of alluvial drainage tied to the Silala fluvial system is from the Lower Pleistocene (ca. 2.6-1.45 Ma BP); the second phase in the evolution of the Silala River system took place in the Late Upper Pleistocene – Lower Holocene (ca.12–7.6 ky BP). While glaciation has affected landscape processes at higher elevations, there is strong evidence that fluvial action was 174
Annex VII
5
the dominant process in the formation of the Silala River ravine (SERNAGEOMIN,
2017).
The principal geological units of Silala River basin, in terms of their hydrological
relevance, are: HU1 (Fluvial deposits), HU2 (Alluvial deposits), HU3 (Ignimbrite),
HU4 (Pyroclastic fall deposits), HU5 (Pleistocene andesitic lavas) and HU6 (weakly
permeable rock) (Arcadis, 2017). The units with greatest presence in the basin are HU5,
HU6, and HU2. HU6 and HU5 are composed mainly of volcanic rocks and HU2 is
composed of unconsolidated deposits of rounded stones, gravels, sands, and silts.
The Silala River ravine is between 10 and 100 m wide, with a mode of 20 m. It has four
erosion terraces (T1, T2, T3 and T4) with associated depositional periods that have
different degrees of development in the study area (Arcadis, 2017; Latorre and Frugone,
2017) indicating past river erosion levels. On one hand, the T2, T3 and T4 terraces were
developed over a period of some 8,400 years or more (Latorre and Frugone, 2017). On
the other hand, the youngest terrace is T1, was formed in the 20th century. The T1
terrace is found along the entire length of the ravine and shows evidence of aggradation
and incision, demonstrating that the river continued to be active geomorphologically
over the recent past. In addition, sedimentation studies have concluded that the Silala
River is currently functioning as an alluvial river with fully developed fluvial processes,
which shape the channel morphology through active sediment dynamics (Mao, 2017).
Although the Silala River is located in an arid environment, its regime is perennial
because it is a groundwater-dominated river, with a very high base flow index (BFI,
which is defined as the ratio of the base flow to the total flow). Its bed sediment
composition (armour ratio) is comparable with values typically found in rivers with
perennial regimes in humid environments (Mao, 2017).
The hydrological behavior of the Silala River basin is affected by climatic phenomena,
the surface topography of the basin, and its surface and subsurface characteristics. The
basin is located in an area that has a climate that has a strong seasonal cycle and high
inter-annual variability, but also changes on an hourly time scale with extreme
conditions of wind, temperature, radiation and precipitation. Therefore, understanding
the dominant processes that affect the water balance of the Silala River basin is
complex. For this reason, the various hydrological processes must be quantified on an
hourly or a daily basis.
Precipitable water can fall in liquid or solid form, depending on the temperature.
Infiltration in the basin is heterogeneous and depends on the geological units and the
permeability of the soil. It is a dominant process in the basin because the infiltration
rates of some units are larger than 1 metre/day, which is a large value compared with
the highest daily rainfall rates (15-20 mm/day). The most permeable geological units
Annex VII
175
6 tested are HU2 (Alluvial) and HU3 (Ignimbrite). Hydrological response varies according to climatic conditions. For example, after a minor rainstorm, depending on the meteorological conditions, the water that falls on the basin can evaporate quickly from the ground surface or from the unsaturated (vadose) zone, with little or no contribution to runoff. In situations where air or surface temperatures fall below zero, precipitation may fall in liquid or solid form and liquid water could freeze. If this is the case, then a small fraction of the water will infiltrate and significant sublimation may occur. For major rainfall events, precipitation will infiltrate and groundwater will be recharged, while localized surface runoff may also occur. The river flow is dominated by groundwater inflows, which include the spring-fed Orientales and Cajones sources of the Silala River in Bolivia, and a large number of springs entering the river as it flows through the ravine in Chile. From the above description it can be seen that temperature and precipitation control important hydrological processes, such as streamflow. The daily variations of temperature are the likely explanation of daily fluctuations that are observed in the Silala River flows. Precipitation in the basin is mainly caused by convective activity. More than 90% of the precipitation in the basin occurs between January and March, as a result of the significant atmospheric moisture coming from the east. During the rest of the year the atmospheric moisture in the area decreases, as dry winds blow from the west. The annual precipitation at the Silala meteorological station is 99  74 mm (mean  standard deviation). The large standard deviation is explained by the natural inter-annual variability, which is affected by the El Niño-Southern Oscillation (ENSO). The highest daily precipitation amounts are around 15-20 mm and take place mainly in January and February. Moreover, a daily cycle in precipitation is observed due to the thermal initiation of convection, which generates an active phase in the afternoon and an inactive phase in the morning. Data from the region show a strong relationship between annual precipitation and elevation. Using this relationship, combined with the local data, the average annual precipitation over the Silala River basin is estimated to be 165 mm. Due to the seasonal variability of precipitation, the vegetation and areal extent of the Orientales and Cajones wetlands have significant seasonal variability (Alcayaga, 2017). During the austral summer (wet season), the surface area of the Cajones and Orientales wetlands is ~0.16 km2, whereas in the winter (dry season) it is ~0.011 km2 (Suárez, Muñoz et al., 2017). Despite significant inter-annual variability, the wetland surfaces did not show any significant long-term change over the last 30 years, considering satellite images from 1987 to 2016 (Alcayaga, 2017). Temperature data are registered at the Inacaliri gauge, located near the lowest point of the basin. Usually, the maximum daily temperature does not exceed 21 °C and the minimum daily temperature frequently reaches values below zero, between 0 °C and -10 °C. The monthly mean temperature 176
Annex VII
7
varies from 1.8 °C (July) to 8.6 °C (March). However, there is a strong gradient of
decreasing temperature with increasing elevation; higher elevation temperatures will
therefore include a higher proportion of sub-zero temperatures. Potential
evapotranspiration (ETo), the maximum water that can be evaporated from the soil and
transpired by the vegetation when water is not limiting, varies between 2 and 8 mm/day
(Suárez, Muñoz et al., 2017). The ETo estimation methods used in this investigation
show that ETo is lower in winter and higher in summer, with a mean standard deviation
of 0.5 mm/day (Suárez, Muñoz et al., 2017) and with annual totals of 1685  142
mm/year (mean  standard deviation). When the evapotranspiration (ET) rate is
compared with the mean annual precipitation of 165 mm, it can be seen that ET losses
from the basin mainly depend on the availability of precipitation. However, where water
is available, for example in the spring-fed wetlands and other riparian areas, ET rates
will be higher.
Actual evapotranspiration (ETr) rates at the Cajones and Orientales wetlands represent
19% and 81.4% of the ETo in winter and summer, respectively. The ETr values are
small compared to the river flow at the Chile-Bolivia international border. Our best
estimate of this loss, using remote sensing data, is that the annual average wetland ETr
is equivalent to an annual average discharge of 1.3 litres per second (l/s) (0.7% of the
river flow at the international border). The maximum monthly value corresponds to a
discharge of 5.9 l/s in February (3.3% of the river flow measured at the international
border). If it is conservatively assumed that the wetlands evaporate at the potential rate
(ETo), the annual ETo represents 2% of the annual streamflow, and the highest water
loss from the wetlands to the atmosphere occurs during January and is equivalent to
approximately 6.5  2.2% (average  standard deviation) of the river flow at the Chile-
Bolivia international border. Therefore, as these percentages are rather small, it is
expected that changes in the magnitude of wetland ET due to channels first constructed
in 1928, would be negligible compared to the river flow.
The flow in the Silala River is nearly constant at a weekly or a monthly time scale. The
monthly average river flow registered at the “Silala River above the FCAB Intake” (the
DGA Fluviometric Station) is ~170-180 l/s (General Directorate of Water’s public
website: http://www.dga.cl), with higher river flows occurring during the wet season.
Nonetheless, the river flow displays daily fluctuations that can be explained by daily
variations of the climatic variables. This behavior is consistent with a groundwaterdominated
river system. Indeed, the BFI of the Silala River is ~0.92, meaning that 92%
of the total of the streamflow at the international border is baseflow, i.e., the streamflow
component that reacts slowly to rainfall and is typically associated with water
discharged from groundwater storage (Eckhardt, 2008).
Annex VII
177
8 The Silala River is a perennial river, and its flows are dominated by groundwater sources. The sources of its perennial flow are found in the groundwater-fed Bolivian springs, but it also receives numerous lateral inputs from groundwater springs along its subsequent path. For instance, Suárez et al. (2017) estimated that ~35.9 l/s of groundwater enter the river from many springs located between the Chile-Bolivia international border and Quebrada Negra. These springs are located in the ravine walls, and are mainly at higher elevations than the river bed. The river is also actively interacting with a shallow underlying fluvial aquifer; Suárez et al. (2017) estimated that ~3.3 l/s of water flows from the river into this fluvial aquifer through the riverbed sediments. A final point to note is that an artesian well exists on Chilean territory that is characterized by having higher water temperatures than those of the river. This artesian well generates a significant additional groundwater contribution into the river, from a deeper groundwater source. 3 STUDY AREA 3.1 Catchment The perennial flows of the Silala River that enter Chilean territory originate in Bolivia, from the Orientales and Cajones wetlands that are ~3.4 km and ~1.3 km east of the Chile-Bolivia international border, respectively (Figure 3-1 and Figure 3-2), at more than 4,323 m.a.s.l. For this analysis, we define the Silala River basin as the area that has its drainage point located at Inacaliri, located ~5 km downstream of the Chile-Bolivia international border (UTM coordinates 596,453 E; 7,563,039 N - datum WGS84-19S). The area of the Silala River basin is 95.5 km2, with 72.2% of this area being in Bolivian territory and 27.8% in Chilean territory (Figure 3-1) and is located between 21°57' S and 22°04' S, and 67°57' W and 68°05' W. The highest elevation in the basin is 5,703 m.a.s.l. (Volcán Apagado), whereas its lowest point is the outlet of the basin at 4,050 m.a.s.l. (Inacaliri). Because of the natural slope, surface waters from the Bolivian portion of the Silala River basin flows naturally to the Silala River, and subsequently cross the international border to reach the Chilean side, as shown on Figure 3-3. The Chile-Bolivia international border corresponds to a line that passes through the highest points at the Cerro Inacaliri o del Cajón (henceforth “Cerro Inacaliri”), Cerrito de Silala, and the Volcán Apagado. A 5-m resolution Digital Elevation Model (DEM), made available for the basin and validated with the Chilean Geographic Military Institute (IGM) cartography and aerial photographs (Alcayaga, 2017), was used for the topographic and hydrological analysis described here. The headwaters of the river are the major water sources of the Silala fluvial system (Alcayaga, 2017) and come from a series of springs located along the edge of the 178
Annex VII
9
Orientales and Cajones wetlands (Figure 3-3). The flow from the Orientales wetland
enters a ravine ~3 km downstream of the water source from the wetland. This ravine
then converges with the Cajones ravine in Bolivia to form a common river channel that
is called the Silala River. This confluence is located ~750 m east of the Chile-Bolivia
international border. The ravine of the Silala River crosses the international border at an
elevation of ~4,277 m.a.s.l. This elevation is based on measurements made by Chile
during a joint field programme carried out by Bolivian and Chilean technicians in
October, 2000. In this field campaign, technical experts from both States measured and
exchanged coordinates and elevations along the ravine on Bolivian and Chilean
territory, using the existing boundary markers (“hitos”), as well as the 1:50,000
common cartography approved by the Chile-Bolivia Mixed Boundary Commission as
reference (Inacaliri 2200-6800, Annex N° 34 and 34A of Minutes N° 38 of the Chile-
Bolivia Mixed Boundary Commission). The 5-m resolution DEM estimate of this
location is 4,282 m.a.s.l. (Alcayaga, 2017). Given the 5-m cell resolution and the 5-m
root mean square error of the DEM, the two estimates of the elevation at which the
Silala River crosses the international border are in good agreement. Once in Chilean
territory, the Silala River receives contributions of a series of ephemeral streams, such
as Quebrada Inacaliri and Quebrada Negra. The Quebrada Negra joins the Silala River
at its left bank, ~2 km downstream of the Chile-Bolivia international border, whereas
the Quebrada Inacaliri joins the Silala River on its right bank at ~2.8 km downstream of
the confluence with the Quebrada Negra. An artesian well (SPW-DQN), drilled for
exploration purposes that is located less than ~50 m downstream of the confluence with
the Quebrada Negra and the Silala River, contributes with an additional flow from a
deeper groundwater system (Figure 3-3). Finally, the Silala River runs for about ~2.2
km from the police station until reaching the San Pedro de Inacaliri River.
Annex VII
179
10 Figure 3-1. Drainage network of the Silala River basin. 180
Annex VII
11
Figure 3-2. Cajones and Orientales wetlands.
Annex VII
181
12 Figure 3-3. Contour lines of the ground in the Silala River basin in a 3D image. 182
Annex VII
13
Figure 3-4 shows the longitudinal profile of the Silala River constructed by Alcayaga
(2017) using a DEM-5m. A nickpoint around 7,000 m upstream of the Inacaliri Police
Station is identified, which divides the profile into two reaches (from Orientales to the
nickpoint and from the nickpoint to the Inacaliri Police Station). Table 3-1 compares the
slope of these two reaches as well as that of the water source Cajones with the values
from another profile built by Chile based on the joint programme of fieldwork by
Chilean and Bolivian technicians in October 2000. The slopes computed from both
sources of information are very similar.
Based on
Slopes
Source
Orientales
Silala downstream of
Orientales
Source Cajones
wetland
DEM-5m 1.7% 5.0% 8.1%
DIFROL profile based
on October 2000 joint
fieldwork with Chilean
and Bolivian technicians
1.5% 4.2% 8.0%
Table 3-1. Average slope comparison for the Silala longitudinal profile using two information
sources.
Annex VII
183
14 Figure 3-4. Longitudinal profile of the Silala River and main tributaries, extracted from the DEM-5m (Alcayaga, 2017). 184
Annex VII
15
Figure 3-5. Photographs of de Silala River basin. a) Silala River. b) River flow gauge station of
Silala River at the Chile-Bolivia international boundary (DGA Fluviometric Station). c) Chile-
Bolivia international border where at the Silala River crossing. d) Riparian wetlands near the
international border. e) View of Orientales wetlands (Bolivian territory) from the Chilean side
of the boundary on the Cerrito de Silala.
Figure3-5. Photographs of the Silala River basin. A) Silala River. B) River flow gauge station at theSilala River. C)Chile-Bolivia international borderwhere the Silala River flows. D) Riparian wetlandsof the Silala River.E) View of Orientaleswetlands (Bolivian territory) from the Chilean side of the boundary on the Cerrito de Silala. A B C D E Annex VII
185
16 Figure 3-5 shows photographs taken on 2016-2017 field expeditions, of the Silala River basin showing different geographic features. In general, the photographs show the Chilean zone of the basin, the Silala River at the border and the vegetation that surrounds it. A photograph of the Orientales ravine (Bolivian territory) taken from the Cerrito de Silala is also presented. 3.2 History of water development concessions in the basin In 1906, Chile granted a concession for the use of the waters of the Silala River on its territory to the British company “The Antofagasta (Chili) and Bolivia Railway Company Ltd.” (FCAB), for an indefinite period of time and for the purpose of increasing the flow of drinking water serving Antofagasta. Two years later, in 1908, FCAB also secured the rights of use of the waters of the Silala from the Bolivian government, again for an indefinite period of time. At that time, no water works existed on the Silala River in either Bolivia or Chile. During the years ~1909-1910, FCAB built the necessary waterworks and pipeline to Antofagasta for the intake and transportation of the water of the Silala River. The impoundments are small weirs that raise water levels to facilitate the pipeline offtakes, while also allowing river flow to be transmitted downstream. The FCAB former Intake in Bolivia was constructed precisely 595 m upstream from the Chile-Bolivia international border from where the pipeline (“Pipeline N° 1”) began (Figure 3-6). Pipeline N°1 was used to transport water suitable for domestic purposes to storage reservoirs at San Pedro Station (~60 km downstream). The Silala–San Pedro Station reservoirs pipelines were 11, 10, 9, 8 inches in size and had a maximum capacity of ~6,600 m3/d (Fox, 1922). The remainder of the water still flowed down the Silala River into Chile. No channels were constructed at that time. Only in 1928, after seventeen years of operation, did FCAB decide to improve the construction of the original intake because it had deteriorated. At that time, FCAB built open channels lined with stones (Figure 3-7 (b) and (c)) near the headwaters of the Silala River, in the Bolivian Orientales and Cajones wetlands, to protect water quality and to avoid contamination by fly’s eggs hatching in the vegetation growth along the river (FCAB Letter N°143 - dated 27 January 1928). The stone channels were approximately 60 cm deep and 60 cm wide. They collect the water from the Orientales and Cajones wetlands and conduct it to the impoundment, following the path of the natural river channel (Figure 3-8). In the main ravine, Orientales, the channel has a length of 3,584 m from its beginning to the Chile-Bolivia international border. The channel in the Cajones ravine is 602 m long and joins the main ravine (Orientales) ~700 m from the international border. 186
Annex VII
17
A second and smaller FCAB Intake was constructed in 1942, located in Chilean
territory ~20 m downstream of the Chile-Bolivia international border, together with a
second pipeline (“Pipeline N° 2”) (Figure 3-6). Pipeline N° 2 is 56 km long and also
goes to the San Pedro Station reservoirs, where the water was stored in two tanks.
Pipeline N° 2 has a maximum capacity of ~3,000 m3/d.
Figure 3-6. FCAB former Intake in Bolivia, FCAB Intake in Chile and pipelines constructed
and used by FCAB. The FCAB former Intake in Bolivia and Pipeline N°1 (orange line)
conducted water from Bolivian Territory to the San Pedro Station reservoirs. FCAB Intake and
Pipeline N°2 (green line) conducted water from Chilean territory, also to the San Pedro Station
reservoirs.
Annex VII
187
18 Figure 3-7. Channels and impoundment constructed in 1928 on Bolivian territory. a) Bolivian impoundment looking down the valley (towards Chile). b) Looking upstream Cajones main channel from Cajones ravine. c) Junction of the channels that come from the Cajones and Orientales wetlands. Photographs (undated) provided by FCAB. A B C 188
Annex VII
19
Figure 3-8. View of channels and FCAB former Intake in Bolivia.
Annex VII
189
20 In 1956 a further withdrawal of river water, downstream of the FCAB sites, was initiated. The Chilean Exploration Company (Chilex, now State-owned mining company CODELCO) began the operation of their current hydraulic system that collects and transports water to the San Pedro Station reservoirs. This water was then used to supply drinking water and domestic uses in the Chuquicamata open-pit copper mine (~12,000 inhabitants). The water withdrawal was initially permitted by provisional legal decrees dated 1954 and 1956. A permanent decree for the use of the waters by Chilex was approved in June, 1958. Chilex was authorized to transfer 87.5 l/s of their water rights (of a total of 165 l/s) from the San Pedro River to the Silala River (referred to in the legal permission as the Inacaliri river), and to collect an additional 31.5 l/s for drinking purposes, thus supporting a total water withdrawal of 119 l/s. The location of the intake was ~4,900 m to the west of the Chile-Bolivia international border (CODELCO intake in Figure 3-6 and Figure 3–9). According to the plans of the 1955 project, the waters collected at the CODELCO intake were transported to the Chuquicamata mine using a 16-inch steel pipe over a course of ~110 km. The current hydraulic system has the same design and civil works, except for the last 27 km of the pipeline, which was relocated in 1986 due to expansion of the Chuquicamata mine. In 1990, CODELCO obtained additional water rights of 41 l/s; increasing the total rights of use up to 160 l/s. Figure 3-9 shows a simplified diagram of the CODELCO intake and the layout of the pipeline that transports water to Chuquicamata. The average amount effectively extracted by CODELCO is ~140 l/s. 190
Annex VII
21
Figure 3-9. CODELCO Intake and pipeline to Chuquicamata.
The use of the water collected at the CODELCO Intake has been extended to include
industrial uses. CODELCO has stated that, since 1995, these waters have been used at
the CODELCO’s Radomiro Tomic Division both for drinking water (~10 l/s) and
industrial purposes (~35 l/s). From 2012, these waters have also been used for drinking
water (~5 l/s) at the CODELCO’s Ministro Hales Division. The water consumption at
the CODELCO’s Chuquicamata Division is of ~65 l/s of drinking water and ~25 l/s of
water used for industrial purposes.
Figure 3-6 shows the FCAB former Intake in Bolivia and the pipelines that were used
until ~1997. In 1997, Bolivia revoked FCAB’s concession of the use of the waters of
the Silala River. Since then, FCAB improved and modified the FCAB Intake in Chile
(Figure 3-10). Since 1997 and until very recently, the channels on Bolivian territory
Annex VII
191
22 have no longer been maintained, without any evidence of impacts on the flow of the Silala River entering Chile. Figure 3-10. FCAB’s current system used to transport water from the Silala River to the San Pedro Station reservoirs. The water is collected from the FCAB Intake and transported using two pipelines (Pipelines N°1 and N°2), both of which are joined into a single pipe (Pipeline N°3) until it meets the pre-existing two pipes (Pipelines N°1 and N°2). Additional weirs were installed by FCAB in the study zone for monitoring purposes in 1993-1994 (weir 4 to weir 10) and in 1999 (weir 11), as shown in Figure 3-11. These weirs have a trapezoidal shape and are used to determine the flow rate in different reaches of the Silala River. 192
Annex VII
23
FCAB takes water from the FCAB Intake that is located between weirs 4 and 5, near the
Chile-Bolivia international border (Figure 3-11). Historical data provided by FCAB
shows that between June 2010 and September 2016, a flow of 125.8 ± 9.4 l/s (average ±
standard deviation) was collected at the FCAB intake (Suárez et al., 2017).
Figure 3-11. FCAB weirs used to measure water flow in the Silala River.
Annex VII
193
24 4 CLIMATE AND METEOROLOGY This section presents a climatic and meteorological characterization of the Silala River basin and its surroundings, including the global and regional influences on the climate. 4.1 Geographical aspects The climate of Chile is highly influenced by the Andes mountains (Figure 4-1), which run along the western edge of South America for more than 5,000 km, from north of the Equator (10°N), to Tierra del Fuego (53°S). Mountain heights in the tropical and sub-tropical sector tend to exceed 4,000 – 5,000 m.a.s.l. and then decrease abruptly at 35°S latitude to elevations ranging between 1,000 – 1,500 m.a.s.l. Nonetheless, some mountains and volcanoes in the Patagonian sector also exceed 3,000 m.a.s.l. In contrast with its great elevation, the Andes is a narrow mountain range which has a width of 200 km on average, except in the central sector known as the Altiplano, where its width is greater and where the study area is located. Because of their size and extent, the Andes substantially affect the atmospheric circulation in the region, resulting in turn in a variety of local and regional phenomena and in a remarkable climatic contrast between the lowlands on each side of the mountain range. The South American Altiplano is a semi-closed plain located in the middle of the Central Andes between 15° and 22°S (Figure 4-1) that has an overall NW-SE orientation. Its elevations vary between 3,600 and 4,200 m.a.s.l. The western limit of the Altiplano is the Cordillera Occidental (Western Mountain Range), a mountain range with volcanoes and peaks exceeding 5,000 m.a.s.l. The terrain then gradually descends towards the desert in southern Peru and northern Chile, at elevations of approximately 1,500 - 2,000 m.a.s.l., followed by a rapid fall towards the Pacific coast located about 200 km from the Altiplano. The eastern limit of the Altiplano is the Cordillera Real (Eastern Mountain Range), a less continuous range that has summits also surpassing 5,000 m.a.s.l. The Boopi, Grande, Chaparé, and Beni rivers originate in the Bolivian Altiplano sector and drain towards the lowlands of east Bolivia (at about 300 m.a.s.l.) and subsequently to the great Amazon basin. Two sectors are identified in the Altiplano. The northern sector corresponds to a narrow and high basin (3,900 m.a.s.l.) that contains the Titicaca Lake. The southern sector is a wider and slightly lower basin (3,600 m.a.s.l.) containing the Uyuni and Coipasa salt flats. During the last several thousands of years, the southern sector has been drier (Gayo et al., 2012), although in the more remote past this sector was likely more humid (Hastenrath and Kutzbach, 1985). 194
Annex VII
25
Figure 4-1. General map of the Andes Mountains showing the ground elevation. The inset
shows a view of the Altiplano region and the location of the Silala River basin (at latitude
22°05’ S and longitude 68°01’ W).
4.2 Large-scale atmospheric circulation
Figure 4-2 describes the large-scale atmospheric circulation pattern that explains the
climate of the Altiplano. At surface level and up to an elevation of 1,500 m, the Central
Andes (15 - 25°S) separate the anticyclonic circulation over the southeast Pacific from a
Annex VII
195
26 continental system east of the Mountain Range. The Pacific anticyclone generates persistent winds from the south along the coast of northern Chile and southern Perú. Furthermore, the air descends over the anticyclone, confining the moisture to a marine boundary with a thickness of ~1,000 m over the ocean (Rahn and Garreaud, 2010). Above 1,500 m, the air is extremely dry (relative humidity < 10%) and stable (Garreaud, 2011). Lastly, the Pacific anticyclone inhibits the entry of frontal systems from the south, which generates the hyper-arid conditions of the coastal desert of northern Chilean and southern Peru (Lettau and Costa, 1978; Garreaud et al., 2010). East of the Central Andes, the low-level atmospheric circulation is dominated by winds from the northwest that tend to blow parallel to the mountain range. These winds have a maximum intensity at about 1,000 m above the surface, forming a jet stream at low levels (Marengo et al., 2004) induced by a thermo-orographic low pressure that prevails in the Chaco area (southeastern Bolivia, Paraguay). This jet stream sustains the moisture movement from the Amazon basin toward the South American sub-tropical plains (Vera et al., 2006). Finally, at elevations of 3 km and above, the atmospheric circulation begins to be dominated by west-east winds (zonal flow), as shown in the altitude-latitude cross section of Figure 4-3. Throughout the year, winds from the Pacific towards the continent prevail in mid latitudes, while winds from the east towards the Pacific take place in low latitudes. However, there is a significant seasonal variation of high relevance for the Altiplano sector. During the southern summer (December to February), the winds from the east reach up to about 25°S, and thus the air flow over the Altiplano mainly comes from the continent towards the Pacific. This air flow brings moisture from the continent towards the Central Andes, which feeds the convective summer storms that give rise to the Altiplano winter or Bolivian winter (Fuenzalida and Rutland, 1987; Hardy et al., 1998; Vuille, 1999; Garreaud, 1999; Vuille et al., 2003; Falvey and Garreaud, 2005). During the rest of the year, the winds from the east recede towards the north and the air column over the Altiplano is dominated by relatively intense winds from the west. These winds transport extremely dry air from the free troposphere over the Pacific that inhibits the development of convection and causes a long dry season from March to November. The following section will discuss in detail the relevance of this seasonal variation of the air flow over the Altiplano and its consequences over the rainfall dynamics in the region. 196
Annex VII
27
Figure 4-2. Atmospheric circulation at low levels (below 1 km) in South America (solid arrows),
main precipitation systems and stratus cloud deck (shaded areas), and the upper-level westerly
jet stream (dashed arrow). Adapted from Garreaud (2011).
Annex VII
197
28 Figure 4-3. Latitude-altitude cross sections of the zonal wind (west-east component), along 80°W for the (a) southern summer months and (b) winter months. W and E indicate maximum winds from the west and east, respectively. Adapted from Garreaud (2009). The southward occurrence of winds from the east during the summer months is caused by an atmospheric feature known as the Alta Boliviana (Bolivian High). This is a high pressure center located in the upper part of the troposphere (above 5 km and more developed near the 10 km level), which is clearly seen in the average air flow lines of the anticyclone circulation shown in Figure 4-4. This phenomenon is the response to the warming triggered by the convective clouds over the Amazon basin, which reaches the highest development and extent precisely during the summer months (Lenters and Cook, 1997). Because of its position, the Alta Boliviana generates winds from the east m.a.s.l.
m.a.s.l.
198
Annex VII
29
on its northern side typically located over the Altiplano, explaining the flow from the
east shown in Figure 4-3.
Figure 4-4. Streamlines of summer (DJF) mean wind at the 200 hPa level (~12 km ASL) in
NCEP–NCAR reanalysis (observations). Also shown in light (dark) blue regions with 200 hPa
wind speed above 25 (40) m/s. Magenta thin lines are the summer mean surface isobars of 1015
and 1020 hPa (innermost) over the SE Pacific signaling the subtropical anticyclone. Adapted
from Garreaud et al. (2010).
4.3 Moisture and local precipitation conditions on the Altiplano
The precipitation on the Altiplano occurs as a result of the moist convection in the
atmospheric column (Garreaud et al., 2003), in which the thermal turbulence raises
large masses of air until they become saturated. Eventually these masses continue rising
due to the warming provided by the condensation process. If that ascent and saturation
is sufficiently vigorous, the clouds will generate precipitation and occasionally
thunderstorms. Both the convection and the eventual precipitation follow a marked daily
cycle, with an active phase during the afternoon and an inactive phase at dawn and
during the morning (Aceituno, 1996; Garreaud and Wallace, 1998).
Although the Altiplano receives large amounts of solar radiation, enough to initiate the
afternoon thermal turbulence during most of the year, the generation of convective
Annex VII
199
30 storms and precipitation is strongly limited by the availability of humidity (Garreaud, 1999). In contrast with the continental lowlands or tropical oceans where high humidity prevails, the air is quite dry most of the year in the Altiplano, especially in the southern sector where the lack of massive water bodies severely limits evapotranspiration. Aceituno (1996) estimated an annual average of the water vapor mixing ratio of w = 2 g/kg (grams of vapor per kilogram of dry air) in the Visviri gauge (17.5°S, 69.5°W) located at 4070 m.a.s.l. This value implies a relative humidity below 30%. From a local standpoint, precipitation over the Altiplano occurs when w > 5 g/Kg, something not very common during the year. This empirical relation between precipitation and humidity using the Visviri gauge data was originally detected by Garreaud (1999); later Garreaud (2001) and Falvey and Garreaud (2005) generalized this relationship using more gauges throughout the Altiplano. 4.4 Precipitation and sources of moisture Figure 4-5 shows the distribution of annual precipitation on the Altiplano. Precipitation reduces with increasing latitude, with values above 600 mm/year near the Titicaca Lake and 100 – 200 mm/yr at the south-western edge of the Altiplano. Moreover, the precipitation decreases drastically outside this region, at the west of the Western Cordillera (Figure 4-6). A second remarkable feature of the precipitation in the region is its strong seasonal variation. Indeed, rainfall can take place in the rainier northern sector throughout the entire year. In contrast, more than 90% of the annual precipitation in the southwestern sector occurs between December - March. In this sector vegetation is very scarce and soil water contents are very low (Norambuena et al., 2011), and thus, evapotranspiration is unlikely to increase air moisture significantly. In fact, a soil moisture flow to the atmosphere of 100 mm/year is estimated for the Altiplano using the Weather Research Forecast (WRF) model (Figure 4-7). The fact that this flow is only 25% of the mean annual precipitation emphasizes the tremendous relevance of the moisture flow from the Amazon basin to explain the precipitations in the Altiplano. 200
Annex VII
31
Figure 4-5. Digital elevation model of the central Andes presenting (a) the spatial distribution
of the annual precipitation and (b) the percentage of precipitation concentrated between
December and March (southern summer months). Adapted from Vuille (1999).
Figure 4-6. West – East cross section showing the mean annual precipitation (diamond
markers) and the topographic profile between 19 and 21°S. Adapted from Houston and Hartley
(2003).
Annex VII
201
32 Figure 4-7. Annual average of the soil moisture flow to the atmosphere (QFX). The prevalence of the moisture flow from the East as a triggering factor for the precipitation over the Altiplano can be seen in Figure 4-8, which shows the average profile of the water vapor mixing ratio w, on both sides of the Central Andes. In the west to east direction, the ratio rapidly decreases at heights above 1,500 m.a.s.l. as a result of the air subsidence over the Pacific anticyclone. Over the Bolivian lowlands a maximum w also occurs in the first 2 km; however w decreases with the elevation is a more controlled manner, as the convective activity in the interior of the continent vertically mixes the water vapor. At the elevation of the Altiplano (i.e. ~4,000 m.a.s.l.), the average w values over the Pacific and the Bolivian lowlands are 1 ± 1 g/Kg and 5 ± 1 g/Kg, respectively. 202
Annex VII
33
Figure 4-8. Average vertical profiles of the water vapor mixing ratio in two air columns 200 km
west from the Central Andes in the Pacific Ocean (EP) and 200 km east over the Bolivian
lowlands (AM). Adapted from Garreaud (1999).
4.5 Inter-annual variability
The natural variation of the climate generates changing conditions through the years
(i.e. inter-annual variability). This variability is illustrated in Figure 4-9, which shows
the spatial average of the annual precipitation from 1962 to 2005 recorded at rain
gauges near the highland basins of the II Region of Chile, where the Silala River basin
is located (see Figure A.1, in Appendix A, for gauge locations). The long-term average
of the mean annual precipitation is 140 mm/year, and the variability implies the
alternation of dry and rainy years, with precipitation values ranging from 40 mm to 340
mm.
Several studies have linked the inter-annual variability of the precipitation over the
Altiplano with the El Niño Southern Oscillation (ENSO) phenomenon (e.g. Vuille,
1999; Vuille et al., 2000; Arnaud et al., 2001; Garreaud and Aceituno, 2001; Vuille and
Keiming, 2004; Ronchail and Gallaire, 2006; Seiler et al., 2013). This linkage
emphasizes the significance of large-scale circulation processes already discussed,
rather than the local evapotranspiration processes, to define the rainfall pattern in the
Altiplano. The ENSO is a form of climate variation due to the alteration of the coupled
Annex VII
203
34 ocean-atmosphere system in the tropical sector of the Pacific Ocean (Garreaud et al., 2009, and Vuille and Garreaud, 2011). Typically, every 3 to 7 years, the abnormal warming of the tropical Pacific for about one year produces a positive ENSO phase known as an El Niño event. Similarly, a negative ENSO phase (La Niña) is produced when the tropical Pacific cools down. During the El Niño years, the warming of the tropical ocean also affects the tropospheric column, thus increasing the thermal contrast between low and mid latitudes. This phenomenon strengthens the winds from the west in subtropical latitudes (20-30°S), which dominate the mid and high troposphere. Under this condition, both the transport of air moisture from the interior of the continent towards the Pacific and the precipitation in the Altiplano decrease (Garreaud and Aceituno, 2001). The opposite takes place in the summers of the La Niña years, when the winds from the west in the subtropical areas are less intense, favoring the flow of moisture from the continent towards the Pacific and increasing the Altiplano precipitation. Although the physical mechanisms relating the ENSO and the precipitation on the Altiplano are well established, the statistical correlation between these two variables is moderate (r2 0.5), as other factors, including the specific charateristics of each ENSO event, also affect the precipitation regime. Figure 4-9. Time series of the mean annual precipitation in the Altiplano sector in Chile. Unfortunately, the low number of consistent long-term climate records over the Altiplano region does not allow an in-depth analysis to detect climate trends in this region. However, during the past decades, no spatially significant precipitation trends have been detected (Seiler et al., 2013). 204
Annex VII
35
Latorre and Frugone (2017) performed regional palaeoclimate reconstructions from
different proxies in the Antofagasta Region. They demonstrated that in the Altiplano
region the hydroclimate has alternated between wet and dry phases. The Altiplano was
affected by a wet phase that existed during the mid-Holocene and by an intense drought
in the 16th and 18th centuries. Then, a large increase in rainfall was documented at the
turn of the 19th century and lasted until the 1850s. These phases correlate well with
groundwater levels, which control the sedimentary processes within aridland drainage
systems such as the Silala River.
5 EVOLUTION OF THE BASIN AND THE RAVINE
5.1 Geological history and evolution of the Silala River basin geology and
geomorphology
Figure 5-1 presents a schematic cross section across the Silala River basin from north to
south and depicts the geological evolution of the valley in which the Silala River basin
is located, from the Upper Miocene (~5.8 Ma BP) (million years before present; ages
are given on Figure 5-1) to the early Holocene (~8.5 ky BP) (thousands of years before
present), and shows the events and processes described below. A detailed analysis and
discussion of the geology can be found in SERNAGEOMIN (2017). In summary, the
Silala River basin geology was formed by a series of volcanic, tectonic and sedimentary
events and processes that have taken place over the last 6 million years.
The radiometric ages of the rocks found in the Silala catchment indicate at least two
major volcanic events, the oldest, dated from about 5.8 Ma, continued until about 2.6
Ma, and was a long period of dominantly acidic volcanism that included the
emplacement of volcanoes, domes, volcanic vents and the extrusion of lavas (Figure
5-1, panel 1). During this period of extensive volcanic activity, a very large eruption in
the east resulted in the deposition of an ignimbrite, in this case named the Cabana
Ignimbrite, which has been dated in the Silala catchment at approx. 4.12 Ma (Figure 5-1
panel 2). This represents part of a voluminous, explosive and extensive volcanic
eruption or series of eruptions that affected this area of the Altiplano
(SERNAGEOMIN, 2017). After this, various volcanoes and volcanic vents were
established through and on top of the Cabana Ignimbrite. The volcanic activity
continued and led to the first development of the Cerro Inacaliri. The products of
eruptions from these volcanoes were mainly lava flows and lava domes. This created the
oldest positive relief in the area (e.g. Cerro Inacaliri and Cerrito de Silala). Subsequently
during the late Pliocene and early Pleistocene (~2.6 Ma - 1.5 Ma) local compressive
tectonic deformation resulted in faulting which exposed and tilted the Cabana
Ignimbrite deposits (Figure 5-1 Panel 3). Dated during this period there is evidence of
Annex VII
205
36 fluvial erosion and deposition, including silt and sand deposits, which are found in the vicinity of the Inacaliri Police Station, and debris and mud flow deposits, found at depth in borehole cores at a location a few metres downstream of the international boundary beneath the Silala River ravine. These deposits can be thought of as the first phase of the Silala River development, which might be called the proto-Silala River (Figure 5-1 Panel 4 Silala 1). Into the palaeo-valley of the proto-Silala (contrained by the hills of Inacaliri, Silala and Volcán Apagado) in the same period there flowed a further less extensive ignimbrite deposit, named the Silala Ignimbrite. This welded ash deposit thins to the west so is interpreted as originating in the east in what is now Bolivia, but is likely to have more or less filled the valley (Figure 5-1 panel 5). After this, further volcanic activity resulted in further volcanic edifice development (Inacaliri, Apagado) and in the deposition of an extensive lava flow on the side of the Inacaliri volcano. This flowed into the headwaters of the proto-Silala River truncating a previously established drainage system (Figure 5-1 Panel 6). There are no outcrops of deposits that can be dated younger than 1.48 Ma until the late Pleistocene (ca. 40 – 12 ky BP). During the period of the last glacial maximum and until the recession of the glaciers there is evidence of glacial moraines at levels of over 4,400 m.a.s.l. on the side of Cerro Inacaliri. Alluvial fan deposits can be found interdigitating with the glacial till and were formed contemporaneously (Figure 5-1 Panel 7). In the early Holocene (~11.5 ky BP) the last evidence of volcanic activity can be found in thin deposits of volcanic ash from a eruption of the San Pedro volcano (30 km east of Inacaliri Police Station). After this period the current geomorphology of the Silala River, its deposits and ravine began to be established (Figure 5-1 Panel 8). Radiocarbon dates indicate that the Silala River has been active since before about 8.5 ky BP. 206
Annex VII
37
Figure 5-1. Schematic geological evolution of the Silala River basin, from the Upper Miocene to
recent times (SERNAGEOMIN, 2017).
Annex VII
207
38 5.2 Geology and geological processes Figures 5-2 shows the geological units of the Silala River basin (SERNAGEOMIN, 2017). Volcanic Sequences from the Lower Pleistocene (Pliv(a)), Volcanic Sequences from the Upper Miocene-Pliocene (MsPvd) and Alluvial deposits from the Upper Pleistocene (PlHa) are the most common units, covering 42%, 20%, and 14% of the total area, respectively. Each geological unit is described in more detail in SERNAGEOMIN (2017). The geological units from oldest to youngest are: Cabana Ignimbrite, Silala Ignimbrite, Fluvial deposits and Alluvial deposits. 208
Annex VII
39
Figures 5-2 and 5-3. Geological units of the Silala River basin (SERNAGEOMIN, 2017).
5.3 Fluvial geomorphology
In this section we present the results obtained from field tests performed by the studies
of Arcadis (2017), Latorre and Frugone (2017) and SERNAGEOMIN (2017) to define
the development of the Silala River terraces and the different sedimentary depositions of
the river. The terraces were identified by these studies and were mapped in detail using
the 5-m resolution DEM and a high resolution mosaic. The sedimentary deposits within
the Silala River ravine were investigated using detailed mapping and 14C dating of
plants and organics in the sediments (Latorre and Frugone, 2017).
Annex VII
209
40 Four terraces, named T1, T2, T3 and T4, can be observed at different elevations (Figure 5-4). From youngest to oldest, they are: T1 (~1-2 m above river level), T2 (~5-6 m above river level), T3 (~10 m above river level), and T4 (~20 m above river level). Their development is variable in different sections of the ravine (Arcadis, 2017). Figure 5-5 clearly shows the abrasion terraces in the walls of the Silala ravine. Four depositional units (U1, U2, U3 and U4) have been associated with the terrace development. These terraces have been dated to the early Holocene (more than about 8,500 years BP) (for more detail see the work of Latorre and Frugone (2017)). Figure 5-5 shows a summary of the development of the terraces and sediment fill of the Silala River ravine and Table 5-1 shows the sedimentological properties, age and depositional environment of each depositional units. Terrace T1 formed in the 20th century but a more recent (2000 AD onwards) incision of Unit 4 (>1.5 m in some areas) is currently visible throughout the Silala River where standing vegetation has dried out (Latorre and Frugone, 2017). T1 is found along the entire length of the ravine. Terrace T2 is well developed in the area where the Quebrada Negra converges with the Silala River. The level of terrace T2 can be followed for several hundred meters along the river bed. This terrace does not show much lateral development, perhaps because of the hardness of the Ignimbrite rock. Terraces T3 and T4 are more laterally developed than terrace T2, having erosion surfaces of several tens of meters wides. Terrace T3 forms at the coordinates 600.050E, 7.565.185N, a 15-20 m flood plain along a 100 m axis. In this area various signs of erosion, including excavation overhangs caused by a water course, are visible. 210
Annex VII
41
Figure 5-4. Abrasion terraces on top of Cabana Ignimbrite due to activity of the Silala River
(Arcadis, 2017).
The depositional units formed during three aggradational (deposition of sediment)
phases dated to > ca. 8.5 - 1.9 ky BP (Unit 1), > ca. 0.65 - 0.2 ky BP (Unit 2) and < ca.
0.2 ky to recent (Unit 4). These phases of aggradation are for the most part, coeval with
elevated groundwater tables that have been well-documented and dated throughout the
Atacama Desert (Latorre and Frugone, 2017).
The oldest and thickest section (U1) indicates that sediments accumulated to
approximately 4 m above the current river level before being incised to their current
level. After this incision, accumulation resumed with the deposition of ~1.8 m of
sediment, which corresponds to Unit 4 that is currently visible throughout the Silala
River where standing vegetation has dried out, most likely due to a recent fall in the
groundwater table. These Units represent widespread Holocene wetland deposition of
organic peat, black mats and silty-sand channels.
Unit 1 (ca. 8,500 to 1,900 yrs BP) is a dark brown peat (high organic matter content)
intermixed with massive, medium to coarse sand and abundant plant macrofossils, such
as leaves, terrestrial roots and wetland vegetation. It presumably postdates terrace T2.
The most probably explanation for such deposits is that they were formed by an
elevated water table, which would have promoted the fine sediment infill of the Silala
channel with organic peat, black mats and silty-sand channels.
Annex VII
211
42 Unit 2 starts with the deposition of median to coarse light brown sand intercalated by coarse brown to light grey sand with sub-angular gravels. The unit lies unconformably on top of Unit 1. These sediments are characteristic of in-stream wetland deposits (see Pigati et al., 2014) that likely formed during a phase of elevated (+3 m above the current river level) groundwater table. 14C dating show ages of 530 years ago for these wetland deposits. The very top of Unit 2 is eroded and overlain by Unit 3. Unit 3 consists of a 120-cm thick massive light grey, fine to medium matrix-supported and poorly sorted gravels. It corresponds to the abrupt deposition of a mudflow (possibly originating laterally from Quebrada Negra) sometime after 530 years before present, and lies unconformably on top of Unit 2. Unit 4 consists of fine sub-rounded gravels interbedded with loamy sand. The unit has abundant plant remains of the waiya grass Calamagrostic Eminens and clumps of dried grasses can often be seen on the surface of this unit. Unit 4 sits unconformably atop Unit 1. This implies that an episode of massive incision (several meters) occurred in the Silala after the deposition of Unit 3. The incision is capped by a final phase of aggradation that possibly began in the early 1800s. High energy fluvial environments were followed by a final phase of finer fluvial aggradation forming the historical riverbed (T1 terrace). 14C dates on plant remains from this unit show that it formed until very recently (as recent as 2000 AD). The dried-out surface of Unit 4 is equivalent to the T1 and lies ~2 m above the current river level (Latorre and Frugone, 2017). The geomorphology of the Silala River and its ravine has been evolving over more than about 8,400 years as a result of changes in the hydrology promoted by changes in the prevailing climatic regime. This has resulted in the four terraces and depositional units found along and in the base of the ravine. 212
Annex VII
43
Figure 5-5. Summary schematic diagram the Holocene sedimentary history of the Silala River.
Unit U4 is inset into U1 indicating a major phase of incision after deposition of U3, after 500
years ago. All ages are in calibrated years before 1950 AD (cal yrs BP) (Latorre and Frugone,
2017).
Annex VII
213
44 Unit Sedimentological properties Age (calibrated 14C yrs BP) Depositional environment Unit 4 Sandy loam with large angular clasts. Forms the recent T1 terrace <~200 (?) cal yr BP to recent Modern fluvial environment surrounded by waiya grasses Unit 3 Massive matrix-supported gravels, heavily altered by pedogenesis <~530 cal yr BP Mudflow Unit 2 Massive coarse to medium sands, interbedded gravels, uppermost 20 cm with interbedded laminated silts and black mats <~1900-530 cal yr BP Extensive river floods that evolved into a short-lived wetland Unit 1 Dark brown clayey peat, interbedded sand lenses, abundant plant remains >~8500-1900 cal yr BP Bofedal, a high altitude wetland Table 5-1. Summary of Stratigraphic Units and ages described for in-stream sedimentary infill of the Silala River (Latorre and Frugone, 2017). 5.4 Fluvial processes of the Silala River Sediment transport experiments in the Silala River have shown that the river is still geomorphologically active (Mao, 2017). Observations from two experimental reaches (Figure 5-6) have shown that finer sediments are moved in higher percentages and for longer distances than coarser sediment fractions. Also, sediments are transported in more quantity and for longer displacement lengths during higher discharges. A natural discharge experiment was carried out, with abstraction by FCAB temporarily stopped. As expected, the bedload transport rate was higher during this period. There is strong evidence of size-selective transport dynamics typical of natural rivers, which is expected under limited sediment supply conditions, i.e., the case of the Silala River. The evidence of size-selective transport justifies the presence of a static armour layer and reinforces the indication that fluvial processes are ongoing in the Silala River. The stream features a cascade-step/pool morphology, which is expected because of the local range of slope and grain size. This reinforces the consideration that the Silala is an 214
Annex VII
45
alluvial river with fully-developed fluvial processes, which shape the channel
morphology through active coarse sediment dynamics. The Morphological Quality
Index classifies the river as featuring overall good morphological conditions, thus
exhibiting forms and processes that are close to the full potential of the river, given the
existing boundary conditions (Mao, 2017).
Figure 5-6. Location of the study reaches of the Mao (2017) study.
The riverbed of the Silala River has a static armour layer (Mao, 2017), meaning that the
stream flows are strong enough to move mainly the finer sediments under limited
sediment supply conditions. The sediment supply is relatively small due to the lack of
major sediment sources along the main channel and the dense mat of Calamagrostis
Eminens that grows on the banks, protecting them from being eroded (Mao, 2017). The
Annex VII
215
46 armour ratio measured by Mao (2017) is 2.5 for one reach and 2 for the other reach (Figure 5-6 shows the location of the reaches). These values are similar to values found in rivers with perennial regimes located in humid environments (Figure 5-7). Although the Silala River is located in an arid environment, its regime is perennial because it is a groundwater-dominated river, with a very high base flow index (more details are provided in the next section). Indeed, its flow discharge is very stable and only slightly fluctuates during the day. Although more significant discharge pulses are likely to occur due to rainfall events, floods are not as flashy as one would expect in an ephemeral environment. Figure 5-7. Armour ratio of rivers with different regimes. Values obtained for reaches A and B of the Silala River are plotted in red (Mao, 2017). 6 HYDROLOGICAL CHARACTERISTICS This chapter presents an analysis of the most important hydrological and hydrometeorological processes occurring in the Silala River basin. Precipitation, temperature, evapotranspiration, infiltration, and streamflow in the basin are analyzed at different time scales to understand the hydrological behavior of the basin. 6.1 Hydrological processes in the basin The flow in the Silala River at the Chile-Bolivia international border is determined by the balance of precipitation inputs to the basin and losses of water to the atmosphere by evapotranspiration. Precipitation in the study region is caused by convective winds that 112244661101001000110100D50(mm)D50S(mm)Armour ratioAridHumidSnow meltSilala47bring 216
Annex VII
47
bring moisture from the East (Amazonas) and can fall in liquid or solid form, depending
on the temperature. Precipitation mainly occurs during the austral summer. A daily
cycle in precipitation due to the thermal initiation of convection is also observed, with
an active phase in the afternoon and an inactive phase in the morning. The precipitation
that falls over the basin can moisten the ground surface and either evaporates, infiltrates
into the soil or loose rock surface materials, or travels as surface runoff. Infiltrated water
can evaporate from the soil’s unsaturated zone as direct evaporation or plant
transpiration, or it can percolate and recharge the aquifer. These processes depend on
the meteorological conditions and the type of soil. In the Silala River basin, soil and
rock surfaces are generally permeable, so that infiltration is a dominant process,
providing recharge to groundwater systems, which emerge as springs to feed the river
discharge. Surface runoff is relatively limited, which can be observed from the patterns
of streamflow, discussed below.
Evapotranspiration is the combination of water evaporated by the soil surface and
transpired by the plants; therefore, it depends on the meteorological conditions, the
presence and activity of vegetation, and water availability. The vegetation in the
Altiplano region is located mainly where surface water occurs, either in groundwaterfed
wetlands, or near-stream riparian areas. The highest evaporation rates are observed
during the austral summer. On the same summer day, there may be large values of both
evaporation and precipitation, due to the diurnal cycles of precipitation and the
atmospheric conditions.
In the Altiplano area, the atmospheric temperature presents a high thermal variability
during the day. During the night, the temperature may often have values below 0 °C.
Also, in the Silala River basin the solar radiation is very high (especially in summer)
because the basin is located at a high altitude.
As a result of the extreme environmental conditions of the Silala River basin,
quantification of the water budget and of hydrological processes is challenging. Due to
the daily cycle of most of the hydrological variables, data must be collected on a daily,
hourly or even subhourly basis. Unfortunately, in isolated areas such as the Silala River
basin, there is generally little instrumentation and those instruments that are available
require more maintenance due to the extreme weather conditions.
6.2 Instrumentation in the basin to measure hydrometeorological and hydrological
processes
Table 6-1 and Figure 6-1 show the existing instrumentation in the Silala River basin,
which belongs to different entities and have data collected over different time periods.
The Chilean General Directorate of Water (DGA, for its acronym in Spanish), has
meteorological and fluviometric stations installed in the study basin. The data of the
Annex VII
217
48 DGA stations are the national official data. In general the data recorded by DGA stations have a daily time scale, with the exception of the fluviometric station, which in the middle of 2016 began to measure the streamflow at 15 min intervals. Some of these stations have several years of records and others were recently installed. As can be seen from Table 6-1, the Inacaliri and Silala meteorological stations have more years of record than the other stations. Different instruments were also installed in the area of study by the UC during 2016 (UC meteorological station). These instruments were installed to validate the DGA data, extend the spatial coverage and to incorporate measurements of other variables such as wind speed and net radiation. These data are typically collected at time intervals of 1 h. In addition, the water level at the DGA Fluviometric Station is also being measured by the UC science team using 15-min intervals (Suárez, Muñoz et al., 2017). Mao (2017) performed temporary measurements of streamflow and temperature of the river in Reach A of Figure 5-6. These data were measured between September 2016 and January 2017. 218
Annex VII
49
Station Name Source Latitude S Longitude W
Elevation
(m.a.s.l.)
Start date
Measured
variable
Silala DGA -22.01 -68.03 4,300 jun-77 Precipitation
Inacaliri DGA -22.03 -68.07 4,040 jan-1969
Precipitation,
temperature,
humidity, and
evaporation
Quebrada Negra DGA -22.03 -68.04 4,240 nov-16
Precipitation,
temperature,
humidity, and
evaporation
Mirador Silala DGA -22.05 -67.99 4,900 nov-16
Precipitation,
temperature,
humidity, and
evaporation
UC meteorological
station
UC -22.03 -68.04 4,191 sep-16
Precipitation, Wind
Speed, Relative
Humidity, Air
Temperature,
Atmospheric
Pressure and Net
Radiation.
DGA Fluviometric
Station (“Río Siloli
antes de Bocatoma
FCAB”)
DGA -22.01 -68.03 4,300 jan-2001 Streamflow
Table 6-1. List of gauges in the Silala River basin.
Annex VII
219
50 Figure 6-1. Location of the gauges in the Silala River basin. 6.3 Precipitation Monthly precipitation recorded in gauges located in and near the study area were used to characterize the precipitation regime (Figure A.1). These gauges belong to either the Chilean General Directorate of Water (DGA, Spanish acronym) or the Bolivian National Service of Meteorology and Hydrology (SENAMHI, Spanish acronym), and have at least 10 hydrologic years (from October to September) of record. Table A.1 220
Annex VII
51
presents these stations. Missing data in the records were infilled using equation 1, which
considers the correlation with two nearby stations.
𝑌𝑌𝑖𝑖
∗ = 𝛼𝛼𝑖𝑖𝑃𝑃1𝑖𝑖 + 𝛽𝛽𝑖𝑖𝑃𝑃2𝑖𝑖 + √1 − 𝑅𝑅2 · 𝜀𝜀𝑖𝑖 · 𝑆𝑆𝑦𝑦
Equation 1, where 𝑌𝑌 𝑖𝑖
∗corresponds to precipitation to be estimated at month i, using the known
precipitation for the same month in two nearby highly correlated gauges, 𝑃𝑃1𝑖𝑖 and 𝑃𝑃2𝑖𝑖. 𝛼𝛼 and 𝛽𝛽
are weighting parameters estimated using the least squares error method. Lastly, the term
√1 − 𝑅𝑅2 · 𝜀𝜀𝑖𝑖 · 𝑆𝑆𝑦𝑦 is added to preserve the natural variance of the completed series. In this term
𝑅𝑅2 is the coefficient of determination of the multiple regressions, 𝜀𝜀𝑖𝑖 is a normal random
standard variable and 𝑆𝑆𝑦𝑦 is the standard deviation of the original series.
Table A.2 compares the averages and standard deviations of the original precipitation
series and the infilled precipitation series, while Table A.3 shows the availability of
information for each hydrological year at each station.
The mean annual precipitation at the two gauges inside the basin, i.e. at the Inacaliri and
the Silala meteorological stations, is 119 mm and 99 mm, respectively (obtained from
the infilled data). It is observed that 5 years of data was infilled for the Inacaliri gauge
and 19 years of data for the Silala gauge, to complete a total of 45 and 37 hydrological
years, respectively.
6.3.1 Spatial variation of precipitation
The mean annual precipitation gradient was estimated using the regional rain gauges
presented in Table A.1. Figure 6-2 shows that annual precipitation measured in
raingauges located below 2,250 m is very minor and no gradient is clearly identified.
Above 2,250 m precipitation is more significant, and a gradient of 57 mm/km is
observed. Overall, precipitation increases in the southwest - northeast direction, as a
function of elevation.
Annex VII
221
52 Figure 6-2. Mean annual precipitation gradients for the stations located in the Second Region, Chile. Figure 6-3 shows the map of the isohyets of mean annual precipitation for the Silala River basin. The values in this particular basin range from just less than 100 mm at the lower elevations to 250 mm in the highest areas, with 165 mm/yr being the spatial average. The tracing of the isohyets was made using a spatial interpolation of the annual average precipitation for the hydrological year, after infilling the missing data. Initially, isohyets were drawn using the Surfer V7.0 program (Golden Software, 1999), which were then corrected based on topography and the gradient of precipitation with elevation. 02040608010012014016018005001,0001,5002,0002,5003,0003,5004,0004,500Mean annual precipitation (
mm) Elevation (m) 222
Annex VII
53
Figure 6-3. Isohyets of annual precipitation in the Silala River basin.
Annex VII
223
54 6.3.2 Temporal variation of precipitation Figure 6-4 shows the annual precipitation (infilled data) and the associated inter-annual variability at the Silala rain gauge located inside the Silala River basin, from 1978 to 2014. The long-term average of the mean annual precipitation is 99 mm/year (based on 37 years of data), and the standard deviation of the annual precipitation is 74 mm. This variability is associated with the alternation of dry and rainy years, with precipitation values ranging from 11 mm (year 2010) to 305 mm (year 1997). The years with no value of annual precipitation are because data of monthly precipitation were missing and could not be infilled. Figure 6-4. Annual precipitation of Silala gauge. Figure 6-5 and Table 6–2 show the mean monthly precipitation at the Silala rain gauge using the data between 1978 and 2016 (the missing data was infilled. See Table A.3 for availability of information). It can be seen that more than 92% of the precipitation occurs between December and March. The maximum monthly precipitation exceeded 120 mm in January and February at Silala Station. In the same months, the standard deviation is large (42 mm on average). 224
Annex VII
55
Figure 6-5. Mean monthly precipitation of the Silala gauge (1978-2016 – infilled data) (mm).
Error bars represent minimum and maximum monthly precipitation.
GAUGE Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Annual
MEAN 0.2 0.5 6.1 32.1 36.0 17.0 1.3 0.9 1.1 0.7 1.5 1.3 99
MAX 5.4 14.9 47.1 121.2 217.1 83.2 14.3 13.8 13.5 9.8 16.3 14.4
MIN 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
STAN.DEV 0.9 2.5 10.3 36.0 47.6 21.5 3.3 2.6 2.9 2.3 4.0 3.3
Table 6-2. Average mean, maximum, minimum and standard deviation of the monthly
precipitation measured at the Silala gauge (1978-2016 – infilled data) (mm).
Figure 6-6 shows the daily precipitation at the Silala rain gauge from January 2001 to
October 2016. These data were not infilled. The maximum daily precipitation of the
entire series is 50 mm (January, 2012) but in general the maximum daily precipitation
for each year reaches ~15-20 mm. The average daily precipitation is ~5 mm, without
considering the days with null precipitation.
Annex VII
225
56 Figure 6-6. Daily precipitation at the Silala gauge (mm) (2001-2016). 6.4 Temperature Monthly temperature data from DGA (Chile) and SENAMHI (The Bolivian National Service of Meteorology and Hydrology) gauges located in and near the study area were used to calculate the spatial variation of temperature with altitude (see Figure B.1 in the appendices). Five or more hydrologic years of records (from October to September) were considered for this analysis. The information of the locations of these gauges and the available information for each hydrological year at each station are presented in the appendices (Table B.1 and Table B.3). The Inacaliri gauge was chosen for the analysis of the temporal variation of the minimum, mean and maximum temperatures (at both daily and monthly time scales). This station was chosen because it is the only station within the Silala River basin with historical temperature records (from 1969 to 1992). Temperature data from the UC meteorological station (Table 6-1) were also used to determine the hourly distribution of temperature. The mean annual temperature at the Inacaliri gauge is 5.3 °C. This annual temperature is compared to those values from the other temperature gauges in Table B.2. There are six gauges with an elevation greater than 4,000 m.a.s.l. The mean annual temperature of these stations ranges from 0.06 °C (Laguna Colorada gauge at 4,278 m.a.s.l.) and 10.41 °C (El Tatio gauge at 4,370 m.a.s.l.). 6.4.1 Spatial variation of temperature with altitude Figure 6-7 shows the annual mean temperature gradient using the regional data of the gauges selected for this study, which are all above 1,500 metres of elevation. At an 226
Annex VII
57
altitude ranging between 1,500 and 4,500 m.a.s.l., a reduction of 4.6 ºC/km was
estimated.
Figure 6-7. Annual mean temperature gradient in the study region.
6.4.2 Temporal variation of temperature
Figure 6-8 shows the last 12 years of daily maximum and minimum temperature
measured at the Inacaliri gauge (January 1980 to October 1992). These data were not
infilled and it is useful to have a notion of the maximum and minimum temperatures
registered at that station, because this variable affects other variables that are essential to
understand the hydrological behavior of the basin, e.g., evapotranspiration, infiltration,
and the physical state of precipitated water (water or snow). The maximum daily
temperature of the time series is 27.5 °C but in general the maximum daily temperature
does not exceed 21°C. The minimum temperature recorded at Inacaliri often reaches
values below zero. The minimum daily temperature of the time series is -14.7 °C but
rarely falls below -10 °C. Table 6-3 shows the statistics of the monthly temperature
(mean, maximum, minimum and standard deviation). Table 6-4 shows the average
standard deviation of the minimum, maximum and mean daily temperature for each
month in the same period of time. The maximum daily temperature has a greater
standard deviation in August (2.6 °C) and the minimum daily temperature has a greater
standard deviation in December (2.6 °C).
Annex VII
227
58 Figure 6-8. Daily temperature of the Inacaliri gauge (°C) (1980-1992). GAUGE Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep MEAN 5.3 6.6 7.9 8.3 8.4 8.6 6.4 3.9 2.1 1.8 2.7 3.7 MAX 10.3 9.1 9.7 9.8 10.5 12.2 8.6 7.7 5.8 4.8 5.6 6.3 MIN 2.5 4.8 6.2 5.6 6.3 6.3 4.2 1.7 -0.7 -0.8 1.0 1.4 STAN.DEV 1.8 1.2 0.9 1.0 1.1 1.5 1.1 1.5 1.7 1.3 1.1 1.3 Table 6-3 Monthly mean temperature at the Inacaliri gauge (°C) (1969-1992). Monthly standard deviation of: Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Average Mean Temperature 1.8 1.9 2.2 1.6 1.5 2.0 2.5 1.9 1.5 1.9 2.0 1.7 1.9 Maximum Temperature 1.9 2.0 2.5 1.9 1.9 2.4 2.3 2.3 2.0 2.2 2.6 2.1 2.2 Minimum Temperature 2.1 2.3 2.6 2.2 1.8 2.2 2.4 2.0 1.9 2.2 1.9 2.0 2.1 Table 6-4. Monthly standard deviation of the maximum, minimum and mean daily temperature (°C) at the Inacaliri gauge. 228
Annex VII
59
Figure 6-9 shows the maximum, minimum and mean monthly temperature. The
monthly mean temperature varies from 1.8 °C (July) to 8.6 °C (March) with a mean
standard deviation of 1.3 °C.
Figure 6-9. Mean monthly temperature at the Inacaliri gauge (°C) (1969-1992).
6.4.3 Hourly temperature
The mean, maximum and minimum hourly temperatures for November and December
2016 and January 2017, as well as the minimum and maximum values of the record
were computed from the data collected by the gauge (UC meteorological station)
installed for this study (Figure 6-10). During the day the temperatures followed the
same trend, as they begin to rise at 8:00 a.m. and decrease at approximately 7:00 p.m.
Temperatures were lower in November, whereas the mean hourly temperatures in
January were higher than those in December between 9:00 p.m. and 8:00 a.m.
Nonetheless, the mean average temperatures in January were lower, because this month
has more days of precipitation, i.e., it is a month with colder and more humid
conditions. Furthermore, in every month, minimum negative temperatures (up to -10°C
in November) occurred between 2:00 a.m. and 8:00 a.m. At this temperature water
freezes and during many nights, from 10:00 p.m. until 8:00 a.m., the river waters may
partially freeze.
Annex VII
229
60 Figure 6-10. Monthly hourly temperatures as well overall maximum and minimums temperatures at the gauge installed in this study. 6.5 Evapotranspiration Various data were analyzed to estimate the evapotranspiration (ET) losses from the basin. We note that potential evapotranspiration (ETo) is a widely used concept and represents the rate at which evaporation can occur from an idealized vegetated surface, actively growing, and not short of water. ETo is representative of the evaporating power of the atmosphere at a specific location and time and only depends on local climatic variables. It does not take into account the effects of different crops and their stage of development, nor of soil characteristics and management practices, nor of water limitation. On the other hand, actual evapotranspiration (ETr) is the amount of water that is actually evaporated from the soil and transpired by the plants, and it is usually smaller (and often very much smaller) than ETo. It depends on the specific vegetation type, its stage of growth and whether it is actively transpiring, and most importantly, includes the effects of water limitation in restricting plant transpiration (Allen et al., 1998; Campbell and Norman, 2012). In the Silala basin, ETo is large and the available precipitation is small. Hence the ET losses from most of the basin are small, limited by the available precipitation. However, where water is readily available, for example at locations of spring emergence, wetland vegetation develops, and since water is not limiting, for these areas ETo can be considered as an upper bound estimate of ETr. Potential evapotranspiration can be estimated using standardized pans designed to measure evaporation (Epan). Evaporation pans provide a very simple measurement of the integrated effects of the relevant atmospheric variables, e.g., radiation, wind, 230
Annex VII
61
temperature and humidity, on the evaporation from an open water surface. The
evaporation rate from pans filled with water is easily obtained, but requires qualified
personnel and rigorous maintenance of the station. In the absence of rain, the amount of
water evaporated during a period (mm/day) corresponds to the decrease in water depth
in that period.
The US Class A pan is probably the most widely used internationally, and has proved
its practical value to estimate reference or potential evapotranspiration by observing the
evaporation loss from a water surface. However, the Class A pan yields evaporation
estimates that are consistently greater than open water evaporation from shallow lakes
(Eagleman, 1967), and is also different from evaporation from a vegetated surface. Thus
to obtain ETo estimates, the pan evaporation has to be corrected using a pan coefficient
(Kp), i.e., ETo = Kp · Epan (Doorenbos and Priutt, 1977; Allen et al., 1998). This method
may give acceptable results for ETo, depending on the position of the pan and the pan
environment.
Another method to estimate ETo is the FAO Penman-Monteith method, widely
considered to offer the best results with minimum error in relation to a living grass
reference crop (Allen et al., 1998). This method explicitly represents the effects of the
most important atmospheric variables, and is maintained as the standard method for the
computation of ETo where the relevant meteorological data are available. The Penman-
Monteith approach calculates ETo (mm /day) using the following equation (Allen et al.,
1998):
ETo =
0.408Δ(Rn − G) + γ 900
T + 273 u2(es − ea)
Δ + γ(1 + 0.34u2)
Equation 2, where Rn is the net radiation at the crop surface [MJ m-2 day-1]; G is the soil heat
flux density [MJ m-2 day-1]; T is the mean daily air temperature measured at 2 m height [℃]; u2
is the wind speed at 2 m height [m s-1 ]; es and ea are the saturation and actual vapor pressure,
respectively, and the term (es – ea) is called the vapor pressure deficit [kPa]; Δ is the slope of
the saturated vapor pressure- temperature curve [kPa ℃-1] and γ is the psychometric constant
[kPa ℃-1].
To address situations where detailed meteorological data are not available (which is
commonly the case in remote areas), many simplified methods have been developed,
using basic measurements such as temperature. However, these methods are empirical,
generally have reduced accuracy, and should be corrected or recalibrated for the study
region (Xu and Singh, 2000). In this study, the Turc, Priestley-Taylor, Taylor de Bruin,
Annex VII
231
62 and Jensen-Haise methods were adopted to assess the uncertainty associated with the evaporation estimates (Suárez, Muñoz et al., 2017). Actual evapotranspiration is the amount of water that is actually evaporated from the soil and transpired by the plants, and as noted above, it is usually smaller (and often much smaller) than ETo. Various methods exist to estimate ETr from ETo (e.g. Doorenbos and Pruitt, 1977). Recent work on wetland evaporation in arid areas has led to new methods, based on remote sensing of vegetation characteristics (Groeneveld et al., 2007; Senay et al., 2011). This report shows the results of the Suárez, Muñoz et al. (2017) study that used the approach of Groeneveld et al. (2007) to determine ETr, which related actual evapotranspiration (ETr) to ETo using Equation 3: ETr=(ETo−Pp)∗ NDVI+Pp Equation 3 where NDVI (-) is the normalized difference vegetation index and Pp is the monthly precipitation (mm). In the Silala River basin there is one meteorological station that has historical records of pan evaporation (1969-1991): the Inacaliri gauge located at the downstream end of the basin. Two stations from DGA were recently installed with pan evaporation measurements (Mirador Silala and Quebrada Negra gauges) (Table 6-1) and an evaporation pan was recently reinstalled at the Inacaliri station (it had been discontinued due to lack of manpower for daily readings). The pan evaporation methods are susceptible to the microclimatic conditions under which the pans are operating and the results depend on the rigor of station maintenance, hence qualified personnel are needed to perform the measurements daily and who know the procedures to follow. We note that the isolated location and climatic conditions of extreme wind, rain, temperature and radiation of the Siala River basin make these measurements very uncertain. 6.5.1 Pan evaporation regime A characterization of annual pan evaporation in the study area was carried out using monthly pan evaporation data from DGA and SENAMHI (see Figure C.1 for gauge locations). Only stations that had five or more years of records were used in this analysis. Table C.1 presents the information for these pans and shows the mean annual pan evaporation values for the stations selected and their respective altitudes. There are four gauges with an elevation greater than 4,000 m.a.s.l. The mean annual pan evaporation of these stations ranges from 1,583 mm/year (El Tatio gauge at 4,370 m.a.s.l.) to 2,376 mm/year (Inacaliri station - the only station inside the Silala River 232
Annex VII
63
basin with pan evaporation records- at 4,040 m.a.s.l.); these rates can be compared with
the available average basin precipitation of 165 mm/year, as noted in 6.3 above.
Figure 6-11 shows the historical daily evaporation at the pan evaporation of Inacaliri
station. Daily evaporation has a variation within the year, being higher in the summer
months and lower in the winter months. The maximum daily evaporation does not
exceed 16.7 mm.
Figure 6-11. Daily Pan evaporation at the Inacaliri station (1969-1992).
Table 6–5 shows the monthly pan evaporation (mean, minimum, maximum) at the
Inacaliri gauge from January 1969 to October 1992. Figure 6-12 shows the mean
monthly pan evaporation. The monthly mean pan evaporation varies from 109.8
mm/day (June) to 281.6 mm/d (December) with a mean standard deviation of 37
mm/day.
Annex VII
233
64 GAUGE Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep MEAN 255 277 282 235 193 211 189 157 110 119 144 181 MAX 339 318 356 312 244 349 240 224 164 175 212 247 MIN 171 179 174 157 101 160 140 112 52 53 65 97 STANDARD DEVIATION 34 36 46 49 38 41 27 32 33 37 40 37 Table 6-5. Monthly mean pan evaporation at the Inacaliri gauge (mm/month). Figure 6-12. Mean monthly pan evaporation at the Inacaliri gauge (1969-1992). Figure 6-13 shows the preliminary results of daily evaporation calculated based on pan evaporation and daily precipitation measurements at the Inacaliri meteorological station between January and February 2017. Before January 14, pan evaporation was not measured. The average and maximum daily evaporation rates were 6.8 mm and 15 mm respectively, whereas the highest precipitation event was of 12.3 mm (January 19th). Because of the difficulties related to evaporation measurements described above, these data are uncertain. Nonetheless, it is clear that evaporation also occurs during the same days when rainstorms happen. As rainstorms usually do not last all day, there is time for evaporation to occur from the water that fall into the canopy, from the water retained in the ground surface, or from the water that infiltrated in the first soil layers. Indeed, a fraction of the water infiltrated in previous rainstorms may be available to support evaporation in the basin for a couple of days after a rainfall event, wheres another fraction of water may percolate and recharge the aquifer. 234
Annex VII
65
Figure 6-13. Daily Pan evaporation and precipitation at the Inacaliri station, 2017.
6.5.2 Potential ET calculated from meteorological measurements in the study area
Both AGRIMED (2015) and Suárez, Muñoz et al. (2017) used the Penman-Monteith
equation to calculate ETo at various stations and towns near the study basin, on an
annual, monthly, and daily basis (Figure C.2).
Figure C.3 presents the spatial distribution of mean annual ETo in the Region of
Antofagasta, Chile (AGRIMED, 2015). ETo on the Chilean side of the Silala River
basin (at a latitude of 22°S on the international border with Bolivia) varies between
1,000 and 1,350 mm/year. Table 6-6 presents monthly ETo values estimated by
AGRIMED (2015) for the locations in areas near or with similar characteristics to the
study zone (located in the altiplano and above 2,450 m.a.s.l. elevation). ETo varies from
3 mm/day in winter to 6.3 mm/day in summer.
Annex VII
235
66 Town Putre San Pedro de Atacama Socompa Geographic location 18.2°S 69.6°W 22.9°S 68.2°W 24.5°S 68.3°W Month ETo, for the 1950-2010 period (mm/day) Jan 4.0 6.3 4.1 Feb 3.9 6.0 4.0 Mar 3.7 5.5 3.7 Apr 3.5 5.1 3.4 May 3.3 4.6 3.2 Jun 3.2 4.3 3.0 Jul 3.4 4.5 3.1 Aug 3.3 4.5 3.1 Sep 3.5 5.1 3.4 Oct 3.7 5.6 3.6 Nov 3.9 6.0 3.9 Dec 4.0 6.3 4.1 Annual ETo (mm/year) 1,317.0 1,942.0 1,294.0 Table 6-6. ETo at selected locations (AGRIMED, 2015). Figure 6-14 and Table 6-7 compare monthly ETo values calculated at the Chiu-Chiu station (22°20’S, 68°38’W) with different methods (Penman–Monteith, Turc, Priestley-Taylor, Taylor de Bruin, and Jensen-Haise). The Chiu-Chiu station is the DGA’s meteorological station with the best quantity and quality of data near the study site (70 km southwest of the Silala meteorological station). For this reason, the DGA recommends using the temperature and relative humidity data from the Chiu-Chiu meteorological station to estimate ETo at the Silala River basin. Since the DGA did not measure wind speed at the Silala River basin, to estimate ETo for the historical calculations using the Penman-Monteith method, Suárez, Muñoz et al. (2017) used the data provided by the Wind Energy Explorer web platform (http://walker.dgf.uchile.cl/Explorador/Eolico2/), which publicly delivers wind speed data throughout the Chilean territory at different timescales (hourly and monthly) and for different heights above the ground (it has 12 vertical levels from 0 to 200 m). The wind speed information was constructed with numerical simulations of the Weather Research and Forecasting (WRF) model version 3.2, with a horizontal resolution of 1 236
Annex VII
67
km and vertical resolution of 10 m. The resulting ETo values varied from 2 mm/day in
winter to 8 mm/day in summer. An average ETo monthly variability of 0.5 mm/day was
identified from the different methods compared in this analysis. Table 6-7 also presents
the annual ETo of each method, which are greater than 2,000 mm/year with the Penman-
Monteith and Taylor de Bruin approaches.
Figure 6-15 presents the daily ETo measured at the UC meteorological station installed
in the Silala River basin for this study, for the period between November 2016 and
January 2017, and the preliminary data of daily precipitation at the Inacaliri gauge (data
from Dec. 29th, 2016). ETo varies between 2.1 and 7.6 mm/day. The monthly averages
values of ETo are 5.6 mm/day, 5.6 mm/day and 4.2 mm/day for Nov. 2016, Dec. 2016
and Jan. 2017 respectively. It is important to note that both monthly and daily ETo
ranges reported in AGRIMED (2015) and Suárez, Muñoz et al. (2017) are similar to the
recent results from the UC meteorological station.
The annual totals in Table 6-7 ranged from 1304 to 2164 mm, and these rates of ETo can
be compared with the mean annual precipitation of 165 mm/year. Hence for most of the
basin, ETr will be somewhat less than 165 mm/year, with evaporation losses occurring
from wetted surfaces after rainfall, and from infiltrated water retained in the near
surface unsaturated zone and lost as soil evaporation and transpiration from sparsely
vegetated areas. In contrast, wetland areas, where water is available year-round to
support evaporation, can support much higher evaporation losses, as discussed below.
Figure 6-14. Monthly mean ETo computed using different methods (Suárez, Muñoz et al., 2017)
at Chiu-Chiu station.
Annex VII
237
68 ETo (mm/day) Average Month Penman- Monteith Turc Priestley- Taylor Taylor de Bruin Jensen- Haise Jan 6.7 ± 0.8 4.7 ± 0.5 5.6 ± 0.4 7.4 ± 0.5 5.1 ± 0.5 5.9 ± 0.5 Feb 6.2 ± 0.9 4.1 ± 0.6 5.2 ± 0.4 6.8 ± 0.5 4.7 ± 0.5 5.4 ± 0.6 Mar 5.9 ± 0.6 3.3 ± 1.4 4.4 ± 0.4 5.8 ± 0.5 4.2 ± 0.5 4.7 ± 0.7 Apr 5.6 ± 0.5 3.4 ± 0.7 3.5 ± 0.4 4.8 ± 0.3 3.5 ± 0.6 4.2 ± 0.5 May 5.2 ± 0.6 2.9 ± 0.7 2.7 ± 0.3 3.7 ± 0.3 2.8 ± 0.4 3.5 ± 0.5 Jun 5 ± 0.5 2.6 ± 0.6 2.2 ± 0.1 3.1 ± 0.1 2.5 ± 0.2 3.1 ± 0.3 Jul 5.2 ± 0.6 2.5 ± 0.7 2.3 ± 0.2 3.3 ± 0.2 2.4 ± 0.3 3.1 ± 0.4 Aug 5.7 ± 0.5 3.2 ± 0.7 3.1 ± 0.2 4.2 ± 0.2 3.1 ± 0.4 3.9 ± 0.4 Sep 6.4 ± 0.9 3.9 ± 1 4.1 ± 0.2 5.7 ± 0.2 3.9 ± 0.5 4.8 ± 0.6 Oct 6.7 ± 0.9 4 ± 1.9 5.1 ± 0.3 6.9 ± 0.3 4.7 ± 0.4 5.5 ± 0.8 Nov 7.1 ± 0.6 5.1 ± 0.7 5.6 ± 0.4 7.6 ± 0.4 5.2 ± 0.5 6.1 ± 0.5 Dec 7.3 ± 0.5 5.2 ± 0.5 5.9 ± 0.4 7.7 ± 0.4 5.5 ± 0.4 6.3 ± 0.4 Average 6.1 ± 0.6 3.7 ± 0.8 4.1 ± 0.3 5.6 ± 0.3 4 ± 0.4 4.7 ± 0.5 Annual (mm/year) 2164 ± 175 1304 ± 210 1506 ± 93 2033 ± 95 1419 ± 138 1685 ± 142 Table 6-7. Monthly average values and variability of ETo for different methods (Suárez, Muñoz et al., 2017). 238
Annex VII
69
Figure 6-15. Daily ETo values (yellow) estimated using the Penman-Monteith method (Suárez,
Muñoz et al., 2017) and preliminary data of daily precipitation (blue) at the Inacaliri gauge
(DGA).
6.5.3 Actual evapotranspiration from the Cajones and Orientales wetlands
Actual evapotranspiration (ETr) was determined as a fraction of ETo, which was
estimated on a monthly basis by Suárez, Muñoz et al. (2017), using the Penman-
Monteith (Equation 2) and the simplified methods at the Chiu-Chiu station.
Subsequently, the results of this station were extrapolated to the Silala River basin. To
compare the ETr losses from the wetlands with the streamflow of the Silala River
measured at the international border, it was necessary to consider the total area of the
wetlands, including the riparian areas surrounding the river (this area was multiplied by
the ETr values).
The monthly NDVI index determines the surface extent of the vegetation (NDVI values
greater than 0.1) and was obtained by Alcayaga (2017) through the processing of
Landsat images. The area of the wetlands around the Silala River was obtained from the
normalized difference vegetation index (NDVI). Table 6–8 shows the monthly average
of the NDVI for selected years, where it can be seen that the months with the largest
wetland area are January, February and March, with an area of ~0.16 km2. During the
winter months the wetlands have an extent of ~0.011 km2. The standard deviation of the
total area of the wetlands for each month is ~0.06 km2.
0
2
4
6
8
10
12
0.0 14
2.0
4.0
6.0
8.0
10.0
12.0
14.0
11/03/16
11/06/16
11/09/16
11/12/16
11/15/16
11/18/16
11/21/16
11/24/16
11/27/16
11/30/16
12/03/16
12/06/16
12/09/16
12/12/16
12/15/16
12/18/16
12/21/16
12/24/16
12/27/16
12/30/16
01/02/17
01/05/17
01/08/17
01/11/17
01/14/17
01/17/17
01/20/17
01/23/17
01/26/17
01/29/17
Precipitation [mm/day]
ETo [mm/day]
Annex VII
239
70 Surface (km2) per NDVI Values Range Month ≤ 0.1 0.1 - 0.2 0.2 - 0.3 0.3 - 0.4 0.4 - 0.5 0.5 - 0.6 > 0.6 Total January 0.781 0.102 0.042 0.014 0.006 0.001 0.000 0.165 February 0.799 0.068 0.051 0.022 0.006 0.003 0.000 0.147 March 0.793 0.090 0.041 0.016 0.005 0.001 0.000 0.153 April 0.891 0.039 0.013 0.004 0.000 0.000 0.000 0.055 May 0.920 0.025 0.002 0.000 0.000 0.000 0.000 0.027 June 0.932 0.013 0.001 0.000 0.000 0.000 0.000 0.014 July 0.936 0.010 0.000 0.000 0.000 0.000 0.000 0.010 August 0.938 0.008 0.000 0.000 0.000 0.000 0.000 0.008 September 0.932 0.014 0.000 0.000 0.000 0.000 0.000 0.014 October 0.923 0.023 0.001 0.000 0.000 0.000 0.000 0.023 November 0.932 0.011 0.004 0.000 0.000 0.000 0.000 0.015 December 0.909 0.027 0.011 0.001 0.000 0.000 0.000 0.038 Average 0.890 0.036 0.014 0.005 0.001 0.000 0.000 0.056 Maximum 0.938 0.102 0.051 0.022 0.006 0.003 0.000 0.165 Minimum 0.781 0.008 0.000 0.000 0.000 0.000 0.000 0.008 Standar Dev. 0.062 0.033 0.019 0.008 0.003 0.001 0.000 0.062 Table 6-8. Average of the monthly wetland area (km2) as estimated using the NDVI (Alcayaga, 2017). Table 6-9 shows the monthly mean ETr rates calculated at the Chiu-Chiu station by Suárez, Muñoz et al. (2017) with Equation 3. ETr ranges between 0.4 (winter) and 4.4 mm/day (summer) and represents at a minimum 19% of ETo (winter) and at most 81.4% of ETo (February). The ET losses as a daily rate can be converted to a volumetric loss by multiplying them by the wetland area. This water, if not lost to the atmosphere, could appear as streamflow, and hence we consider the relative effect of ET on flow discharge in the Silala River at the international border. Table 6-9 shows the monthly average of ETr obtained combining the Groeneveld et al. (2007) approach (Equation 3) and the ETo methods used by Suárez, Muñoz et al. (2017) (Penman–Monteith, Turc, Priestley-Taylor, Taylor de Bruin, and Jensen-Haise). Table 6-9 also presents the ET monthly average of all the methods, and the wetland ET expressed as the percentage of the monthly streamflow at the international border (records of the DGA Fluviometric Station stream flow gauge). ETr values are higher in the summer months reaching 5.9 l/s 240
Annex VII
71
on February. This value represents 3.3% of the monthly flow. The average standard
deviation of ETr for all methods in January (month with more NDVI data) is 0.74 l/s.
Note that ETr values are small compared with the streamflow at the international border.
The ETr flow represents 2.1  0.4 % (3.7  0.74 l/s) (average  standard deviation) of
the streamflow in January. As an annual average, annual ETr represents 0.7% (1.3 l/s) of
the annual flow.
While the ETr values represent best estimates of the evaporation loss from the wetlands,
we also present ETo results, as an upper bound estimate of the wetland evaporation
losses. Table 6-10 shows the monthly average of ETo obtained for each method. The
ETo reach values of 11.5 l/s. These values represent 6.5% of the monthly river flow. The
average standard deviation of ETo for all methods in January is 3.9 l/s. Under this more
conservative scenario in which evaporation equals the ETo rate, wetland ET losses
corresponds to a 6.5  2.2 % (11.5  3.9 l/s) (average  standard deviation) of the
streamflow in January, and 2% (3.4 l/s) of the streamflow over the year.
This result clearly shows that wetland evaporation is a relatively small component of the
basin water balance. Any historical effects of channelization on the wetland area can
therefore be expected to represent a very minor influence on the streamflow at the
border.
Annex VII
241
72 Month METHOD Average Percentage of stream discharge Penman-Monteith Turc Priestley-Taylor Taylor de Bruin Jensen-Haise ETr (l/s) Jan 4.1 3.1 3.6 4.4 3.3 3.7 2.1 Feb 5.8 5.3 6.0 6.8 5.7 5.9 3.3 Mar 3.1 2.2 2.3 2.9 2.3 2.6 1.5 Apr 0.8 0.6 0.6 0.8 0.6 0.7 0.4 May 0.3 0.2 0.2 0.2 0.2 0.2 0.1 Jun 0.2 0.1 0.1 0.1 0.1 0.1 0.1 Jul 0.1 0.1 0.0 0.1 0.1 0.1 0.0 Aug 0.1 0.1 0.1 0.1 0.1 0.1 0.0 Sep 0.2 0.1 0.1 0.2 0.1 0.2 0.1 Oct 0.3 0.2 0.3 0.4 0.3 0.3 0.2 Nov 0.5 0.4 0.4 0.5 0.3 0.4 0.2 Dec 1.0 0.7 0.7 0.9 0.7 0.8 0.5 Annual 1.4 1.1 1.2 1.4 1.1 1.2 0.7 Table 6-9. Average monthly ETr, expressed as an equivalent streamflow (l/s) using the method of Groeneveld et al. (2007), obtained with the five different methods and the percentage of the monthly and yearly streamflow of the mean of all methods. 242
Annex VII
73
Month
METHOD
Average
Percentage of
stream
discharge
Penman-
Monteith Turc Priestley-
Taylor
Taylor
de Bruin
Jensen-
Haise
ETo (l/s)
Jan 13.2 9.0 10.9 14.4 9.9 11.5 6.5
Feb 8.1 6.3 8.8 11.6 7.6 8.5 4.7
Mar 11.0 7.3 7.9 10.3 7.8 8.9 5.0
Apr 3.7 2.5 2.4 3.3 2.5 2.9 1.7
May 1.6 0.9 0.8 1.1 0.9 1.1 0.6
Jun 0.8 0.4 0.4 0.5 0.4 0.5 0.3
Jul 0.6 0.3 0.3 0.4 0.3 0.4 0.2
Aug 0.6 0.3 0.3 0.4 0.3 0.4 0.2
Sep 1.1 0.6 0.7 0.9 0.6 0.8 0.5
Oct 1.8 1.2 1.4 2.0 1.3 1.6 0.9
Nov 2.5 1.8 1.7 2.4 1.5 2.0 1.1
Dec 3.4 2.2 2.4 3.2 2.2 2.7 1.6
Annual 4.0 2.7 3.1 4.2 2.9 3.4 2.0
Table 6-10. Average monthly ETo, expressed as an equivalent streamflow (l/s), obtained with the
five different methods and the percentage of the monthly and yearly streamflow of the mean of
all methods.
Annex VII
243
74 Figure 6-16. Monthly mean actual evapotranspiration (ETr) using the Groeneveld et al. (2007) method (Suárez, Muñoz et al., 2017). 6.6 Infiltration As discussed in 6.1 above, infiltration is an important process in the basin. Where precipitation exceeds the infiltration capacity of the surface, overland flow will be generated. This overland flow is associated with a rapid runoff contribution to the streamflow. If rainfall is infiltrated, it can be stored in the unsaturated zone and subsequently evaporated, or, when rainfall rates are substantial, it can percolate through the subsurface to recharge the shallow or deep groundwater. Visual inspection of the basin indicates that soils are coarse textured and rock slopes fragmented, which suggests large infiltration rates. However, to provide a quantification of infiltration capacity, a set of infiltration experiments was undertaken, for each of the geological units named below. These experiments were performed by Arcadis (2017) in which the Guelph permeameter and the double-ring method were used to determine the infiltration capacity of each of the following hydrogeological unit (Figure 6-17). HU1 (Fluvial deposits): these correspond to sandy gravels to gravel rich sandstones, with some clay content. These deposits can be classified as an unconfined aquifer, i.e., with a free water surface (water table) (Arcadis, 2017), and have infiltration capacities of ~1x10-1 m/d. 244
Annex VII
75
HU2 (Alluvial deposits): these are poorly consolidated deposits composed of gravel,
sand and silt. The infiltration capacities in these deposits ranged between ~5x10-1 and 2
m/d.
HU3 (Ignimbrite): these comprise two geological units: Cabana Ignimbrite (beneath)
and Silala Ignimbrite (above). Thus, HU3 could be divided into a lower unit composed
of crystal rich tuff and an upper unit composed of welded lithic tuff. HU3 crops out in
the walls of the Silala River ravine directly surrounding the Silala River valley (Figure
6-17). Arcadis (2017) determined that HU3 unit can be a confined or semi-confined
aquifer with an infiltration rate of ~4x10-1 m/d for the 15 cm thick incipient soil that
covers the Cabana and Silala Ignimbrites.
HU4 (Pyroclastic fall deposits): HU4 corresponds to pyroclastic fall deposits as well
as stratified, non-consolidated deposits (SERNAGEOMIN, 2017). These deposits have
the lowest infiltration capacity that is of the order of ~0.02 m/d.
HU5 (Andesitic and dacitic volcanic rocks): these correspond to the volcanic
sequences from the lower Pleistocene (Pliv(a)) including an andesitic lava flow dated
ca. 1.48 Ma. The infiltration capacity of these deposits was not measured.
HU6 (Weakly permeable rock): these comprise all the other rocks that outcrop in the
proximity of the Silala Valley (Figure 6-17), mainly Miocene-Piocene Volcanic
Sequences. The surface infiltrations tests (Arcadis, 2017) indicate an infiltration rate
ranging from 1x10-2 to 1 m/d for the 15 cm thick incipient soil. Underneath this soil,
very low permeability rock was found.
It can be seen that several of the geological units have infiltration capacities larger than
1 m/d. These rates can be compared with the highest daily rainfall rates of around 15–20
mm/day, i.e., six years of average precipitation could be infiltrated in a single day when
the infiltration capacity is of ~1 m/d. We can therefore conclude that infiltration is likely
to be a dominant process, especially in HU2.
Annex VII
245
76 Figure 6-17. Hydrogeological units defined by Arcadis (2017). 6.7 River flow A DGA Fluviometric Station (also known as “Río Silala antes de Bocatoma FCAB”) is located in the Silala River at the Chile-Bolivia international border. The gauge is located at an elevation of 4,300 m.a.s.l. at coordinates 600,440 E and 7,565,684 N. This station transmits the water stage by satellite, and the river flow is calculated through a stage-discharge curve, which is calibrated manually every month. These data are publically 246
Annex VII
77
available from the DGA website. The fluvial station has a record of 9 years of complete
data (from 2001 to 2007 and from 2010 to 2011). For the following analysis, the gauge
data were used without filling the data gaps to avoid alteration of the field information.
Thus, only complete hydrological years with no missing data were used in the analysis.
Note that there are difficulties in flow measurement due to the extreme conditions of the
Silala River. An important feature is the low night time temperatures and thus, the
measuring devices can be frozen and can crack. Therefore, more frequent maintenance
of the station is necessary, but this is not easy to do since the basin is located in an
isolated area with difficult access. In addition, partial freezing of the flow cross-section
may affect the stage-discharge relationship.
6.7.1 Annual and Monthly flow discharge
Figure 6-18 shows that the average annual streamflow discharge measured by the
fluviometric gauge is 0.17 m3/s, with a maximum of 0.20 m3/s (year 2004) and a
minimum of 0.15 m3/s (year 2011). This discharge represents 32.7% of the annual
precipitation (165 mm over a surface area of 95.5 km2), i.e., the runoff coefficient of the
basin is 0.327.
Figure 6-18. Annual streamflow at the DGA Fluviometric Station flow gauge, from 2001 to
2015.
Figure 6-19 shows the monthly streamflow measured by the DGA from 2001 to 2016.
The streamflow ranges between 0.12 and 0.23 m3/s with a flow of 0.17  0.02 m3/s
Annex VII
247
78 (average  standard deviation). The monthly mean and standard deviation of the flow of this period is shown in Table 6-11. In comparison with most perennial rivers, the streamflow is relatively constant, although higher discharges are observed from January to May when precipitation occurs. The average standard deviation of the monthly discharges is 23 l/s (i.e., a 13.4% of the monthly average discharge), with maximum and minimum values of 27.9 and 18.7 l/s occurring in September and December, respectively. Figure 6-19. Monthly streamflow at the DGA Fluviometric Station flow gauge, from 2001 to 2016. Oct Nov Dec Jan Feb Mar Apr May Jun Jul Agu Sep Mean Mean 0.17 0.17 0.17 0.18 0.18 0.18 0.18 0.18 0.17 0.17 0.17 0.17 0.17 Stan. Dev 0.02 0.02 0.02 0.02 0.02 0.03 0.03 0.03 0.02 0.02 0.02 0.03 0.02 Table 6-11. Mean monthly flow discharges and standard deviation at the DGA Fluviometric Station (m3/s). 6.7.2 Daily flow discharge Figure 6-20 shows the water discharge monitored downstream of the FCAB abstraction by Mao (2017) in reach A (Figure 5-6) for four months, showing the diurnal variability of the flow. Results show that at reach A there is always at least 11 l/s. In the flow rate time series, anomalous oscillations and peaks (especially in November) are likely to be 248
Annex VII
79
due to artificial hydraulic operations upstream (cleansing of water tank and
experimental aquifer pumping test operations). The very high streamflow measured on
the 11th January 2017 corresponds to the experiment of Mao (2017), with the natural
flow discharge restored by FCAB, where the liquid discharge at reach A reached 120
l/s. These records have daily fluctuations most likely due to daily variations of the
meteorological variables, freezing effects, the turbulent nature of the flow, and due to
the daily variations of hydrological processes. For instance, Lundquist and Cayan
(2002) studied the diurnal cycle of river flows in many rivers in the western United
States and relate it to the variations of other variables such as temperature, infiltration
and evaporation. Therefore, we infer that similar processes may be occurring at the
Silala River. Nonetheless, we do not have enough data to infer the main mechanisms
that explain the daily cycles in river flow.
Figure 6-20. High-resolution (15 min) discharge monitoring at reach A (September 2016 to
January 2017 (Mao, 2017).
6.7.3 Base Flow Index
Baseflow is a streamflow component which reacts slowly to rainfall and is usually
associated with water discharged from groundwater storage (Eckhardt, 2008). The Base
Flow Index (BFI) is the ratio of the base flow to the total flow calculated from a
hydrographic smoothing and separation procedure using daily discharges. The BFI is
thus a measure of the river’s runoff, which is derived from stored sources and, as a
general catchment descriptor, has found many areas of application, including low flow
estimation and groundwater recharge assessment (Brušková, 2008).
Annex VII
249
80 We computed the BFI using the algorithm proposed by the Institute of Hydrology (1980) that uses equations 4 and 5: 𝐵𝐵𝐵𝐵𝐵𝐵= Σ𝑏𝑏𝑖𝑖𝑖𝑖Σ𝑑𝑑𝑖𝑖𝑖𝑖 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑜𝑜𝑜𝑜 𝑎𝑎𝑎𝑎𝑎𝑎𝑀𝑀𝑀𝑀𝑀𝑀𝑎𝑎𝑎 𝑎𝑎 𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵= Σ𝐵𝐵𝐵𝐵𝐵𝐵𝑎𝑎𝑎𝑎𝑛 Equations 4 and 5,where 𝑏𝑏𝑖𝑖 are the base flow values that are part of the selected season, 𝑑𝑑𝑖𝑖 are the total streamflow values that are part of the selected season, 𝐵𝐵𝐵𝐵𝐵𝐵𝑎𝑎 is the annual or seasonal BFI where 𝑎 denotes the year or season and 𝑛 is the total number or years/seasons. The daily flows measured at the DGA Fluviometric Station gauge were used to determine the BFI from 2006 to 2015. The BFI was 0.924 and the mean of annual BFI’s was 0.920. Figure 6-21 shows the base flow calculated for the streamflow at this gauge for the year 2007 using the algorithm proposed by the Institute of Hydrology (1980). The year 2007 was selected because it has a complete data set. It can be seen that the base flow is similar to the river flow. Therefore, the Silala River is dominated by groundwater discharge (Brušková, 2008). Figure 6-21. Streamflow and base flow of the Silala River at the DGA Fluviometric Station gauge (year 2007). 250
Annex VII
81
6.8 Groundwater-surface water interactions
Suárez et al. (2017) studied groundwater-surface water interactions in a 2-km reach of
the Silala River downstream of the Chile-Bolivia international border. This
investigation used temperature rods, distributed temperature sensing methods, and was
supplemented with river flow data measured at the FCAB weirs that are located along
the study reach (Figure 6-22). The results show that the Silala River reach is a gaining
reach (i.e., dominated by groundwater inputs), where ~35.9 l/s of groundwater enters
the river from various springs that discharge water from the walls of the ravine at
elevations higher than the river stage. These springs have a higher temperature than the
river water. Suárez et al. (2017) estimated that ~3.3 l/s of water is lost from the river to
a shallow fluvial aquifer below the river, by percolation through the riverbed sediments.
Near the end of the reach, groundwater is discharged into the river from an artesian well
(SPW-DQN), which discharges ~91.6 l/s of warm water (at a higher temperature than
the springs noted above). Therefore, the water balance shows a net gain in the
investigated reach of the order of ~32.5 l/s from natural springs and bed losses, and
~124 l/s when the effect of the artesian well is included. The data collected in this study
reveal that river-aquifer interactions in the Silala River are undeniable and support the
hypothesis that the Silala River is a perennial river sustained by groundwater.
Arcadis (2017) constructed a hydrogeological diagram that summarizes the
understanding of the hydrogeology of the Silala River basin (Figure 6-23 and Figure
6-24). They found that the precipitation in the surroundings of the Silala River area
infiltrates into the surface rocks and deposits (mainly HU1 and HU5) and flows
underground to shallow-unconfined aquifers, discharging the water through springs in
the vertical rock walls of the river valley (Figure 6-23). This water interacts with the
alluvial/colluvial sediments adjacent to the rock walls, travelling over and through the
fluvial deposits (HU1), until it reaches the river. Precipitation at higher elevation in the
basin may infiltrate into different volcanic rocks and recharge the ignimbrite aquifer
HU3 (Figure 6-23).
The Silala River headwaters are in Bolivia, where high altitude wetlands are fed by
springs. At the Orientales wetland the springs appear to emerge either from the Silala
Ignimbrite or the edges of the overlying Pleistocene andesitic lava, which was deposited
on top of the ignimbrite (Figure 6-23) (Hauser, 2004; SERNAGEOMIN, 2017). The
flow from these springs could be due to infiltration into the lava flow, the Silala or
Cabana Ignimbrite, or alluvial/colluvial deposits, which are found on the hillsides
leading down to the wetland. The springs feeding the Cajones wetland could be sourced
from the alluvial deposits on the side of Cerro Inacaliri or possibly the ignimbrite
bedrock. The discharge from the springs flows as surface water in the river channel,
following the natural topographic gradient from high to low altitude zones (from Bolivia
Annex VII
251
82 to Chile) and enters the Silala River ravine. The Silala River enters the Chilean territory with a discharge of ~170 l/s. Shortly after this ~126 l/s are taken by FCAB. Between this point and the confluence with Quebrada Negra, the Silala River receives a water influx of ~35.9 l/s, provided by valley wall springs. A few meters downstream the confluence of the Silala River and the Quebrada Negra, an artesian well (SPW-DQN) feeds ~ 91.6 l/s of groundwater into the river. The river continues until the Codelco intake, which is very close to the Inacaliri Police Station where most of the surface water is taken. It is likely that deep groundwater flows through the HU3 ignimbrite rock from NE to SW, in more or less the same direction as the Silala River as indicated by the measured water levels from deep boreholes (Arcadis, 2017). Groundwater in the fluvial deposits, HU1, beneath the river is fed by direct recharge and river leakage. The sediments are unconfined as a whole but may have variable perched water tables on the lower permeability deposits (peat, black organic mats, and silt) but the direction of flow must follow the base of the fluvial deposits. Each of these systems should be considered as a separate aquifer. 252
Annex VII
83
Figure 6-22. Location of the flow discharge gauges (weirs 4, 5, 6, 7, 8 and 9, and Flow gauge at
Reach A), the temperature rods (TR1-TR5) and the fiber-optic cable (Suárez et al., 2017).
Annex VII
253
84 Figure 6-23. Longitudinal schematic diagram along the course of the Silala River from its headwaters in Bolivia (NE to the right) to Codelco intake in Chile (SW to the left). This diagram, together with Figure 6-24 provides a summary of the understanding of the hydrogeology of the Silala River basin (Arcadis, 2017). Figure 6-24. Transverse schematic diagram along the course of the Silala River. Approximate location of the diagram is shown by a green bold vertical line labelled “cross section” in previous figure. This diagram, together with Figure 6-23 provides a summary of the understanding of the hydrogeology of the Silala River basin (Arcadis, 2017). 254
Annex VII
85
7 CONCLUSIONS
The main objective of this study is to present an up-to-date physiographic, hydrological,
and hydrogeological characterization of the Silala River basin. This study presents an
up-to-date characterization of the study area, and describes the climatic, geological,
geomorphological and hydrological processes present in the basin.
The Silala River basin is located on the western edge of the Andean Altiplano, and
drains from Bolivian territory to the Antofagasta Region of Chile. Elevations in the
basin headwaters generally exceed 4,000 m.a.s.l. The Silala River becomes one of the
main tributaries of the San Pedro River, which flows into the Loa River. The Loa River
is the longest Chilean river (440 km long) and the main watercourse in the Atacama
Desert, draining from the Andes Mountains to the Pacific Ocean.
The waters of Silala River originate from a series of springs located in the Orientales
and Cajones wetlands in Bolivia, at more than 4,323 m.a.s.l. The waters from the
Orientales wetland flow into a ravine which converges with the Cajones ravine to form
a common ravine and associated stream called the Silala River ~750 m east of the
Chile-Bolivia international border. The ravine of the Silala River crosses the
international border at an elevation of ~4,277 m.a.s.l. The Silala River basin has a
drainage area of 95.5 km2, of which 72.2% is located in Bolivia and the remainder is
located in Chile (Alcayaga, 2017). This basin drains into a location known as Inacaliri,
located ~5 km downstream of the Chile-Bolivia international border (596,453 E;
7,563,039 N datum WGS84-19S).
In addition to the Bolivian springs that form the source of the perennial flows, a series
of natural springs in Chilean territory also discharge their waters into the Silala River
(Suárez et al., 2017). In addition, a well previously drilled on the Chilean territory for
exploration purposes also discharges groundwater to the river, under artesian flow
conditions (SPW-DQN well). The waters of the Silala River flow naturally towards
Chile from the Bolivian highlands following the natural topographic gradient of the
basin; the channel slope is relatively steep (5%), calculated from a nickpoint (7,000 m
upstream of the Inacaliri Police Station) to the Inacaliri Police Station along the river
(Alcayaga, 2017).
The Silala River basin formed as the result of a series of geological events, including
volcanic processes, which took place episodically over the past 12 Million years before
present (Ma BP) (SERNAGEOMIN, 2017). The Silala River ravine was carved out by
the action of the river. The first evidence of the existence of alluvial drainage tied to the
Silala fluvial system is found beneath the Silala Ignimbrite (dated approx. 4.12 – 2.6 Ma
BP) but the drainage system was truncated by a lava flow from Inacaliri volcano (ca.
Annex VII
255
86 1.48 Ma BP); the second phase in the evolution of the Silala River system began in the Late Upper Pleistocene – Lower Holocene (ca. 12–8.5 ky BP). While glaciation has affected landscape processes at higher elevations, there is strong evidence that fluvial action was the dominant process in the formation of the Silala River ravine (SERNAGEOMIN, 2017). The principal geological units of Silala River basin, in terms of their hydrological relevance, are: HU1 (Fluvial deposits), HU2 (Alluvial deposits), HU3 (Ignimbrite), HU4 (Pyroclastic fall deposits), HU5 (Andesitic and dacitic volcanic rocks) and HU6 (weakly permeable rock) (Arcadis, 2017). The units with greatest presence in the basin are HU5, HU6, and HU2. HU6 and HU5 are composed mainly of volcanic rocks and HU2 is composed of unconsolidated deposits of rounded stones, gravels, sands, and silts. The Silala River ravine is between 10 and 100 m wide, with a mode of 20 m. It has four erosion terraces (T1, T2, T3 and T4) with associated depositional periods that have different degrees of development in the study area (Arcadis, 2017; Latorre and Frugone, 2017) indicating past river erosion levels. On one hand, the T2, T3 and T4 terraces were developed over a period of about 8,500 years (Latorre and Frugone, 2017). On the other hand, the youngest terrace is T1, which was formed in the 20th century. This terrace is found along the entire length of the ravine. Sedimentation studies concluded that the Silala River is an alluvial river with fully developed fluvial processes, which shape the channel morphology through active sediment dynamics (Mao, 2017). Although the Silala River is located in an arid environment, its regime is perennial because it is a groundwater-dominated river, with a very high base flow index (BFI, which is defined as the ratio of the base flow to the total flow). Its bed sediment composition (armour ratio) is comparable with values typically found in rivers with perennial regimes in humid environments (Mao, 2017). The hydrological behavior of the Silala River basin is affected by climatic phenomena, the surface topography of the basin, and its surface and subsurface characteristics. The basin is located in an area that has a climate that has a strong seasonal cycle and high inter-annual variability, but also changes on an hourly time scale with extreme conditions of wind, temperature, radiation and precipitation. Therefore, understanding the dominant processes that affect the water balance of the Silala River basin is complex. For this reason, the various hydrological processes must be quantified on an hourly or a daily basis. Precipitable water can fall in liquid or solid form, depending on the temperature. Infiltration in the basin is heterogeneous and depends on the geological units and the permeability of the soil. It is a dominant process in the basin because the infiltration 256
Annex VII
87
rates of some units are larger than 1 m/d, which is a large value compared with the
highest daily rainfall rates (15-20 mm/d). The most permeable geological units are HU2
(Alluvial) and HU3 (Ignimbrite). Hydrological response varies according to climatic
conditions. For example, after a minor rainstorm, depending on the meteorological
conditions, the water that falls on the basin can evaporate quickly from the ground
surface or from the unsaturated (vadose) zone, with little or no contribution to runoff. In
situations where air or surface temperatures fall below zero, precipitation may fall in
liquid or solid form and liquid water could freeze. If this is the case, then a small
fraction of the water will infiltrate and significant sublimation may occur. For major
rainfall events, precipitation will infiltrate and groundwater will be recharged, while
localized surface runoff may also occur. The river flow is dominated by groundwater
inflows, which include the spring-fed Orientales and Cajones sources of the Silala River
in Bolivia, and a large number of springs entering the river as it flows through the
ravine. From the above description it can be seen that temperature and precipitation
control important hydrological processes, such as streamflow. The daily variations of
temperature are the likely explanation of the daily fluctuations that are observed in the
Silala River flows.
Precipitation in the basin is mainly caused by convective activity. More than 90% of the
precipitation in the basin occurs between January and March, as a result of the
significant atmospheric moisture coming from the east. During the rest of the year the
atmospheric moisture in the area decreases, as dry winds blow from the west. The
annual precipitation at the Silala precipitation station is 99  74 mm (mean  standard
deviation). The large standard deviation is explained by the natural inter-annual
variability, which is affected by the El Niño-Southern Oscillation (ENSO). The highest
daily precipitation amounts are around 15-20 mm and take place mainly in January and
February. Moreover, a daily cycle in precipitation is observed due to the thermal
initiation of convection, which generates an active phase in the afternoon and an
inactive phase in the morning. Data from the region show a strong relationship between
annual precipitation and elevation. Using this relationship, the average annual
precipitation over the Silala River basin is estimated to be 165 mm.
Due to the seasonal variability of precipitation, the vegetation and areal extent of the
Orientales and Cajones wetlands have significant inter-annual variability (Alcayaga,
2017). During the austral summer (wet season), the surface area of the Cajones and
Orientales wetlands is ~0.16 km2, whereas in the winter (dry season) it is ~0.011 km2
(Suárez, Muñoz et al., 2017). The wetland surfaces did not show any significant longterm
change over the last 30 years, considering satellite images from 1987 to 2016
(Alcayaga, 2017). Temperature data are registered at the Inacaliri gauge, located near
the lowest point of the basin. Usually, the maximum daily temperature does not exceed
Annex VII
257
88 21 °C and the minimum daily temperature frequently reaches values below zero, between 0 °C and -10 °C. The monthly mean temperature varies from 1.8 °C (July) to 8.6 °C (March). However, there is a strong gradient of decreasing temperature with increasing elevation; higher elevation temperatures will therefore include a higher proportion of sub-zero temperatures. Potential evapotranspiration (ETo), the maximum water that can be evaporated from the soil and transpired by the vegetation when water is not limiting, varies between 2 and 8 mm/day (Suárez, Muñoz et al., 2017). The ETo estimation methods used in this investigation show that ETo is lower in winter and higher in summer, with a mean standard deviation of 0.5 mm/day (Suárez, Muñoz et al., 2017) and with annual totals of 1685  142 mm/year (mean  standard deviation). When the evapotranspiration (ET) rate is compared with the mean annual precipitation of 165 mm, it can be seen that ET losses from the basin mainly depend on the availability of precipitation. However, where water is available, for example in the spring-fed wetlands and other riparian areas, ET rates will be higher. Actual evapotranspiration (ETr) rates at the Cajones and Orientales wetlands represent 19% and 81.4% of the ETo in winter and summer, respectively. The ETr values are small compared to the river flow at the Chile-Bolivia international border. Our best estimates of this loss, using remote sensing data, are that the annual average wetland ETr is equivalent to an annual average discharge of 1.3 l/s (0.7% of the river flow at the international border). The maximum monthly value corresponds to a discharge of 5.9 l/s in February (3.3% of the river flow measured at the international border). If it is conservatively assumed that the wetlands evaporate at the potential rate (ETo), the annual ETo represents only 2% of the annual streamflow, and the highest water loss from the wetlands to the atmosphere occurs during January and is equivalent to approximately 6.5  2.2% (average  standard deviation) of the river flow at the Chile-Bolivia international border. Therefore, as these percentages are rather small, it is expected that changes in the magnitude of wetland ET due to channel constructed in 1928, would be negligible compared to the river flow. The flow in the Silala River is nearly constant at a weekly or a monthly time scale. The monthly average river flow registered at the DGA Fluviometric Station gauge is ~170-180 l/s (General Directorate of Water public website: http://www.dga.cl, where it is referred to as “Río Silala antes de Bocatoma FCAB”), with higher river flows occurring during the wet season. Nonetheless, the river flow displays daily fluctuations that can be explained by daily variations of the climatic variables. This behavior is consistent with a groundwater-dominated river system. Indeed, the BFI of the Silala River is ~0.92, meaning that 92% of the total of the streamflow at the international border is baseflow, i.e., the streamflow component that reacts slowly to rainfall and is typically associatedwith water discharged from groundwater storage (Eckhardt, 2008). 258
Annex VII
89
The Silala River is a perennial river, and its flows are dominated by groundwater
sources. The sources of its perennial flow are found in the groundwater-fed Bolivian
springs, but it also receives numerous lateral inputs from groundwater springs along its
subsequent path. For instance, Suárez et al. (2017) estimated that ~35.9 l/s of
groundwater enter the river from many springs located between the Chile-Bolivia
international border and Quebrada Negra. These springs are located in the ravine walls,
and are mainly at higher elevations than the river bed. The river is also actively
interacting with a shallow underlying fluvial aquifer; Suárez et al. (2017) estimated that
~3.3 l/s of water flows from the river into this fluvial aquifer through the riverbed
sediments. A final point to note is that an artesian well exists on Chilean territory that is
characterized by having higher water temperatures than those of the river. This artesian
well generates a significant additional groundwater contribution into the river, from a
deeper groundwater source.
Annex VII
259
90 8 REFERENCES Aceituno, P., 1996. Elementos del clima en el altiplano sudamericano. Revista Geofísica, 44, 37-55. AGRIMED, 2015. Evapotranspiración de Referencia para la Determinación de las Demandas de Riego en Chile. Facultad de Ciencias Agronómicas, Universidad de Chile. Alcayaga, H., 2017. Characterization of the Drainage Patterns and River Network of the Silala River and Preliminary Assessment of Vegetation Dynamics Using Remote Sensing. (Vol. 4, Annex I). Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evapotranspiration-Guidelines for Computing Crop Water Requirements-FAO Irrigation and Drainage Paper 56. FAO, Rome, 300(9), D05109. Arcadis, 2017. Detailed Hydrogeological Study of the Silala River. (Vol. 4, Annex II). Arnaud, Y., Vuille, M., Ribstein, P., 2001. El Niño-Southern Oscillation (ENSO) influence on a Sajama volcano glacier (Bolivia) from 1963 to 1998 as seen from Landsat data and aerial photography. Journal of Geophysical Research, 106 (16), 17773. Brušková, V., 2008. Assessment of the base flow in the upper part of Torysa River catchment. Slovak Journal of Civil Engineering, 2, 8-14. Campbell, G.S. and Norman, J.M., 2012. An introduction to environmental biophysics. 2nd Edition. Springer Science and Business Media. Doorenbos, J. and Priutt, W.O., 1977. Guidelines for predicting crop water requirements. Irrigation and Drainage Paper 24. Food and Agriculture Organization of the United Nations. Eagleman, J.R., 1967. Pan evaporation, potential and actual evapotranspiration. Journal of Applied Meteorology, 6, 482-488. Eckhardt, K., 2008. A comparison of baseflow indices, which were calculated with seven different baseflow separation methods. Journal of Hydrology, 352(1), 168-173. Falvey, M. and Garreaud, R.D., 2005. Moisture variability over the South American Altiplano during the South American Low Level Jet Experiment (SALLJEX) observing season. Journal of Geophysical Research-Atmospheres, 110, D22105. 260
Annex VII
91
Fox, R.H., 1922. The Waterworks Department of the Antofagasta (Chili) and Bolivia
Railway Company, South African Journal of Science, 19, 120-131.
Fuenzalida, H. and Rutlant, J., 1987. Origen del Vapor de Agua que Precipita sobre el
Altiplano de Chile. Paper presented at II Congreso InterAmericano de Meteorología,
Am. Meteorol. Soc., Buenos Aires.
García, M., Riquelme, R., Farías, M., Hérail, G., and Charrier, R., 2011. Late Miocene-
Holocene canyon incision in the western Altiplano, northern Chile: tectonic or climatic
forcing? Journal of the Geological Society, 168, 1047-1060.
Garreaud, R.D. and Wallace, J., 1998. Summertime incursions of midlatitude air into
subtropical and tropical South America. Monthly Weather Review, 126(10), 2713-2733.
Garreaud, R.D. and Aceituno, P., 2001. Interannual rainfall variability over the South
American Altiplano. Journal of Climate, 14, 2779-2789.
Garreaud, R.D., Vuille, M., and Clement, A.C., 2003. The climate of the Altiplano:
observed current conditions and mechanisms of past changes. Palaeogeography,
Palaeoclimatology, Palaeoecology, 194, 5-22.
Garreaud, R.D., 2009. The Andes climate and weather. Advances in Geosciences, 22, 3-
11.
Garreaud, R.D., Molina, A., and Farias, M., 2010. Andean uplift, ocean cooling and
Atacama hyperaridity: A climate modeling perspective. Earth and Planetary Science
Letters, 292, 39-50.
Garreaud, R.D., 2011. The Climate of Northern Chile: Mean State, Variability and
Trends. Institute of Astronomy, UNAM, 41, 5–11.
Gayo, E. M., Latorre, C., Jordan, T.E., Nester, P.L., Estay, S.A., Ojeda, K.F. and
Santoro, C.M., 2012. Late quaternary hydrological and ecological changes in the
hyperarid core of the northern Atacama Desert (~ 21° S). Earth-Science Reviews, 130
(3-4), 120-140.
Groeneveld, D.P., Baugh, W.M., Sanderson, J.S., Cooper, D.J., 2007. Annual
groundwater evapotranspiration mapped from single satellite scenes. Journal of
Hydrology, 344, 146-156.
Hardy, D.R., Vuille, M., Braun, C., Keimig, F., and Bradley, R.S., 1998. Annual and
daily meteorological cycles at high altitude on a tropical mountain. Bulletin of the
American Meteorological Society, 79, 1899-1913.
Hastenrath, S. and Kutzbach, J., 1985. Late Pleistocene climate and water budget of the
South American Altiplano. Quaternary Research, 24, 249-256.
Annex VII
261
92 Hauser, A., 2004. Marco Morfológico, Geológico, Tectónico, Hidrogeológico e Hidroquímico: Morfogénesis, Evolución y Modalidades de Aprovechamiento del Sistema Hidrográfico Compartido Chileno-Boliviano del Río Silala. Servicio Nacional de Geología y Minería, Informe consolidado. Santiago. (Vol. 4, Annex II, Appendix A). Herrera, C. and Aravena, R., 2017. Chemical and Isotopic Characterization of the Surface Water and Groundwater of the Silala River Transboundary Basin, Second Region, Chile. (Vol. 4, Annex III). Houston, J.and Hartley, A., 2003. The Central Andean west-slope rainshadow and its potential contribution to the origin of hyper-aridity in the Atacama Desert. Int. J. Climatol. 23, 1453. Institute of Hydrology, 1980. Low Flows Studies Report, 3 volumes. Institute of Hydrology, Wallingford, UK. Latorre, C., Frugone, M., 2017. Holocene Sedimentary History of the Río Silala (Antofagasta Region, Chile). (Vol. 5, Annex IV). Lenters, J.D. and Cook, K.H., 1997. On the origin of the Bolivian high and related circulation features of the South American climate. Journal of the Atmospheric Sciences, 54, 656-678. Lettau, H.H. and Costa, J.R, 1978. Characteristic winds and boundary layer meteorology of the arid zones in Peru and Chile. In Lettau, H.H. and Lettau, K. (eds.) Exploring the World’s Driest Climates. Center for Climatic Research, pp. 163-181. Lundquist, J. and Cayan, D., 2002. Seasonal and spatial patterns in diurnal cycles in streamflow in the Western United States. Journal of Hydrometeorology, 3, 591–603. Mao, L., 2017. Fluvial Geomorphology of the Silala River, Second Region, Chile. (Vol. 5, Annex V). Marengo, J. A., Soares, W.R., Saulo, C. and Nicolini, M., 2004. Climatology of the low-level jet east of the Andes as derived from the NCEP-NCAR reanalyses: Characteristics and temporal variability. Journal of Climate, 17(12), 2261-2280. Norambuena, P., Luzio, W., Zepeda O., Stern, J. and Reinoso, F., 2011. Preliminary survey of some soils from Chilean Altiplano near Iquique. Journal of Soil Science and Plant Nutrition, 11(2), 62-71. Rahn, D. A. and Garreaud, R.D., 2010. Marine boundary layer over the subtropical southeast Pacific during VOCALS-REx - Part 1: Mean structure and diurnal cycle. Atmospheric Chemistry and Physics, 10, 4491-4506. 262
Annex VII
93
Ronchail, J., Gallaire, R., 2006. ENSO and rainfall along the Zongo Valley (Bolivia)
from the Altiplano to the Amazon Basin. Int. J. Climatol. 26, 1223.
Seiler, C., Hutjes, R. W., and Kabat, P., 2013. Climate variability and trends in
Bolivia. Journal of Applied Meteorology and Climatology, 52(1), 130-146.
Senay, G.B., Leake, S., Nagler, P.L., Artan, G., Dickinson, J., Cordova, J.T. and Glenn,
E.P., 2011. Estimating basin scale evapotranspiration (ET) by water balance and remote
sensing methods. Hydrol. Process. 25, 4037.
SERNAGEOMIN, 2017. Geology of the Silala River Basin. (Vol. 5, Annex VIII).
Suárez, F., Muñoz, J.F., Maass, T., Mendoza, M., 2017. Evapotranspiration Estimation
in the Silala River Basin - Methods Review and Estimation of Wetland Evaporation.
(Vol. 5, Annex IX).
Suárez, F., Sandoval, V., Sarabia, A., 2017. River-Aquifer Interactions Using Heat as a
Tracer in the Transboundary Basin of the Silala River. (Vol. 5, Annex X).
Vera, C., Higgins, W., Amador, J., Ambrizzi, T., Garreaud, R.D., Gochis, D., Gutzler,
D., Lettenmaier, D., Marengo, J., Mechoso, C.R., Nogues-Paegle, J., Dias, P.L.S., and
Zhang, C., 2006. Toward a unified view of the American Monsoon Systems. Journal of
Climate, 19, 4977-5000.
Vuille, M., 1999. Atmospheric circulation over the Bolivian Altiplano during dry and
wet periods and extreme phases of the Southern Oscillation. International Journal of
Climatology, 19, 1579-1600.
Vuille, M., Bradley, R.S., Keiming, F., 2000. Interannual climate variability in the
Central Andes and its relation to tropical Pacific and Atlantic forcing. Journal of
Geophysical Research, 105(10), 12447.
Vuille, M., Keiming, F., 2004. Interannual variability of summertime convective
cloudiness and precipitation in the Central Andes derived from ISCCP-B3 data. Journal
of Climate,17, 3334.
Vuille, M., Bradley, R.S., Healy, R., Werner, M., Hardy, D.R., Thompson, L. and
Keimig, F., 2003. Modeling δ18O in precipitation over the tropical Americas: 2.
Simulation of the stable isotope signal in Andean ice cores. Journal of Geophysical
Research, 108, 4175.
Vuille, M., and R.D. Garreaud., 2012. Ocean-atmosphere interactions on interannual to
decadal timescales. In Matthews, J.A., P.J. Bartlein, K.R. Briffa, A.G. Dawson, A. De
Vernal, T. Denham, S.C. Fritz and F. Oldfield (Eds.), Handbook of Environmental
Change. Sage Publications, London, Los Angeles, New Delhi, Singapore, p 471-496.
Annex VII
263
94 Xu, C.-Y., and Singh, V.P., 2000. Evaluation and generalization of radiation-based methods for calculating evaporation. Hydrological Processes, 14(2), 339-349. 264
Annex VII
95
APPENDIX A – DATA USED IN THE CHARACTERIZATION OF THE
PRECIPITATION REGIME
Figure A.1. Location of rain gauges selected for the study.
Annex VII Appendix A
265
96 Source Code Station Name Elevation UTM (WGS 84-19S) m.a.s.l. East North DGA 01080001-3 UJINA 4300 538722 7680944 DGA 01080002-1 COLLAHUASI 4250 520585 7678240 DGA 01700009-8 COPOSA 3760 531847 7710041 DGA 01770001-4 COPAQUIRE 3540 511206 7685535 DGA 02000001-5 OLLAGUE 3700 577892 7652920 DGA 02020001-4 CEBOLLAR 3730 568495 7618895 DGA 02020002-2 ASCOTAN 3970 574925 7597374 DGA 02101003-0 LEQUENA 3320 535065 7605259 DGA 02102005-2 QUINCHAMALE 3080 541690 7577571 DGA 02103007-4 SAN PEDRO DE CONCHI 3217 547879 7574448 DGA 02103008-2 PARSHALL Nº 2 3318 549826 7573397 DGA 02103009-0 OJOS SAN PEDRO 3800 570954 7570241 DGA 02103010-4 INACALIRI 4040 596385 7563772 DGA 02103012-0 SILALA 4305 600236 7565285 DGA 02104007-K OLD CONCHI 3491 528511 7572650 DGA 02104008-8 CONCHI RESERVOIR 3010 538788 7564326 DGA 02104009-6 CONCHI RESERVOIR WALL 3000 539161 7561141 DGA 02104010-K CHIU-CHIU 2524 537415 7529707 DGA 02105002-4 SALADO RIVER AT AYQUINA SIPHON 2980 567504 7534956 DGA 02105014-8 CUPO 3370 570481 7554284 DGA 02105015-6 TURI 3070 573367 7540311 DGA 02105016-4 LINZOR 4100 600714 7541388 DGA 02105017-2 TOCONCE 3310 585178 7538221 DGA 02105018-0 AYQUINA 3031 570229 7536512 DGA 02105020-2 SALADO RESERVOIR 3200 582129 7535316 DGA 02105021-0 CASPANA 3260 581126 7529664 DGA 02105022-9 EL TATIO 4370 601585 7525730 DGA 02110013-7 CALAMA 2300 509832 7517389 DGA 02111004-3 SLOMAN DAM 985 446897 7583594 DGA 02112008-1 QUILLAGUA 802 444557 7604648 DGA 02112009-K COYA SUR 1250 435920 7523409 266
Annex VII Appendix A
97
Source Code Station Name
Elevation UTM (WGS 84-19S)
m.a.s.l. East North
DGA 02113005-2 GUATACONDO DGA 2460 495262 7685876
DGA 02210002-5 TOCOPILLA 150 377253 7557838
DGA 02500015-3 TOCONAO POLICE STATION 2460 601376 7434668
DGA 02500016-1 TOCONAO EXPERIMENTAL 2500 602573 7435152
DGA 02500017-K CAMAR 2700 605954 7411015
DGA 02500019-6 SOCAIRE 3251 613011 7391031
DGA 02500020-K PEINE 2460 596017 7380388
DGA 02500021-8 TALABRE 3300 613727 7421415
DGA 02510006-9 SAN PEDRO DE ATACAMA 2450 582019 7466710
DGA 02510007-7 RIO GRANDE 3250 585709 7494705
DGA 02700001-0 SIERRA GORDA 1616 466978 7468680
DGA 02710002-3 BAQUEDANO 1032 414151 7419265
DGA 02710003-1 ANTOFAGASTA 50 358510 7389614
SENAMHI COLCHA K 3780 639691 7706417
SENAMHI LAGUNA COLORADA 4278 633991 7542616
SENAMHI SAN AGUSTIN 4230 629774 7658896
SENAMHI SAN PABLO DE LIPEZ 4230 746604 7600327
SENAMHI UYUNI 3669 729513 7737180
Table A.1. List of rain-gauge stations selected for the study.
Annex VII Appendix A
267
98 Station ORIGINAL INFORMATION COMPLETED INFORMATION Years* Average [mm] Deviation [mm] Years* Average [mm] Deviation [mm] UJINA 20 169.8 126.8 29 165.8 109.8 COLLAHUASI 15 146.2 62.4 33 134.1 70.5 COPOSA 13 94.6 82.7 24 81 66.3 COPAQUIRE 25 85.5 119.7 25 85.5 119.7 OLLAGUE 16 82.2 58.2 18 79 56 CEBOLLAR 23 61 39.7 29 57.2 38.8 ASCOTAN 28 70.9 49.6 35 68.9 46.3 LEQUENA 34 85.7 126.9 36 81.6 124.5 QUINCHAMALE 14 21.2 23.5 29 24.7 20.5 SAN PEDRO DE CONCHI 19 33.6 29.9 19 33.6 29.9 PARSHALL Nº 2 37 28.5 25.2 38 28 25 OJOS SAN PEDRO 29 63,3 42.3 34 70.6 49.1 INACALIRI 34 123,3 96.1 39 120.9 92.5 SILALA 9 60 43.5 28 96.8 76.1 OLD CONCHI 30 39.4 41.2 31 38.4 40.9 CONCHI RESERVOIR 31 18.3 14.4 31 18.3 14.4 CONCHI RESERVOIR WALL 11 19.7 13.3 27 17.5 14.2 CHIU-CHIU 33 6 5.6 33 6 5.6 SALADO RIVER AT AYQUINA SIPHON 7 16.3 8.9 31 24.3 23 CUPO 27 63.6 74.6 29 64.2 72.9 TURI 9 39.3 39.1 27 40.8 41.3 LINZOR 25 183 118.6 34 161.4 113.2 TOCONCE 34 92.5 73.4 35 90.8 73.1 AYQUINA 37 37.1 42.4 41 37.6 42.1 SALADO RESERVOIR 30 68.8 68 35 81.9 83.7 CASPANA 32 84.8 82.7 35 81.1 80.7 EL TATIO 27 149.4 104.2 32 141.2 100 CALAMA 32 3.6 4.4 32 3.6 4.4 SLOMAN DAM 8 0 0 8 0 0 QUILLAGUA 32 0.2 0.6 32 0.2 0.6 268
Annex VII Appendix A
99
Station
ORIGINAL
INFORMATION
COMPLETED
INFORMATION
Years*
Average
[mm]
Deviation
[mm]
Years*
Average
[mm]
Deviation
[mm]
COYA SUR 14 0.7 1.7 14 0.7 1.7
GUATACONDO DGA 31 21.7 27.9 31 21.7 27.9
TOCOPILLA 16 1 4 16 1 4
TOCONAO POLICE STATION 15 38.4 29.7 28 31.2 27.3
TOCONAO EXPERIMENTAL 27 35.1 32.7 34 29.9 31.6
CAMAR 30 33.4 32.5 34 34.1 31.3
SOCAIRE 33 37.3 36.2 35 39.1 38.2
PEINE 33 19.2 20 33 19.2 20
TALABRE 13 56.6 46.5 30 64.5 50.8
SAN PEDRO DE ATACAMA 25 25 21.7 25 25 21.7
RIO GRANDE 29 67.3 59.1 37 71.7 58.8
SIERRA GORDA 15 1 2.1 15 1 2.1
BAQUEDANO 23 1.5 3.1 23 1.5 3.1
ANTOFAGASTA 32 3.3 6.3 32 3.3 6.3
COLCHA K 25 187.9 125.2 30 196.3 127.7
LAGUNA COLORADA 8 50.4 30 8 50.4 30
SAN AGUSTIN 14 169.3 157.7 14 169.3 157.7
SAN PABLO DE LIPEZ 29 306.4 133 29 306.4 133
UYUNI 29 190.3 114.7 30 189.5 112.7
(*) Number of hydrological years with complete information.
Table A.2. Statistical comparison of the original and completed precipitation series.
Annex VII Appendix A
269
100 Table A.3. Availability of information for precipitation gauges used in this study. STATION1011UJINACOLLAHUASICOPOSACOPAQUIREOLLAGUE CEBOLLAR ASCOTAN LEQUENA QUINCHAMALE SAN PEDRO DE CONCHI PARSHALL Nº 2 OJOS SAN PEDRO INACALIRI SILALA CONCHI VIEJO CONCHI EMBALSE CONCHI MURO EMBALSE CHIU-CHIU RIO SALADO EN SIFON AYQUINA CUPO TURI LINZOR TOCONCE AYQUINA SALADO EMBALSE CASPANA EL TATIO CALAMA TRANQUE SLOMAN QUILLAGUA COYA SUR GUATACONDO DGA TOCOPILLA TOCONAO RETEN TOCONAO EXPERIMENTAL CAMAR SOCAIRE PEINE TALABRE SAN PEDRO DE ATACAMA RIO GRANDE SIERRA GORDA BAQUEDANO ANTOFAGASTA Colcha KLaguna ColoradaSan AgustinSan Pablo de LipezUyuniYear with information for 10 or 11 monthsYear with information for 7 to 9 months Year with information for less than 7 monthsLegend:1960-19691970-19791980-19891990-19992000-2009Year with complete information101 270
Annex VII Appendix A
101
APPENDIX B – DATA USED IN THE CHARACTERIZATION OF THE
TEMPERATURE
Figure B.1. Location of meteorological stations that measure temperature selected for the study.
Annex VII Appendix B
271
102 Source Code Station Name Elevation UTM (WGS84 – 19S) m.a.s.l. East North DGA 02000001-5 OLLAGUE 3700 577892 7652920 DGA 02020001-4 CEBOLLAR 3730 568495 7618895 DGA 02020002-2 ASCOTAN 3970 574925 7597374 DGA 02101003-0 LEQUENA 3320 535065 7605259 DGA 02103008-2 PARSHALL Nº 2 3318 549826 7573397 DGA 02103009-0 OJOS SAN PEDRO 3800 570954 7570241 DGA 02103010-4 INACALIRI 4040 596385 7563772 DGA 02104008-8 CONCHI RESERVOIR 3010 538788 7564326 DGA 02104009-6 CONCHI RESERVOIR WALL 3000 539161 7561141 DGA 02104010-K CHIU-CHIU 2524 537415 7529707 DGA 02105015-6 TURI 3070 573367 7540311 DGA 02105016-4 LINZOR 4100 600714 7541388 DGA 02105017-2 TOCONCE 3310 585178 7538221 DGA 02105018-0 AYQUINA 3031 570229 7536512 DGA 02105021-0 CASPANA 3260 581126 7529664 DGA 02105022-9 EL TATIO 4370 601585 7525730 DGA 02110013-7 CALAMA 2300 509832 7517389 DGA 02112008-1 QUILLAGUA 802 444557 7604648 DGA 02112009-K COYA SUR 1250 435920 7523409 DGA 02113005-2 GUATACONDO DGA 2460 495262 7685876 DGA 02500016-1 TOCONAO EXPERIMENTAL 2500 602573 7435152 DGA 02500019-6 SOCAIRE 3251 613011 7391031 DGA 02500020-K PEINE 2460 596017 7380388 DGA 02510006-9 SAN PEDRO DE ATACAMA 2450 582019 7466710 DGA 02660001-4 MONTURAQUI 3430 557288 7307823 DGA 02700001-0 SIERRA GORDA 1616 466978 7468680 DGA 02710002-3 BAQUEDANO 1032 414151 7419265 SENAMHI COLCHA K 3780 639691 7706417 272
Annex VII Appendix B
103
SENAMHI LAGUNA COLORADA 4278 633991 7542616
SENAMHI SAN AGUSTIN 4230 629774 7658896
SENAMHI SAN PABLO DE LIPEZ 4230 746604 7600327
Table B.1. List of meteorological stations that measure temperature selected for the study.
Source Code Station Name Annual T (°C)
DGA 02000001-5 OLLAGUE 6.81
DGA 02020001-4 CEBOLLAR 6.61
DGA 02020002-2 ASCOTAN 4.52
DGA 02101003-0 LEQUENA 8.69
DGA 02103008-2 PARSHALL Nº 2 10.17
DGA 02103009-0 OJOS SAN PEDRO 3.04
DGA 02103010-4 INACALIRI 5.26
DGA 02104008-8 CONCHI RESERVOIR 10.05
DGA 02104009-6 CONCHI RESERVOIR WALL 9.95
DGA 02104010-K CHIU-CHIU 11.96
DGA 02105015-6 TURI 9.29
DGA 02105016-4 LINZOR 4.18
DGA 02105017-2 TOCONCE 10.98
DGA 02105018-0 AYQUINA 14.71
DGA 02105021-0 CASPANA 12.21
DGA 02105022-9 EL TATIO 10.41
DGA 02110013-7 CALAMA 2.14
DGA 02112008-1 QUILLAGUA 13.50
DGA 02112009-K COYA SUR 17.28
DGA 02113005-2 GUATACONDO DGA 20.21
DGA 02500016-1 TOCONAO EXPERIMENTAL 16.05
DGA 02500019-6 SOCAIRE 10.56
DGA 02500020-K PEINE 16.51
DGA 02510006-9 SAN PEDRO DE ATACAMA 14.40
DGA 02660001-4 MONTURAQUI 7.29
DGA 02700001-0 SIERRA GORDA 17.54
Annex VII Appendix B
273
104 DGA 02710002-3 BAQUEDANO 16.53 SENAMHI COLCHA K 10.17 SENAMHI LAGUNA COLORADA 0.06 SENAMHI SAN AGUSTIN 7.67 SENAMHI SAN PABLO DE LIPEZ 5.60 Table B.2. Mean annual temperatures at the meteorological stations selected for the study. Table B.3. Availability of temperature data at the meteorological stations used in this study. CODESTATION101102000001-5OLLAGUE 02020001-4CEBOLLAR02020002-2ASCOTAN02101003-0LEQUENA02103008-2PARSHALL Nº 202103009-0OJOS SAN PEDRO02103010-4INACALIRI02104008-8CONCHI EMBALSE02104009-6CONCHI MURO EMBALSE02104010-KCHIU-CHIU 02105015-6TURI 02105016-4LINZOR02105017-2TOCONCE02113005-2GUATACONDO DGA02105018-0AYQUINA02105021-0CASPANA02105022-9EL TATIO02110013-7CALAMA02112008-1QUILLAGUA02112009-KCOYA SUR02500016-1TOCONAO EXPERIMENTAL02500019-6SOCAIRE02500020-KPEINE02510006-9SAN PEDRO DE ATACAMA02660001-4MONTURAQUI02700001-0SIERRA GORDA02710002-3BAQUEDANOCOLCHA KLAGUNA COLORADASAN AGUSTINSAN PABLO DE LIPEZYear with complete informationYear with information for 10 or 11 monthsYear with information for 7 to 9 months Year with information for less than 7 months1960-19691970-19791980-19891990-19992000-2009Legend:274
Annex VII Appendix B
105
APPENDIX C – DATA USED IN THE CHARACTERIZATION OF THE
EVAPOTRANSPIRATION
Figure C.1. Location of the evaporation pans selected for this study.
Annex VII Appendix C
275
106 Source Code Station Name Elevation UTM (WGS84-19S) Mean annual T Mean annual E m.a.s.l. East North °C mm/year DGA 01080001-3 UJINA 4,300 538722 7680944 - 1,874 DGA 02000001-5 OLLAGUE 3,700 577892 7652920 6.81 2,645 DGA 02020001-4 CEBOLLAR 3,730 568495 7618895 6.61 2,497 DGA 02101003-0 LEQUENA 3,320 535065 7605259 8.69 2,683 DGA 02103007-4 SAN PEDRO DE CONCHI 3,217 547879 7574448 - 3,674 DGA 02103008-2 PARSHALL Nº 2 3,318 549826 7573397 10.17 3,868 DGA 02103009-0 OJOS SAN PEDRO 3,800 570954 7570241 3.04 2,339 DGA 02103010-4 INACALIRI 4,040 596385 7563772 5.26 2,376 DGA 02104008-8 CONCHI EMBALSE 3,010 538788 7564326 10.05 3,509 DGA 02104009-6 CONCHI MURO EMBALSE 3,000 539161 7561141 9.95 2,950 DGA 02104010-K CHIU-CHIU 2,524 537415 7529707 11.96 2,547 DGA 02105015-6 TURI 3,070 573367 7540311 9.29 3,014 DGA 02105016-4 LINZOR 4,100 600714 7541388 4.18 1,935 DGA 02105018-0 AYQUINA 3,031 570229 7536512 12.21 3,474 DGA 02105021-0 CASPANA 3,260 581126 7529664 10.41 2,197 DGA 02105022-9 EL TATIO 4,370 601585 7525730 2.14 1,583 DGA 02110013-7 CALAMA 2,300 509832 7517389 13.5 3,247 DGA 02112009-K COYA SUR 1,250 435920 7523409 20.21 3,715 DGA 02500016-1 TOCONAO EXPERIMENTAL 2,500 602573 7435152 16.05 3,547 DGA 02500019-6 SOCAIRE 3,251 613011 7391031 10.56 3,270 DGA 02500020-K PEINE 2,460 596017 7380388 16.51 3,373 DGA 02510006-9 SAN PEDRO DE ATACAMA 2,450 582019 7466710 14.4 3,094 DGA 02660001-4 MONTURAQUI 3,430 557288 7307823 7.29 2,972 - : no temperature record is available. Table C.1. Mean annual evaporation values at the stations selected for the study. 276
Annex VII Appendix C
107
Figure C.2. Stations and towns where ETo is calculated in this study.
Annex VII Appendix C
277
108 Study area Potential Evapotranspiration Antofagasta Region Figure C.3. Reference crop evapotranspiration maps for the Second Region of Antofagasta (AGRIMED, 2015).
ETo (mm/year)
278
Annex VII Appendix C
Annex VIII
SERNAGEOMIN (National Geology and Mining Service), 2017. Geology of the Silala River Basin
279
280
Annex VIII
GEOLOGY OF THE SILALA RIVER BASIN
Nicolás Blanco P. (MSc)
Geologist Project Manager of the Regional Geology Unit of the Department of Basic
Geology
Edmundo Polanco V. (DSc)
Project Geologist of the Regional Geology Unit of the Department of Basic Geology
May, 2017
Annex VIII
281
GLOSSARY This glossary of geologic terms is based on the glossary in Earth: An Introduction to Geologic Change, by S. Judson and S.M. Richardson (Englewood Cliffs, NJ, Prentice Hall, 1995). Where possible, definitions conform generally, and in some cases specifically, to definitions given in Robert L. Bates and Julia A. Jackson (editors), Glossary of Geology, 3rd ed., American Geological Institute, Alexandria, Virginia, 1987. 14C method: A method for determining the age in years of organic matter by calculating the amount of radioactive carbon still remaining, as compared to the stable isotope, 12C. 39Ar/40Ar method: A different method that was invented to supersede K/Ar method, to be more accurate. 40K/40Ar method: A method used for the dating of potassium-bearing rocks by using the ratio of radioactive 40K to its daughter, 40Ar. Absolute time: Geologic time expressed in years before the present. Alveoli: (Honeycomb weathering) Term used to describe numerous small pits or alveoli, no more than a few centimetres wide and deep, separated by an intricate network of narrow walls and resembling a honeycomb. They are often thought of as a small-scale version of multiple tafoni. Honeycomb weathering is particularly evident in semiarid and coastal environments where salts are in ready supply and wetting and drying cycles are common (Huggett, 2007). Alluvial fan: Land counterpart of a delta. An assemblage of sediments marking place where a stream moves from a steep gradient to a flatter gradient and suddenly loses transporting power. Typical of arid and semiarid climates but not confined to them. Andesite: A fine-grained volcanic rock of intermediate composition, consisting largely of plagioclase and one or more mafic minerals. Angular unconformity: An unconformity in which the beds below the unconformity dip at a different angle than the beds above it. Aquifer: Geological formation capable of storing, transmitting and yielding exploitable quantities of water. Assemblage: The collection of minerals that characterize a rock or a facies. 282
Annex VIII
Basalt: A dark colored extrusive igneous rock composed chiefly of calcium plagioclase
and pyroxene. Extrusive equivalent of gabbro, underlies the ocean basins and comprises
oceanic crust.
Bedding: A collective term used to signify presence of beds, or layers, in sedimentary
rocks and deposits.
Bedding plane: Surface separating layers of sedimentary rocks and deposits. Each
bedding plane marks termination of one deposit and beginning of another of different
character, such as a surface separating a sandstone bed from an overlying mudstone bed.
Rock tends to breaks or separate readily along bedding planes.
Bedrock: Any solid rock exposed at the Earth’s surface or overlain by unconsolidated
material.
Breccia: A clastic rock in which the gravel-sized particles are angular in shape and
make up an appreciable volume of the rock.
Caliche: Also called Hardpan, calcium-rich duricrust, a hardened layer in or on a soil. It
is formed on calcareous materials as a result of climatic fluctuations in arid and semiarid
regions. Calcite is dissolved in groundwater and, under drying conditions, is precipitated
as the water evaporates at the surface. Rainwater saturated with carbon dioxide acts as
an acid and also dissolves calcite and then redeposits it as a precipitate on the surfaces
of the soil particles; as the interstitial soil spaces are filled, an impermeable crust is
formed (https://global.britannica.com/science/calcrete).
Cavetto: Is essentially a parallel-sided furrow on the outside of a bend, eroded into a
near-vertical face so that the upper rim of the furrow is strongly overhanging. Cavettos
vary greatly in size, and in deep, narrow channels their vertical dimension may
approach the width of the channel. In the case of dip-parallel reaches, they are often
structurally influenced, tending to pick out less resistant beds (Richardson and Carlin,
2005).
Caldera: A large, basin-shaped volcanic depression, more or less circular in form.
Typically steep-sided, found at the summit of a shield volcano.
Cirque: A steep-walled hollow in a mountain side, shaped like an amphitheater, or
bowl, with one side partially cut away. Place of origin of a mountain glacier.
Clastic: Refers to rock or sediments made up primarily of broken fragments of preexisting
rocks or minerals.
Composite: Volcano see stratovolcano.
Conformable: Lying parallel to, rather than cutting across surrounding strata.
Annex VIII
283
Crater: 1. A steep-walled, usually conical depression at the summit or on the flanks of a volcano, resulting from the explosive ejection of material from a vent. 2. A bowl-shaped depression with a raised, overturned rim produced by the impact of a meteorite or other energetic projectile. Cross-lamination: Layering within a stratum and at an angle to the main bedding plane. Crystal: The multi-sided form of a mineral, bounded by planar growth surfaces, that is the outward expression of the ordered arrangement of atoms within it. Debris flow: Fast-moving, turbulent mass movement with a high content of both water and rock debris. The more rapid debris flows rival the speed of rock slides. Differentiation: The process of developing more than one rock type, in situ, from a common magma. Dip: The angle that a structural surface such as a bedding plane or fault surface makes with the horizontal, measured perpendicular to the strike and in the vertical plane. Dome: An uplift or anticlinal structure, roughly circular in its outcrop exposure, in which beds dip gently away from the center in all directions. Extrusive: Pertaining to igneous rocks or features formed from lava released on the Earth’s surface. Fault gouge: Soft, uncemented, pulverized clay-like material found along some faults. Foot wall block: The body of rock that lies below an inclined fault plane -compare hanging wall block-. Geologic time scale: The chronological sequence of units of Earth time. Glaciation: The formation, advance and retreat of glaciers and the results of these activities. Glacier: A mass of ice, formed by the recrystallization of snow, that flows forward, outward or downwards, or has flowed at some time in the past. Glassy: A texture of extrusive igneous rocks that develops as the result of rapid cooling, so that crystallization is inhibited. Ground moraine: Till deposited from main body of glacier during ablation. Hanging wall block: The body of rock that lies above an inclined fault plane. Igneous rock: A rock that has crystallized from a molten state. Joint: A surface of fracture in a rock, without displacement parallel to the fracture. Lateral moraine: Moraine formed by valley glaciers along valley sides. 284
Annex VIII
Lava: Molten rock that flows at the Earth’s surface.
Lava dome: A steep-sided rounded extrusion of highly viscous lava squeezed out from
a volcano and forming a dome-shaped or bulbous mass above and around the volcanic
vent. The structure generally develops inside a volcanic crater.
Magma: Molten rock, containing dissolved gases and suspended solid particles. At the
Earth’s surface, magma is known as lava.
Mineral: A naturally occurring inorganic solid that has a well-defined chemical
composition and in which atoms are arranged in an ordered fashion.
Moraine: Any glacially formed accumulation of unconsolidated glacial debris (soil and
rock) that occurs in both currently and formerly glaciated regions on Earth (i.e. a past
glacial maximum), through geomorphological processes. Moraines are formed from
debris previously carried along by a glacier and normally consist of somewhat rounded
particles ranging in size from large boulders to minute glacial flour. Lateral moraines
are formed at the side of the ice flow and terminal moraines at the foot, marking the
maximum advance of the glacier. Other types of moraine include ground moraines, tillcovered
areas with irregular topography, and medial moraines which are formed where
two glaciers meet.
Normal fault: A geological fault where the hanging wall block has moved downwards
relative to the foot wall block.
Oxidation: The decomposition process by which iron or other metallic elements in a
rock combine with oxygen to form residual oxide minerals.
Phenocryst: Any relatively large, conspicuous crystal in a porphyritic igneous rock.
Pipe: A vertical conduit through the Earth’s crust below a volcano, through which
magma has passed.
Porphyritic: A texture of an igneous rock in which large crystals (phenocrysts) are set
in a matrix of relatively finer-grained crystals or of glass.
Potholes: Round to oval shaped holes in the bedrock of a river bed. They are created
where sediment accumulates within naturally occurring small depressions on the rock
surface on the river bed. Turbulent flow swirls the stones (called grinders) around in the
depressions, widening and deepening them through the prolonged process of abrasion.
As the holes gets bigger even bigger debris can become trapped in the pothole, and this
material is again used as an abrasive tool (http://www.coolgeography.co.uk/Alevel/
AQA/Year%2012/Rivers_Floods/Landforms/Landforms.htm).
Annex VIII
285
Pyroclastic: Pertaining to clastic material formed by volcanic explosion or aerial expulsion from a volcanic vent. Pyroclastic flow: A dense, hot (sometimes incandescent) cloud of volcanic ash and gas produced in a Pelean eruption. Reverse fault: A dip-slip fault on which the hanging wall block is offset upward relative to the foot wall block. Rhyolite: A fine-grained silica-rich igneous rock, the extrusive equivalent of granite. Rift (graben): A valley caused by extension of the Earth’s crust. Its floor forms as a portion of the crust moves downward along normal faults. Rock: An aggregate of one or more minerals in varying proportions. Sedimentary rock: Rock formed from the accumulation of sediment, which may consist of fragments and mineral grains of varying sizes from pre-existing rocks, remains or products of animals and plants, the products of chemical action, or mixtures of these. Silica: Silicon dioxide (SiO2) as a pure crystalline substance makes up quartz and related forms such as flint and chalcedony. More generally, silica is the basic chemical constituent common to all silicate minerals and magmas. Sorting: The range of particle sizes in a sedimentary deposit. A deposit with a narrow range of particle sizes is termed “well-sorted”. Stratovolcano (composite volcano): A volcano that is composed of alternating layers of lava and pyroclastic material, along with abundant dikes and sills. Viscous, intermediate lava may flow from a central vent. Example: Mt. Fuji in Japan. Tafoni (singular tafone): Large weathering features that take the form of hollows or cavities on a rock surface. They tend to form in vertical or near-vertical faces of rock. They can be as little as 0.1 m to several metres in height, width, and depth, with arched-shaped entrances, concave walls, sometimes with overhanging hoods or visors, especially in case-hardened rocks, and smooth and gently sloping, debris-strewn floors. The origins of tafoni are complex. Salt action is the process commonly invoked in tafoni formation, but researchers cannot agree whether the salts promote selective chemical attack or whether they promote physical weathering, the growing crystals prising apart grains. Tafoni are common in coastal environments but are also found in arid environments. Some appear to be relict forms (Huggett, 2007). Tephra: A general term for all pyroclastic material. 286
Annex VIII
Terrace: A relatively flat surface along a valley, with a steep bank separating it either
from the floodplain, or from a lower terrace.
Texture: The general appearance of a rock as shown by the size, shape, and
arrangement of the materials composing it.
Tuff: A general term for all consolidated pyroclastic rock. Not to be confused with tufa.
Turbulent flow: Fluid flow in which the flow lines are confused and mixed. Fluid
moves in eddies and swirls.
Unconformity: A buried erosion surface separating two rock masses.
Vesicle: A cavity in a lava, formed by the entrapment of a gas bubble during
solidification of the lava.
Vesicular: A textural term applied to an igneous rock containing abundant vesicles,
formed by the expansion of gases initially dissolved in the lava.
Vergence: Inclination direction of a fold or fault
(http://ceramica.wikia.com/wiki/Glosario_de_Geolog%C3%ADa_Ingl%C3%A9s_%E2
%80%93_Espa%C3%B1ol:_V).
Volcanic ash: The dust-sized, sharp-edged, glassy particles resulting from an explosive
volcanic eruption.
Volcano: A vent in the surface of the Earth, from which lava, ash, and gases erupt,
forming a structure that is roughly conical.
Weathering: The destructive process by which earth materials on exposure to
atmospheric agents (water, wind, temperature etc.) at or near the Earth's surface are
changed in color, texture, composition, firmness, or form, with little or no transport of
the loosened or altered material.
Welded tuff: A pyroclastic rock in which glassy clasts have been fused by the
combination of the heat retained by the clasts, the weight of overlying material, and hot
gases.
Annex VIII
287
TABLE OF CONTENTS 1.INTRODUCTION ........................................................................................................ 11.1. Context ............................................................................................................... 1 1.2. Location of the investigated area........................................................................ 1 1.3. Objectives of the report ...................................................................................... 2 1.3.1. General Objective ....................................................................................... 2 1.3.2. Specific Objectives ..................................................................................... 2 1.4. Summary of the methodology ............................................................................ 3 1.5. Structure of the report......................................................................................... 4 2.SUMMARY OF THE REGIONAL GEOLOGY AROUND THE SILALA RIVERBASIN ............................................................................................................................... 5 3.GEOLOGICAL SUMMARY OF THE SILALA RIVER BASIN ............................... 84.DETAILED STRATIGRAPHY ................................................................................. 125.STRUCTURAL GEOLOGY ...................................................................................... 276.EVOLUTION OF THE SILALA RIVER RAVINE .................................................. 297.EVIDENCE OF THE FORMATION AND EROSION OF THE SILALA RIVERRAVINE.......................................................................................................................... 36 8.CONCLUSIONS ......................................................................................................... 38REFERENCES................................................................................................................ 41 APPENDIX A APPENDIX B APPENDIX C APPENDIX D APPENDIX E APPENDIX F APPENDIX G 288
Annex VIII
1
1. INTRODUCTION
1.1. Context
In connection with Note OF. SEC. No. 43 dated September 13, 2016, from the Ministry
of Foreign Affairs (National Department for State Borders and Boundaries, DIFROL)
concerning technical cooperation with the National Geology and Mining Service
(SERNAGEOMIN), the National Deputy Director of Geology, Mr. Mario Pereira,
commissioned geologists Nicolás Blanco (MSc) and Edmundo Polanco (DSc) from the
Department of Basic Geology. This cooperation involved the production of a report on
the geological evolution of the Silala River basin. This was carried out under the
supervision of Difrol’s international experts, Drs. Denis Peach and Howard Wheater.
1.2. Location of the investigated area
The Silala River basin is located in the volcanic arc of the high Andean Mountain
Range in the Second Region of Chile (Antofagasta Region) and the Department of
Potosí of Bolivia, approximately 100 km NE from the city of Calama (Fig. 1). In
particular, the Silala River basin crosses the border between Chile and Bolivia.
Annex VIII
289
2 Fig. 1. Location of study area, in the Silala River basin (outlined in black line) in the border of Chile and Bolivia. 1.3. Objectives of the report 1.3.1. General Objective To conduct a study of the geological evolution of the Silala River basin. 1.3.2. Specific Objectives a.To prepare a geological summary of the regional surroundings of the Silala Riverbasin. b.To produce a detailed geological map of the Silala River basin.c.To establish the recent evolutionary history of the geology in the Silala River basin.290
Annex VIII
3
1.4. Summary of the methodology
The study was carried out in the following manner: a) a review of all geological
information (maps and papers) available in the academic literature for this area, b)
mapping of the different lithological units and collection of rock samples (mainly for
dating analysis) in the field in the study area, and c) preparation of the report that
integrates the analytic data, field observations, petrographic descriptions and
stratigraphic relationships to develop a synthesis of the geological evolution of the
basin.
In detail the methodology consisted of:
 Preparation of a geological summary of the regional geological setting of the Silala
River basin: a broad geological context is presented, at a scale of 1:250,000, with a
brief summary of the evolution of the Upper Miocene – Pleistocene volcanic chain
(5.8 million years to 8.5 thousand years), in which the Silala River basin is located.
To this end, the most updated geological information of the area was compiled – both
information about the geology of the Chilean national territory, which is in part
unpublished, and of the border area of the Republic of Bolivia - using published
reports, papers and maps. For the Chilean territory, the Calama Map was used; scale
1:250,000 (Marinovic and Lahsen, 1984); the geological map of the Ascotán Salt
Flat- Inacaliri Hill area (SERNAGEOMIN, Sellés and Gardeweg, in edition; scale
1:100,000); and the geological map of the Paniri Grant (Polanco, 2012). There was
no fieldwork in Bolivian territory but it was possible to interpret available satellite
images with the support of the 1:250,000 sheets for the Volcán Ollagüe (Almendras
et al., 2002) and the Volcán Juriques and Cerro Zapaleri (Ríos et al., 1997), both
maps produced by the National Geology and Mining Service (SERGEOMIN) of
Bolivia.
 Creation of a detailed geological map of the Silala River basin. A geological survey,
at a detailed scale, of the immediate surroundings of the Silala River was carried out,
to determine the geological elements and units that form the area and the processes
that have shaped the present-day landscape. Two field trips were made (in September
and October 2016) to map the geology at a scale of 1:75.000. The lithostratigraphic
units were differentiated, and rock samples were collected for petrographic
characterization, geochronological determinations and geochemical analysis. The
trips included geological photo-interpretation of high resolution satellite images
supported by geological records, data and interpretations available in the academic
literature.
Annex VIII
291
4 The geological maps in this report (the cartographic composition and databases ofthe maps) were drawn by Mr. Cristián Faunes, geographer of the Department ofBasic Geology of SERNAGEOMIN.Establishment of the recent evolutionary history of the geology in the Silala Riverbasin. The characterization, identification and dating of the geological elements andunits mapped provided the evidence to establish a chronology of events andgeological processes that gave rise to these elements and units as well as theprocesses that shaped the landscape.Although efforts have been made to avoid jargon and the use of specialist terms, this is often not possible. For the purposes of clarity a glossary of geological terms has been included above. 1.5. Structure of the report Chapter 1. Introduction. This is an introductory chapter that includes a presentation of the location of the Silala River, the objectives of this report, a summary of the employed methodology and a brief description of the structure of this report. Chapter 2. Summary of the regional geology around the Silala River basin. In this chapter a summary of regional geology in which the study area lies is made to provide the geological context and environment of the Silala River basin. Chapter 3. Geological Summary of the Silala River basin. Chapter 4. Detailed stratigraphy. In this chapter the stratigraphy of the study area is described in detail, together with descriptions of the different geological units. Chapter 5. Structural geology. In this chapter the deformation and faults recognized in the Silala River basin are explained. 292
Annex VIII
5
Chapter 6. Evolution of the Silala River basin and ravine.
This chapter describes the different kinds of geological evidence observed in the Silala
River basin and ravine that are interpreted as relating to the fluvial genesis of the river
valley and ravine.
Chapter 7. Evidence of the formation and erosion of the Silala River ravine.
Chapter 8. Conclusions.
2. SUMMARY OF THE REGIONAL GEOLOGY AROUND THE SILALA
RIVER BASIN
At a regional scale, in the area identified in Fig. 2 it is possible to identify a series of
volcanic processes and events that have taken place in episodes over the course of the
last 12 Ma (view map “Simplified compilation of the geology in the Silala River area”,
Appendix F).
The oldest rocks that are exposed in the area are the sequences of ignimbrites whose
volume is considerable. These are chemically evolved rocks (dacites and rhyolites) that
filled depressions and valleys in the existing relief. The radiometric ages found in these
rocks indicate that at least two similar geological events took place in the area, their
ages being 10.71 Ma (Lower Río San Pedro Ignimbrite: Ar/Ar in biotite; Salisbury et
al., 2011) and 8.33 Ma (Sifón Ignimbrite: average Ar/Ar; Salisbury et al., 2011). These
dated rocks represent part of a series of voluminous and extensive volcanic events that
affected this part of the Highland region (Altiplano) (Salisbury et al., 2011). These
ignimbrites conform the older geological rocks in the Silala River basin area
(undifferentiated basement in the Fig. 4). Ignimbrites are deposited from explosive
volcanic eruptions. These volcanoes extrude a mix of volcanic gases, molten rock and
ash in a highly fluid pyroclastic flow. They flow under gravity at speeds of at least 100
km/hour and are very destructive.
From 6.2 Ma (Polanco, 2012) several stratovolcanoes formed on the ignimbrite bedrock
from underlying magma chambers. These had intermediate to more evolved
compositions (andesites and dacites) that have been identified in the north, forming a
volcanic chain spanning more than 30 km in a NW-SE direction (Cerro Lailai, Cerros
de Colana and Cerro Inacaliri o del Cajón [henceforth Cerro Inacaliri]: 5.4-5.8 Ma; KAnnex
VIII
293
6 Ar in total rock; Rivera et al., 2015) (Fig. 4.1), as well as in the south as an isolated volcano (Cerro Negro dated to 6.2 Ma; Polanco, 2012) (Fig. 2). The volcanic products associated with these eruptive centers are mainly lava flows and domes. Then, at 4.12 Ma (U-Pb in zircon) the Cabana Ignimbrite (Fig 4.2) was deposited in the Altiplano. This was a voluminous and highly evolved deposit and filled much of the pre-existing topography. Volcanic activity continued to develop several emission centers located NE and SW of Quebrada Negra (eg. Cerrito de Silala) and Inacaliri volcano (Cerro Inacaliri) (2.6 and 1.5 Ma, respectively, including Volcanic Sequences of the lower Pleistocene, referred to later) and, in the south, an intense and episodic volcanism began along the volcanic chain called Parini-Toconce that is over 20 km long and also is aligned in a NW-SE (N130°E) direction (Polanco et al., 2012), the most recent activity of which corresponds to a lava flow from the Paniri Volcano (150 ky; Polanco, 2012) (Fig. 2). Lastly, the most recent volcanic activity in the area corresponds to pyroclastic fall deposits (tephra) that resulted from an eruption of the Volcán San Pedro at the beginning of the Holocene (11.5 ky; Bertin and Amigo, 2015) which is immediately to the west of the Silala catchment area. 294
Annex VIII
7
Fig. 2. Regional geological synthesis in which the Silala River basin is located. Solid black line
corresponds to the catchment area of the Silala River.
Annex VIII
295
8 3.GEOLOGICAL SUMMARY OF THE SILALA RIVER BASINThe stratigraphy of the Silala River basin is summarized in Fig. 3. Fig. 3. Stratigraphic column of the Silala River basin. The blue line represents the morpho-stratigraphic position of the Silala River ravine with its associated river deposits (Holocene evolution of Silala River: T1 to T3; see detailed discussion of its evolution in Latorre and Frugone, 2017). The stratigraphic units found in the Silala River basin are described in detail later in Chapter 4. The interpretation of the rock types found in these units and their contact relationships have led to a robust conceptual understanding of the development of the Silala River (Río Silala) over geologic time. There follows a summary of this geological and geomorphological evolution of the Silala River and basin, which is further summarized in Fig. 4 below. A geological map of the area of the Silala River basin is shown in Fig. 5. The oldest positive relief in the study area was generated by volcanism that began early in the late Pliocene (ca 5.8 Ma) (Fig 4.1). Compositionally this volcanism is dacitic with subordinate andesite and these structures included the Cerro Silala, Cerros de Silaguala and Cerro Inacaliri (stage 1). This activity continued to the Late Pliocene (ca. 2.6 Ma). But during the Early Pliocene (ca. 4.12 Ma), in the Highland region (Altiplano) to the east of the present Silala River basin there was a massive volcanic eruption associated with the creation of a caldera: it was explosive, extended and voluminous, dominated by evolved magma (dacites and rhyodacites). This eruption generated a high-volume 296
Annex VIII
9
pyroclastic flow deposit that forms the Cabana Ignimbrite (Fig 4.2). Through and
overlying this ignimbrite a series of volcanic edifices, domes and lava domes began to
form (Fig 4.3).
This volcanism was of intermediate (or not very evolved) composition (mainly andesitic
rocks).
Subsequently, localized compressive tectonic activity occurred at the end of the
Pliocene or at the beginning of the Pleistocene. Through reverse faulting, aligned in a
NNW-SSE direction, and west vergent, the oldest documented units in the study area
(for example, the Cabana Ignimbrite) were exposed and tilted (Fig. 4.3). After this
compressive event – and during the Pleistocene – an alluvial episode occurred draining
towards the SW and SSW (Fig. 4.4). Evidence of this is found in the western margin
and central part of the study area, represented in geological record by fluvial and debris
flow deposits. This activity supports the existence of an active alluvial drainage course
constrained by the pre-existing volcanic edifices and lava flows. Similarly, at the source
springs of the Silala River, in what is now Bolivia, there is evidence of other alluvial
activity occurring before 1.48 Ma, when an andesitic lava flow truncated a drainage
system that flowed westwards. This is consistent in terms of time and genesis with the
alluvial activity that occurred at the western edge of the study area and indicates the
existence of a ‘Silala palaeo-valley’ that was active during the Lower Pleistocene. In
this same period, several lesser pyroclastic flows, having an andesitic composition and
coming from the east, the Silala Ignimbrite, were channeled through and then filled
most of the Silala palaeo-valley and other depressions formed by the then exiting
topography (Fig. 4.5). This effusive volcanism increased in activity during the late
Lower Pleistocene (Calabrian; ca. 1.6-1.5 Ma), through the formation of stratovolcanoes
(for example, Cerro Inacaliri, stage 2, Volcán Apagado) and the overflow of andesitic
and basaltic-andesite lava flows on top of the volcanic products from the Upper
Miocene - Pliocene and the Silala Ignimbrite, respectively (Fig. 4.6). There was an
important hiatus between the late Lower Pleistocene and the late Upper Pleistocene
(between 40-12 ky BP), in other words, there is a period without a geological record in
the study area. The eruption of lava flows and formation of volcanic edifices ceased and
the physiography gave way to glacial activity (Fig. 4.7). The latter is evident in the area
through narrow valleys that descend from close to the top of the volcanic edifices (for
example, Cerro Inacaliri) down to 4,450 m.a.s.l. Two main phases of deposition of
terminal moraines took place in the area. These are attributed to the period of time that
has been documented in this part of the Highland region (Altiplano) between 40-12 ky
BP (Dylan et al., 2005). Contemporaneously, in periglacial areas wide alluvial fans
developed, coalescing in the lower parts of the surrounding volcanic edifices. Around
11,650 BP, an eruptive pulse from the Volcán San Pedro (located 40 km west of the
Annex VIII
297
10 study area) generated recognizable deposits of pyroclastic fall tephra, which settled and covered a large part of the landscape. Lastly, towards the limit of the Upper Pleistocene - Holocene (post 11.6 ky BP and pre 8430-8350 years BP), the fluvial carving of the present Silala River ravine began (Fig. 4.8), giving rise to at least 4 levels of river terraces associated with erosion and accumulation of fluvial sediments. The penultimate of these levels (T3) shows the greatest development of lateral erosion, with associated aggradational deposits, followed by successive lesser abrasion terracing This positively indicates an initial state of fluvial stability in which lateral erosion predominates: thereafter came a period of rapid incision with prevalence of vertical erosion, in which the present configuration of the Silala River ravine was attained, close to 530 cal years BP (Latorre and Frugone, 2017). 298
Annex VIII
11
Fig. 4. Schematic diagram depicting the geological evolution of the Silala River basin, from the
Upper Miocene to Lower Holocene times (5.8 Ma – 8.5 ky BP).
Annex VIII
299
12 4.DETAILED STRATIGRAPHYThis chapter describes in detail the stratigraphy and lithology, from the oldest to the most recent geological units that form the Silala River basin (Fig. 5). The localities and geographical references mentioned in the text, as well as more details, are included in the geological map attached as folded sheet called “Geology of the Silala River area” (Appendix E). A 300
Annex VIII
13
Fig. 5. Geology of the Silala River basin. (A) geological map and profiles, (B) Legend of
geological units.
UPPER MIOCENE-PLIOCENE
Cabana Ignimbrite (Piic)
The Cabana Ignimbrite corresponds to a volcanic unit that was defined by Marinovic
and Lahsen (1984) and refers to a welded biotite and amphibole tuff that crops out in the
vicinity of the Cabana sulphur plant and that was attributed by Marinovic and Lahsen to
the Pleistocene period. The outcrops of this unit are distributed in a strip having an
approximate north-south alignment, from the area of the small Inacaliri impoundment in
the north to the area of the former Planta Cabana (Cabana sulphur plant) in the south. Its
base does not crop out but it underlies, in apparently conformable contact, dacitic lavas
from the unit called Volcanic Sequences from the Upper Miocene - Pliocene (MsPvd),
B
B
Annex VIII
301
14 which dip approximately 15° to the east (Fig. 6). In the lower course of the Silala River, the Cabana Ignimbrite is covered, in angular and erosive unconformity, by the Silala Ignimbrite (Pliis), from the Lower Pleistocene. The exposed thickness to the east of where the Cabana ravine and Silala River (Quebrada Cabana and Río Silala) converge is of about 350 m. Fig. 6. An outcrop of the Cabana Ignimbrite in small Inacaliri impoundment area (Silala River ravine; view to the south-southeast). The Cabana Ignimbrite (Piic) emerges covered by dacitic lava flows from the Upper Miocene - Pliocene (MsPvd), and both units underlie the Silala Ignimbrite – Pliis - (Lower Pleistocene age) in angular unconformity. The Cabana Ignimbrite is interpreted as outcropping in Bolivian territory, since rocks having a similar stratigraphic position are found there. These are covered by Silala Ignimbrite and andesitic lava flows dated at 1.48±0.02 Ma (K/Ar in biotite; Almendras et al., 2002). The Cabana Ignimbrite outcrops are pale yellowish brown to light reddish brown in colour, and are moderately to poorly welded. The lower part of the deposit contains brown and red dacite fragments, between 2-4 cm in diametre, and a few pumice clasts of up to 1 cm. At the roof of the unit there is a pumice-rich concentration (~15-20%), ranging between 0.5-15 cm in diametre and rounded. Usually the Cabana Ignimbrite appears fractured, in blocks ranging from decimetres to metres, as a result of a conjugate joint system. The fractures are paired in the directions N70°W/N60°E, and also in less frequent secondary conjugate pairing of N5/20°W. An age of 4.12±0.08 Ma (U/Pb in zircon; for analytic details see Appendix A, Table 3) has been obtained for this tuff. Plis MsPvdPiic 302
Annex VIII
15
Volcanic Sequences from the Upper Miocene - Pliocene (MsPvd) (ca. 5.8-2.6 Ma)
In this period a series of volcanic rocks, such as domes, lava domes, lava flows and
autoclastic breccia were emplaced. Their composition is mostly dacitic (Sellés and
Gardeweg, in edition) and they outcrop in the northern and southern edges of the Silala
River basin. These rocks conformably overlie the Cabana Ignimbrite and lie beneath, in
angular and erosive unconformity, the Silala Ignimbrite (PPlis) from the Lower
Pleistocene (Fig. 6). The outcrops of this unit are distributed on the south and east slope
of the Cerro Inacaliri, on the Cerrito de Silala and Cerros de Silaguala, and southwest of
the small Inacaliri water impoundment.
There are two radiometric ages for this unit. One of them – located immediately south
of the small Inacaliri impoundment – provided a younger date of 2.6±0.4 Ma (K/Ar in
groundmass; this work). The other age, located in Bolivian territory on the southwest
side of the Cerro Inacaliri, was dated to 5.84±0.09 Ma (K/Ar in biotite; Almendras et
al., 2002). These data indicate that the volcanism of this period begins in the Upper
Miocene (5.8 Ma) and extends to the Upper Pliocene (2.6 Ma).
Alluvial Deposits from the Upper Pliocene - Lower Pleistocene
(“Unmappable unit”)
This unit is too small scale to map (“unmappable unit”). It occurs as a thin sedimentary
clastic intercalation that has been deposited in erosive (and probably angular)
unconformity on top of the Cabana Ignimbrite, but is in conformable contact beneath
the Silala Ignimbrite (Fig. 7A). Outcrops of this unit are exposed 600 m south of the
Inacaliri Police Station, at a road crossing located on the west slope of the Silala River
ravine. In this area they reach a thickness of just 20 cm (Fig. 7B). Three main levels are
identified. From base to top they correspond to:
• 5 cm of coarse tuffaceous sandstone, with pebbles and volcanic ash matrix,
light grey, with fragments of quartz, biotite, rounded pumice measuring 3-5 mm and
lithics of reddish porphyritic dacites having sub-rounded edges and oblate shapes, 0.3-1
cm in diameter. Unconformable overlying (erosive contact) the top of the Cabana
Ignimbrite.
• 5.5 cm of fine pebbly conglomerate, dark grey, poorly sorted, with sub-angular
clasts measuring 0.3-2 cm in diameter, locally with poorly developed imbricated fabric,
which indicates palaeo-flows directed to the southwest and south-southwest (n = 4
measurements).
Annex VIII
303
16 •15 cm of coarse greyish brown sandstone, with 3-6 mm pebbles, gooddevelopment of low-angle trough crosses lamination, together with 5 cm thick erosive base (Fig. 7C). It has also been recognized in the subsurface (borehole CW-CO; see Appendix B), approximately 100 meters to SW of the international boundary and 51.4 m below the bed of the Silala River. It is in erosional contact with the Cabana Ignimbrite and conformable contact under the Silala Ignimbrite, and reaches a thickness of 12.6 m. It comprises, from bottom to top: 2.0 m of coarse sedimentary breccia, with angular fragments of black scoria up to 6 cm, brown pumice (10% in vol.), included in a silty-clayey sandy matrix (Fig. 7D). This is followed by 10.6 m of very coarse pebble sedimentary breccia, light brown, massive, non-laminated, matrix-supported, with angular fragments of 0.4-5 cm in diameter, consisting of black scoria (25% vol.), andesites, dacites, and monomineral fragments of plagioclase feldspar, biotite, amphibole and scarce quartz, included in a pebbly sandstone matrix that contains approximately 20-25% of fine interstitial silt and clay (Fig. 7E). No direct age is available for these deposits. However, given the stratigraphic position they occupy, their age is limited to the Lower Pleistocene (ca. 2 Ma). In the Inacaliri Police Station area, these facies represent alluvial sedimentation, with a flow directed towards the south-southwest and southwest They are interpreted as being transported by turbulent erosive flows, deposited as sub-aqueous sand dunes (Facies St) and as longitudinal bars (facies Gh) (Miall, 1996). On the other hand, these facies near to the international border represent proximal alluvial facies, where debris flow and mud flow deposits predominate, which have been transported by a gravity flow process (Boggs, 1995). 304
Annex VIII
17
Fig. 7. Clastic alluvial deposits from the Lower Pleistocene, located between the Cabana and
Silala Ignimbrites in the Inacaliri Police Station. (A) Show an erosive phase between both
pyroclastic events; (B) detailed section of the deposit showing a succession, from bottom to top,
of pebbly sandstone, fine dark grey conglomerate and coarse sandstones made of fine pebbles,
with low-angled cross lamination; (C) detail of upper sandstones with low-angled cross
lamination (scale in cm). Borehole samples in international border area, that show; (D) a
matrix-supported sedimentary breccia (debris flow deposits), included in a in a pebbly
sandstone matrix with interstitial silt-clay (E).
PLEISTOCENE
Silala Ignimbrite (Pliis) (Lower Pleistocene) (this report)
The Silala Ignimbrite is defined as a sequence of welded tuffs having an andesitic
composition that outcrops along the course of the Silala River. This unit was
deposited, in angular and erosive unconformity, on top of the Cabana Ignimbrite and
dacitic lava flows from the Volcanic Sequences from the Upper Miocene - Pliocene
Annex VIII
305
18 (MPsvd) (Fig. 6). In the area of the Inacaliri Police Station, the ignimbrite rests paraconformably on a thin deposit of previously described clastic sediments (Fig. 7A). It lies in paraconformity under lava flows from the unit described as Lower Pleistocene Volcanic Sequences (Pliv(a)). It should be noted that the name Silala Ignimbrite has also been used in Bolivia to refer to a welded tuff of 7.8 Ma (Urquidi-Barrau, 2005; 2012). The Silala Ignimbrite varies greatly in thickness, from approximately 15-20 m near the Inacaliri Police Station (Fig. 8) up to 85-90 m in the area close to the international border (locality of the core exploration borehole CW-CO). Fig. 8. Outcrop of Silala Ignimbrite (Pliis) exposed in the Silala River ravine near to the Inacaliri Police Station (Retén de Carabineros de Inacaliri). It is possible to identify the unconformable contact with the Cabana Ignimbrite (Piic) and at least three units of flow and/or cooling that constitute the Silala Ignimbrite (Pliis). This pyroclastic deposit is made of four flow units that together constitute a single cooling unit (Fig. 8). The lower flow unit, about 5 to 7 m thick, is characterised by its dark grey colour and a subvertical fracturing that defines blocks measurable in metres. It corresponds to a flow of scoria, comprising a poorly sorted tuffaceous breccia characterised by angular fragments of varying dimensions (10-90 cm in diameter), and made of strongly vesicular basaltic andesites and andesites, immersed in a matrix of ash, lithic fragments, and pumice measurable in centimetres. 306
Annex VIII
19
The upper flows, light reddish brown in colour, are of variable thickness between 7 and
10 m, and are characterised by open subvertical fractures that define blocks measurable
in metres (2-4 m). They consist of pumice and lithic tuffs, welded, having
approximately 20% vesicular pumice that is 1-30 cm in diameter, 7% andesitic scoria
fragments, and 10% reddish andesitic lithic fragments having a diameter of 1 to 6 cm.
The age of the Silala Ignimbrite is determined by its stratigraphic position, that is,
deposited unconformably on dacitic lava flows (MsPvd), dated at 2.6±0.4 Ma (K/Ar
groundmass, this work) and underlies andesitic lava flows (Pliv(a)) dated at 1.48±0.02
Ma (K/Ar in biotite; Almendras et al., 2002). The Silala Ignimbrite is thereby limited to
the Lower Pleistocene. It thins to the west indicating that this ignimbrite flowed from
east to west.
Volcanic Sequences from the Lower Pleistocene (Pliv (a))
This denomination encompasses a group of volcanic edifices and well preserved lava
flows having an andesitic composition and subordinate dacitic composition, which are
exposed in the central north and southeast of the study area. It includes the volcanic
cone of the Cerro Inacaliri (Fig. 9A), the mountain range of the Volcán Apagado and its
extension into Bolivian territory, and a vast andesitic lava flow that partially fills up the
depression where the springs of the Silala River and Orientales wetlands can be found
(Fig. 9B).
The volcanic edifices of this unit are located on the remains of lesser edifices and domes
from the earlier Upper Miocene - Pliocene (MPsvd) volcanics, as is clearly visible in the
middle part of the south slope of the Cerro Inacaliri (Fig. 8A). At the springs of the
Silala River (in Bolivian territory), there is a large lava flow on top of the Silala
Ignimbrite (relationship described by photo geology). The rocks of this unit are partly
covered at high elevation by glacial deposits (Plg) that are dispersed on the south slope
of the Cerro Inacaliri and to the WNW of the Volcán Apagado.
Lithologically this unit comprises andesitic lava, agglomerates and andesitic tuffs that
are reddish and black (Pliv(a)) (Sellés and Gardeweg, in edition). To north of Cerro
Inacaliri, there are mainly andesitic and dacitic lava flows and dacitic domes (Pliv)
(Sellés and Gardeweg, in edition).
There are two radiometric ages for this unit. One of them, located on the south slope of
the Cerro Inacaliri gave a value of 1.612±0.018 Ma (40Ar/36Ar groundmass; Sellés and
Gardeweg, in edition). In addition, the lava flow located at the Silala River source
springs (Orientales wetland) was dated to 1.48±0.02 Ma (K/Ar in biotite; Almendras et
Annex VIII
307
20 al., 2002). This information indicates that the eruptive phase was active during the Lower Pleistocene (Calabrian). It is important to note, from a morphological perspective, that the andesitic lava flow that partially fills the depression of the Silala River headwater cuts a palaeo-drainage system that appears to converge toward the springs of the Orientales wetland at the present headwaters of the Silala River (Fig. 10). Such relationship indicates that 1.48 Ma ago there already was a drainage system that flowed west, towards present day Chilean territory, through the ‘Silala palaeo-valley’. Fig. 9. Volcanic edifices and lava flows from the Lower Pleistocene. (A) Construction of the andesitic volcanic cone of the Inacaliri Hill (Cerro Inacaliri) (Pliv(a)) on remains of dacitic domes from the Upper Miocene - Pliocene (MsPvd) (view to the northwest); (B) Andesitic lava flow that partially fills the depression where the source springs of the Silala River (Orientales wetland), deposited on the Silala Ignimbrite (Pliis) in Bolivian territory (view to the northeast). 308
Annex VIII
21
Fig. 10. Andesitic lava flow dated to 1.48 Ma (Pliv(a)) that cuts or truncates a palaeo-drainage
system with flow to the west and that converges towards the headwater of the present Silala
River. This indicates the existence of ‘Silala palaeo-valley’ in the Lower Pleistocene. Piic:
Cabana Ignimbrite.
Annex VIII
309
22 Glacial Deposits (Plg) (Pleistocene) The deposits of glacial origin that surround the Silala River basin are confined to the south, southwestern, and southeastern slopes of the Cerro Inacaliri (Fig. 10), and to the northwest end of the Volcán Apagado ridge. In these areas, the glacial geomorphology comprises cirques and narrow glacial valleys. On the northern slope of the Silala River basin, the morainic deposits extend down to 4,400 meters above sea level, with glacial cirques located between 4,970-5,300 m.a.s.l. On the southern slope, the moraines extend down to 4,690 m.a.s.l., with glacial cirques located between 5,574-5,620 m.a.s.l. On the north slope of the basin, the preservation of two terminal morainic ridges (Fig. 11) can be seen. The lower ones are located between 4,400-4,450 m.a.s.l., while the highest frontal ridges are located between 4,550-4,750 m.a.s.l. The Glacial Deposits correspond to deposits that, morphologically, constitute well preserved terminal and lateral moraines that are associated with small glacial valleys extending 1.7-3 km in length by 0.36 – 0.7 km in width. The deposits are formed by blocks and rounded stones that range from meters to decimeters, and gravels with fragments ranging between decimeters to centimeters. They mesh laterally with the Alluvial Deposits from the Upper Pleistocene Pls(a) and are incised by alluvial stream beds from the Upper Pleistocene - Holocene (PlHa) and active river beds (Ha). There are no direct data that allow an accurate dating of these deposits but broadly speaking, they are attributed to the Pleistocene. A new publication of palaeoclimatology of the Andes at 23°S (Chajnantor Plateau, northern Chile), suggest there existed three main phases of moraine formation at regional level: ca. 25-40, 15-17 and 12-14 ky (Ward et al., 2015). In the study area it is unclear to which Ice Age the two identified terminal morainic ridges should be attributed but they will be bracketed by this wide range of between 40 and 12 ky BP. 310
Annex VIII
23
Fig. 11. Glacial deposits on the northern slope of the Silala River basin. Two main phases of ice
stabilization, represented by two terminal moraines ridges (green lines), are attributed to the
Annex VIII
311
24 Upper Pleistocene (approximately 40-12 ky BP). The photo shows the flat surface of alluvial deposits (Pls(a)), the reach of the glaciers and their non-genetic relationship with the size of the Silala River ravine. Alluvial Deposits from the Upper Pleistocene (Pls(a)) This unit comprises a group of blocks, rounded stones, gravels, sands and silts, which are unconsolidated and exposed chiefly on the north slope of the Silala River basin. These deposits mesh laterally with the Glacial Deposits (Plg) and are incised by the Alluvial Deposits from the Upper Pleistocene - Holocene (PlHa) and active lesser river beds. In addition, they are partly covered by Pyroclastic Fall Deposits PlH(pc) from the Upper Pleistocene (ca. ~11 ky AP). These are mostly constituted by rounded stone gravel, gravels and very coarse pebbles. They are unconsolidated, with an interstitial matrix of coarse to very coarse sand, with pebbles. The gravels demonstrate a clast-supported fabric that is moderately imbricated, poorly sorted (1-60 cm in diameter), with clasts having sub-rounded to sub-angular edges and sub-oblong shapes. The clasts are formed by coarse porphyritic dacites of biotite and hornblende (20 vol. %), andesites to basaltic andesites, black, fine porphyritic (30 vol. %) and black and reddish scoriaceous andesites (50 vol. %). Measurements of the palaeo-current direction indicate flows trending towards to S10-20° W and S20°E on the north slope of the Silala River, while on the south slope they have a N70-95°W direction. Since they mesh laterally with the Glacial Deposits, they are attributed to the Upper Pleistocene. Pyroclastic Fall Deposits (PlH(pc)) These are unconsolidated deposits, dark grey with thin strips of lighter colour, well stratified, located in the central and south-central parts of the study area. This unit is deposited on top of the Alluvial Deposits from the Upper Pleistocene Pls(a) and has been eroded by the later Alluvial Deposits from the Upper Pleistocene - Holocene (PlHa). They are characterised by presenting alternating levels of dark scoria and subordinated levels of light coloured pumice. The pyroclastic tephra fragments range between 0.5-4 cm and it is possible to identify pumice clasts of up to 6 cm (Bertin and Amigo, 2015). This fall deposit was dated by Payne (1998) to 11,650±140 years BP by 14C in peat (30 km to west of study area) (calibrated age in Bertin and Amigo, 2015). Its stratigraphic position suggests that it represents the last eruptive activity of the San Pedro volcano that had regional impact (Bertin and Amigo, 2015). 312
Annex VIII
25
PLEISTOCENE - HOLOCENE
Alluvial Deposits from the Upper Pleistocene - Holocene (PlHa)
This corresponds to unconsolidated deposits of rounded stones, gravels, sands and silts,
exposed in the central and southwest sectors of the Silala River basin. These deposits
cut off and partially cover the Alluvial Deposits from the Upper Pleistocene (Pls(a)) and
the Pyroclastic Fall Deposits (PlH(pc)). They have been eroded by alluvial systems
from the Holocene (Ha).
This unit is mainly formed by gravels made up of rounded stones and very coarse
pebbles. It is unconsolidated, with few intercalations of sands with pebbles. The clasts
comprise coarse porphyritic biotite and hornblende dacites, andesites and scoriaceous
andesites, black and red in colour. They were deposited chiefly as a result of
redeposition of the alluvial deposits from the Upper Pleistocene.
Fluvial Deposits from the Upper Pleistocene-Holocene
(“Unmappable unit)”
This unit is too small to map scale and is a thin sedimentary clastic deposit with a
restricted areal extent. It is unconsolidated and located at a high level close to the
convergence of the Quebrada Negra and the Río Silala (Fig. 12, map). The deposit
overlies the Silala Ignimbrite (erosive unconformity; Fig. 12, picture) and is partially
covered by black active aeolian sands. The thickness is estimated at about 1-1.5 m.
This lower section of this unit contains grey and brownish grey sands, coarse grained, in
part having fine pebbles, with cross lamination, followed by a layer of light brown
siltstone 15 cm thick, with parallel flat lamination and fine traces of roots. On top of
these, in erosive contact, is a 1 m thick layer of gravels made of medium pebbles to fine
rounded stones (1-40 cm in diameter), poorly sorted, sub-rounded to sub-angular edges,
to a lesser degree rounded in the coarser fractions. There is a clast-supported fabric,
with moderate development of imbrication. The clasts are essentially made up of
andesites, porphyritic dacites and welded tuffs. The measurements of palaeo-current
(from the imbrication observed) indicate flows with a southwest direction,
approximately parallel to the walls of the incised river bed.
Given that this unit is confined on either side by the Silala Ignimbrite ravine walls and is
located at a lower surface than the Alluvial Deposits from the Upper Pleistocene
(Pls(a)), it is attributed to the interval between the Upper Pleistocene and the Holocene.
Recent 14C datings made in fluvial deposits from residual terraces, located in this area
and topographically lower than the unit in question, have yielded an age of 8430-8350
years BP (Latorre and Frugone, 2017), an age that constitutes a minimum value for this
Annex VIII
313
26 unit. This means that this unit is likely to be of the late Upper Pleistocene or early Lower Pleistocene in the development of the Silala River and represent the precursor (highest terrace) to the carving out of the Silala River ravine. They, therefore, represent palaeo-deposits of the Silala River. Fig. 12. Location of the Fluvial Deposits from the Upper Pleistocene – Holocene (Hf), as precursor stage before the erosion of the present Silala River ravine. Map shows lateral arm of a palaeo-course of the Silala River (blue arrow), with associated clastic deposits. Pls(a): Alluvial Deposits, Pliis: Silala ignimbrite, Hf: Fluvial Deposits. Photo shows details of the fluvial incision in the Silala Ignimbrite and aggradational filling of clastic sediments; view to the southwest. 314
Annex VIII
27
HOLOCENE
Alluvial Deposits from the Holocene (Ha)
These are unconsolidated deposits that have a sporadic occurrence, but which are
mainly distributed in the southwest corner of the study area. Here they form a system of
alluvial river beds and small alluvial fans, in very bright colours, which come from the
hydrothermal alteration of the rocks found on the Cerro del León (10 km to the south of
Pampa del León). They dissect the Alluvial Deposits from the Pleistocene-Holocene
(PlHa) and are cut off by the fluvial deposits of the Río Inacaliri.
Fluvial Deposits from the Holocene (Hf)
These correspond to clastic deposits from the Silala and Inacaliri rivers. These deposits
erode all the aforementioned units. They comprise alternating gravels, sandy gravels,
grits, along with occasional levels of silts and clays (Hauser, 2004) and numerous
organic deposits, many of which have been dated (Latorre and Frugone, 2017).
Hauser states that the larger grain-size fractions, which are predominantly andesites and
dacites, are sub-rounded to sub-angular, unaltered, resistant, and immersed in an
abundant sandy matrix. The deposits are loose, with little compaction, therefore porous
and permeable (Hauser, 2004). They lack a well-developed soil and/or plant coverage,
which suggests that the deposits would be have been deposited by similar processes to
those of the recent past or present-day (Hauser, 2004).
5. STRUCTURAL GEOLOGY
In general, the sedimentary deposits and rocks exposed in the area of the Silala River
basin do not show signs of tectonic deformation. The observed inclines of bedding and
surfaces of volcanic deposits correspond to the primary depositional slope. The only
exception occurs in the contact between the Cabana Ignimbrite and the dacitic lava
flows from the unit called Volcanic Sequences from the Upper Miocene - Pliocene
(MsPvd). This junction is tilted at approximately 15° to the east. This tilt is attributed to
the existence of a reverse fault (inferred and covered), west vergent, with a NNW-SSE
direction, that runs along the east edge of the Pampa del León area, along which the
Quebrada Cabana is channeled, and to the location of springs in the Planta Cabana area.
In this report, this fault is called ‘Cabana Fault’. Its hanging block contains the Cabana
Ignimbrite in contact with the dacitic lava flows. East of this structure, it is also possible
Annex VIII
315
28 to infer the existence of a normal fault, contemporary (Fig. 13), which accommodates the shortening and lifting of the hanging block of the Cabana Fault that is located further west. The descent of the west block of this normal structure creates a depression that is later filled with the Silala Ignimbrite and the Pyroclastic Fall Deposits (PlH(pc)) (Fig. 14). Due to the contact relations of the units involved (Cabana fault affecting Volcanic Sequences from the Upper Miocene - Pliocene (MsPvd; ca. 2.6 Ma) and covered by Silala Ignimbrite), the activity of this structure is attributed to the Lower Pleistocene. Fig. 13. Hanging block of the Cabana Fault (reverse fault) that generates the tilted contact between the Cabana Ignimbrite (Piic) and the dacitic lava flows (MsPvd) and, contemporary, a normal fault that creates a depression that is subsequently filled by the Silala Ignimbrite (Pliis). 316
Annex VIII
29
Fig. 14. Structural model for a rigid layer that has undergone reverse faulting, contemporary
coupled with a normal fault that accommodates said movement (modified from Gordon and
Lewis, 1980).
6. EVOLUTION OF THE SILALA RIVER RAVINE
This Chapter describes the geological evidence observed in the Silala River basin and
ravine, interpreted as relating to the fluvial genesis of the river valley and ravine.
The complete course of the Silala River has a stepped shape in plan view, with 2.3-3 km
sections in a NE-SW direction, and 1.3-2 km sections in a WNW-ESE direction. Within
these sections, the course is relatively winding, with a typical V-shaped or asymmetric
section, a sloped side (< 45°) and another subvertical side, the development of which
relates to the location either on the inside or outside of bends in the river course
(Huggett, 2007) (Figs. 15A and 15B). The width of the ravine varies between
approximately 57 and 147 m. Its depth ranges between 20 to 45 m. On the slopes and
walls of the ravine it is possible to observe, at different heights, a series of features that
are typical of river erosion of a youthful or mountain stream.
In fact, the surface of a palaeo-channel of the Silala River is preserved in the upper part
of this ravine (Fig. 16A). It comprises a sub-vertical erosion escarpment about 2-3 m
high that represents a lateral bank of the river or erosive escarpment of the river bed
(Fig. 16B). Adjacent to this escarpment is a relatively smooth slightly concave surface
Annex VIII
317
30 which corresponds to the substratum or bottom of the palaeo-channel. At the bed level, this surface is rough, lacking in striation and polished surfaces. In it a series of circular to sub circular cavities are developed, measuring decimeters to meters in diameter, their depth ranging between 30-90 cm. and their inner walls show different abrasions. Such cavities generally originate due to turbulent currents or whirlpools that catch pebbles and rounded stones that, because of the circular movement, generate cavities that are commonly known as potholes or pan-holes (Richardson and Carling, 2005; Ortega, 2010). Potholes can also be identified in the walls of the escarpment and are partially coalescent. On the surface of this palaeo-channel it is possible to see circular potholes, open and incomplete, while in the escarpment of the palaeo-channel it is possible to observe closed, lateral, coalescent potholes (Fig. 17), as defined in the classification of Richardson and Carling (2005) (Fig. 18). 318
Annex VIII
31
Fig. 15. View along the Silala River ravine (A) with mixed profile, whether V-shaped or
asymmetric having sloped sides, corresponding to the inner course or side of the river, and
subvertical corresponding to the external course or side of the river (view to the east). (B)
Conceptual model of water flow in a winding river (taken from Huggett, 2007).
A
Cross-sections
Undercut
bank
Slip-off
bank
Gravel bank
Line of maximum velocity B
B
Slip-off
bank
Undercut
bank
Annex VIII
319
32 Fig. 16. Morphology of the highest terrace in the Silala River ravine. (A) Terrace with the greatest lateral development, associated to a palaeo-surface of a river bed (view to the southwest). (B) Detail of the terrace components: the bottom of the river bed, and the erosion escarpment (view to the southwest). Cerro Paniri Bottom surface of palaeo-channel Erosion escarpment of the palaeo-channel 320
Annex VIII
33
Fig. 17. Development of eroded formations of the river bed, carved out in the highest terrace of
the Silala River ravine. (A) Creation of closed, lateral potholes (orange arrows), coalescent
lateral potholes (blue arrow), and open potholes in the palaeo-channel bed (red arrows). (B)
Detail of a pot-hole located in the bottom of the palaeo-channel. (C) Detail of a closed, lateral
pot-hole in the walls of the escarpment; note the differential circular erosion towards the
bottom of the cavity, caused by a smaller lithic fragments in the Silala Ignimbrite (scale in cm).
(D) Present-day carving of potholes in the Kukdi River, India
(https://unexplored.lonelyplanet.in/discovery/entry/1331.html). (E) Genetic diagram showing
the formation of potholes in turbulent flows due to bed load
(http://thebritishgeographer.weebly.com/river-landforms.html).
Annex VIII
321
34 Fig. 18. Classification and typology of potholes of fluvial origin (redrawn from Richardson and Carling, 2005). Another of the eroded formations, typically of fluvial origin, that are preserved at different heights on the subvertical slopes of the Silala River ravine are the so-called cavettos (Richardson and Carling, 2005; Schaller et al., 2005; Ortega, 2010) (Fig. 19). These correspond to concave hanging shapes that are located in the vertical or subvertical face of the side of a river course, and that undermine that surface. They vary from deeply carved formations with a nearly horizontal roof to those having low relief. They present linear and winding trajectories, attached to the subvertical walls located to the external side of a river bend, whose vertical section looks like a semi-circular depression with a prominent upper lip or ledge (Richardson and Carling, 2005). Their dimensions vary greatly and can be approximately the width of the river bed. Often these vertical depressions tend to develop in strata or levels that tend to be less resistant to erosion (Richardson and Carling, 2005). They originated due to abrasion by the bed load and suspended sediment due to lateral undermining by turbulent currents forming whirlpools at the river bends (Richardson and Carling, 2005; Schaller et al., 2005; Ortega, 2010). In the cavettos observed in the walls of the Silala River ravine it is often possible to see irregular patches or impregnations of caliche (Fig. 19A). The caliche represents deposition of calcium carbonate directly from the water table at a stage when evaporation predominated over sediment transport and aggradation. The association of cavettos with caliche impregnations constitutes unequivocal evidence of a river environment related to the different stages of the formation of the Silala River ravine. 322
Annex VIII
35
Fig. 19. Concave hanging formations or cavettos due to river erosion. (A) Middle course of the
Silala River with creation of cavettos or alcoves (blue lines) on top of the present surface of the
river bed, associated with caliche impregnations (lime); north slope. (B) Final third of the Río
Silala (North of Inacaliri Police Station), south slope, with development of at least two well
C
B
A
cal
cal
cal
Annex VIII
323
36 preserved cavettos (blue lines), associated to closed lateral potholes and open bottom potholes (orange arrows). (C) Present-day example of the formation of cavettos in the Fairy Pool River, Scotland; in the box, explanatory genetic diagram (taken from Richardson and Carling, 2005). 7.EVIDENCE OF THE FORMATION AND EROSION OF THE SILALARIVER RAVINE The most obvious and active degradation processes that affect the ravine of the Silala River correspond to those of alluvial action, of mass wasting (Huggett, 2007), and to a lesser extent to wind erosion. Regarding the first, it is possible to identify waterfall erosion in small ephemeral tributaries, in subvertical walls, that generate narrow grooves or chute furrows (Fig. 20A). In gentler slopes, the smaller tributaries give rise to a talus or alluvial and/or colluvial fans that accumulate unconsolidated detritus material from blocks, gravels, and sands (Fig. 20B). The mass wasting process constitutes the biggest degradation effect in the ravine. It is characterized mainly by the falling of blocks, measurable in meters, which accumulate in colluvial talus (Fig. 20C). 324
Annex VIII
37
Fig. 20. Alluvial degradation effects in the Silala River ravine. (A) Erosion by chute furrow due
to waterfall. (B) Alluvial and colluvial fans. (C) Block falls.
Although wind activity is common all year round, the effects are very marginal
compared to those of fluvial activity discussed earlier. The eroded formations observed
are small pillars or arcs (Fig. 21A), associated with surfaces with small impact holes and
abrasive surfaces (Fig. 21B). In the surfaces that are more exposed inside the ravine, a
series of irregular abrasion-weathering structures develop (Fig 21C), like as tafoni and
alveoli (honeycomb) structure (Huggett, 2007) amongst which small formations carved
in high relief stand out. These are characterized by minute ridges and residual
promontories, and by numerous oval hollows that contain sand, revealing the erosive
agent (Fig. 21D). Such structures are normally interpreted as a combination of wind
erosion and chemical weathering (saline) in desert environments (Rodriguez-Navarro et
al., 1999; Huggett, 2007).
Annex VIII
325
38 Fig. 21. Marginal effects of wind erosion on the degradation of the Silala River ravine. (A) Pillars and hollows. (B) Sand grain impact surface, indicating the preferential wind direction. (C) Wear formations in high relief and hollows on surfaces having frontal to oblique exposure to wind. (D) Detail of an oval hollow with sand remaining inside it. 8.CONCLUSIONSThe geological study of the Silala River basin reported here has led to the following conclusions: 1.The geological context of the region where the Silala River basin developedcorresponds to an intra-volcanic arc, which began with the forming of volcanic edifices, lava flow emissions, pyroclastic flows, and falls of tephra, from the Upper Miocene up to the late Upper Pleistocene (e.g., ca. 5.8-1.5 Ma). 2.The first evidence of the existence of alluvial drainage associated with the Silalafluvial system appears in the Lower Pleistocene (2.6-1.48 Ma). Such evidence is provided by alluvial deposits transported towards the SW and SSW in the final course 326
Annex VIII
39
of a palaeo-valley, and at the Bolivian Orientales wetland springs of the Silala River,
where an andesitic lava flow dated at 1.48 Ma cut off a drainage system that flowed
towards the west. This was interrupted by the deposition of the Silala Ignimbrite which
probably filled the first valley developed, but then the fluvial processes continued after
the deposition of the Silala Ignimbrite.
3. The second stage in the evolution of the Silala River system, which has the nature
of a mountain river, took place in the late Upper Pleistocene - Lower Holocene (ca. 12-
8.5 ky BP). This phase began with a period of fluvial stability, represented by the
formation of a wide terrace (T3) (see Fig. 3) located in the highest part of the ravine, in
which it is possible to identify the surface of a palaeo-channel bed and palaeoescarpment
of a river bank erosion. Later, the fluvial conditions became unstable and
gave way to a period of rapid river incision characterized by the formation of narrow
abrasion terraces, locally having associated deposits (T2; T1) (see Fig. 3).
4. Diverse morphological evidence indicates that the Silala River ravine was carved
out due to fluvial activity. Firstly the morphology shows a V-shaped or rather
asymmetrical cross section, with an inclined slope (< 45°) and another subvertical to
vertical, which are related to the internal and external sides of bends in the course of a
river, respectively. Both in the palaeo-channel bed and in the erosion escarpments it is
possible to identify circular erosion structures called potholes that are typical of river
erosion. Furthermore, at different heights in the ravine it is possible to identify
elongated semi-circular depressions, carved in the walls of the river bed, called cavettos
which originate by lateral abrasive undermining by the bed load and suspended sand, in
the areas of the external banks of river bends.
5. The glacial action that took place in the study zone was not involved in the
carving of the Silala River ravine. The glacial valleys are confined exclusively to the
slopes of the highest volcanic edifices (Inacaliri and Apagado volcanoes). The
associated deposits (moraines) lie a considerable distance from the middle course and
headwaters of the river and they do not descend below 4,400 m.a.s.l.
6. The wind action in the basin is a marginal degradational agent of the Silala River
ravine. Mass wasting (block falls) and lateral alluvial activity of minor tributaries are
the main degradation agents in the Silala River ravine.
Annex VIII
327
40 ACKNOWLEDGEMENTS The authors of this Study express their sincere gratitude to Dr. Moyra Gardeweg (Aurum Consultores) for the valuable contribution of geological information on the geologic environment of the study area, through maps, scientific publications, and fruitful discussions about the regional geology of the volcanic area; to Mr. Efraín Olivares (DIFROL) for providing his vast geographic knowledge of the zone and for coordinating field visits; and also to the palaeo-ecologist Dr. Claudio Latorre (Pontificia Universidad Católica de Chile), for sharing the results of his study that provided relevant records for this paper. Our thanks go to our colleague, Dr. Carolina Gómez (Arcadis), for the efficient coordination of the work groups. A special thank you to the SERNAGEOMIN laboratory, particularly to the professionals and personnel of chemistry, geochronology and sample preparation (grinding and separation of minerals), who strove to diligently obtain the results of the analyses, guided by the geologist Mrs. Eugenia Fonseca. We express our gratitude to Messrs. Carlos Ramírez and Germain Rivera (ENEL) for providing preliminary geochronology data on the Cabana Ignimbrite. We also thank María Teresa Cortés, librarian of SERNAGEOMIN, for her valuable and timely contributions. 328
Annex VIII
41
REFERENCES
Almendras, A. O., Balderrama, Z. B., Menacho, L. M., Quezada, C. G., 2002. Mapa
geológico hoja Volcán Ollagüe, escala 1:250.000. Mapas Temáticos de Recursos
Minerales de Bolivia. SERGEOMIN, Bolivia.
Bertin, D. and Amigo, A., 2015. Peligros del volcán San Pedro, región de Antofagasta.
Carta Geológica de Chile, Serie Geología Ambiental, No 25, p., 1 mapa escala
1:50.000, Santiago.
Boggs, S., 1995. Principles of Sedimentology and Stratigraphy, fourth edition. Prentice-
Hall, New Jersey, 662 p.
Hauser, A., 2004. Marco morfológico, geológico, tectónico, hidrogeológico e
hidroquímico: morfogénesis, evolución y modalidades de aprovechamiento del sistema
hidrográfico compartido Chileno-Boliviano del Río Silala. Servicio Nacional de
Geología y Minería, Informe consolidado. Santiago. (Vol. 4, Annex II, Appendix A).
Huggett, R. J., 2007. Fundamental of Geomorphology. Routledge (Ed.), New York, 458
p.
Latorre, C. and Frugone, M., 2017. Holocene Sedimentary History of the Río Silala
(Antofagasta Region, Chile). (Vol. 5, Annex IV).
Marinovic, N., Lahsen, A., 1984. Hoja Calama, Región de Antofagasta. Servicio
Nacional de Geología y Minería, Carta Geológica de Chile 58, 140 p., 1 mapa escala
1:250.000. Santiago.
Miall, A.D., 1996. The Geology of Fluvial Deposits. Sedimentary Facies, Basin
Analysis, and Petroleum Geology. Springer-Verlag, Berlin Heidelberg, 582 p.
Ortega, J.A., 2010. Morfologías en los ríos en roca. Variaciones y tipologías. In: Ortega,
J. A. and Duran, J.J. (Eds.), Patrimonio Geológico: Los ríos en Roca de la Península
Ibérica. Publicaciones del Instituto Geológico y Minero de España, Serie Geología y
Geofísica, No. 4, Madrid, pp. 55-78.
Payne, D., 1998. Climatic implications of rock glaciers in the arid Western Cordillera of
the Central Andes. Glacial Geology and Geomorphology, rp03/1998, 17 p.
Polanco, E., 2012. Geología a escala 1:50.000 del área de la Cadena Volcánica Paniri-
Toconce, Provincia del Loa, Región de Antofagasta (Informe inédito de la Gerencia de
Exploraciones) Energía Andina S.A. 35 p. (Appendix D).
Polanco, E., Clavero J., Giavelli A., 2012. Geología de la Cadena Volcánica Paniri-
Toconce, Zona Volcánica Central, Altiplano de la Región de Antofagasta, Chile. In:
Annex VIII
329
42 Actas del XII Congreso Geológico de Chile, Sesión Temática No. 4, Volcanología y Geotermia. Antofagasta, 462-464. Richardson, K., Carling, P.A., 2005. A Typology of Sculpted Forms in Open Bedrock Channels. Special Paper-Geological Society of America 392, 108 pp. Ríos, H., Baldellón, E., Mobarec, R., Aparicio, H., 1997. Mapa Geológico Hojas Volcán Inacaliri y Cerro Zapaleri, escala 1:250.000. Mapas Temáticos de Recursos Minerales de Bolivia, SGM Serie II-MTB-15B. Sergeomin. Rivera, G., Morata, D., Ramírez, C., 2015. Evolución Vulcanológica y Tectónica del Área del Cordón Volcánico Cerro del Azufre – Cerro de Inacaliri y su Relación con el Sistema Geotérmico de Pampa Apacheta, II Región de Antofagasta, Chile, XIV Congreso Geológico de Chile, La Serena. Rodríguez-Navarro, C., Doehne, E., Sebastian, E., 1999. Origins of honeycomb weathering: the role of salts and wind. GSA Bulletin, 111(8), 1250–1255. http://bulletin.geoscienceworld.org/cgi/content/abstract/111/8/1250. Salisbury, M.J., Jicha, B., de Silva, S., Singer, B., Jiménez, N., Ort, M., 2011. 44Ar/39Ar Chronostratigraphy of Altiplano-Puna volcanic complex ignimbrites reveals the development of a major magmatic province. Geological Society of America Bulletin 123, 821–840. Sellés, D., Gardeweg, M., (in edition). Geología del área Ascotán-Cerro Inacaliri, Región de Antofagasta. Servicio Nacional de Geología y Minería, Carta Geológica de Chile, Serie Geología Básica, Informe de Avance III, Inédito, 1 mapa, escala 1:100.000. Santiago. (Appendix G). Schaller, M., Hovius, N., Willett, S., Ivy-Ochs, S., Sinal, H., Chen, M., 2005. Fluvial bedrock incision in the active mountain belt of Taiwan from in situ-produced cosmogenic nuclides. Earth Surf. Process. Landforms 30, 955–971. Urquidi-Barrau, F., 2005. Recursos Hídricos en la Frontera Boliviana-Chilena (Silala y Lauca). In Política Exterior en Materia de Recursos HÍdricos, Udapex-PNUD (Eds.), Bolivia, p. 225. Urquidi-Barrau, F., 2012. Los Recursos Hídricos en Bolivia. Un Punto de Vista Estratégico Sobre la Problemática de las Aguas Transfronterizas. In Jimenez, B. and Galizia, J. (Eds.). Diagnóstico del Agua en las Américas. Red Interamericana de Academias de Ciencias, Foro Consultivo Científico y Tecnológico, AC, México, 95 p. Ward, D., Cesta, J., Galewsky, J., Sagredo, E., 2015. Late Pleistocene glaciations of the arid subtropical Andes and new results from the Chajnantor Plateau, northern Chile Quaternary Science Reviews 128, 98-116. http://doi.org/10.1016/j.quascirev.2015.09.022. 330
Annex VIII
1
APPENDIX A
RADIOMETRIC DATING
Analytic Procedure and Instrumental Conditions of the SERNAGEOMIN
Geochronology lab
http://www.sernageomin.cl/laboratorio.php
The following were the analytic methods used for the geochronological dating carried
out during this Study:
K-Ar Method
The datings made by the K-Ar Method were carried out by Messrs. César Vásquez and
Marcelo Yañez, at the Geochronology Lab of the National Geology and Mining Service.
In the case of minerals, the material used is selected manually under binocular loupe
with reflected and transmitted light and has a purity close to 100%; while in the total
rock analyses, the material is sieved in a 60/80 mesh. In both cases, the material passes
through a splitter and the fractions are analysed by K and by Ar. The chemical analysis
of K was done in triplicate, in an atomic absorption spectrometer in emission mode,
with lithium internal standard. For the Ar analysis, the sample is placed in a
molybdenum crucible and suspended inside an ultrahigh vacuum Pyrex glass line,
which is heated to 300°C for a period of 13-15 hours to obtain an adequate level of
vacuum (10-8 Torrs). Later, the sample is melted in a radiofrequency induction furnace
and the gases are purified by zeolites, Cu, and Cu and Ti oxides. The Ar volume is
determined by isotopic dilution, with an 38Ar-enriched tracer; the readings of isotopic
ratios were done in a AE1 mass spectrometer model MS-10S.
The constants used correspond to those adopted in the International Geology Congress
No. 25 (1976), Sydney, Australia and subsequently published by Steiger and Jäger
(1977):
λ(40Kԑ) = 0.581 x 10-10 years-1, λ(40Kβ) = 4.962 x 10-10 years-1,
isotopic abundance 40K = 0.01167 atom%, atmospheric ratio 40Ar/36Ar = 295.5
40Ar/39Ar Method
For datings by the 40Ar/39Ar Method that were done at the Geochronology Lab of
SERNAGEOMIN, the following steps were carried out. The minerals are selected
Annex VIII Appendix A
331
2 manually under binocular loupe with reflected and transmitted light. They are placed on a high-purity aluminium disk together with a monitor grain of Fish Canyon tuff sanidine (28.03±0.10 Ma, Renne et al., 1994). Once the disk is completed with samples, it is sealed with an aluminium planchet having similar characteristics as the carrier plate and sent to the 5 MW Herald pool nuclear reactor, operated by the Comisión Chilena de Energía Nuclear (“CCHEN”, Chilean Nuclear Energy Commission). The samples are placed in a stable position inside the reactor (Position A-09), surrounded by a cadmium shield, and radiated for 24 consecutive hours. Once the samples are returned from the reactor, an analysis is conducted separately – by total fusion - of all the monitors that the disk contains and the J parameter value is determined for each of them. These values are entered into a statistics program, which allows the creation of a smoothed radiation plate that assigns its own J value to each of the samples from the disk (Pérez de Arce et al., 2003). The cooled samples are introduced onto a copper disk and covered with a transparent potassium bromide disk. The disk is introduced in a chamber joined to a UHV line that is covered by a ZnS window, which is permeable to the passing of a CO2 laser. The samples are analysed by successive heating with temperature increments, by variation in laser power that can reach a maximum of 30 W. This is carried out with an integrator lens that allows the even heating of a 6 x 6 millimetre plate. Every three steps, a blank sample from the extraction line is analysed, which allows making corrections for the subsequent steps. The noble gases are separated by cold finger trap at -133°C and getter ST101 operated at 2.2 A. Once purified, they are placed inside a MAP 215-50 high resolution mass spectrometer, with its detector system used in its electron multiplier configuration. The 36Ar, 37Ar, 38Ar, 39Ar and 40Ar isotopes are analysed in 10 cycles: the 36/40, 37/40, 38/40 and 39/40 ratios are calculated for time zero (time when gas was introduced in the spectrometer), to eliminate the effects of isotopic fractioning during the analysis. The baseline is analysed at the start and end of the process, for each step, and subtracted from the height of the peaks. The apparent age obtained for each heating step considers the corrections corresponding to the Ar isotopes associated to atmospheric argon and the argon from radiation with K, Ca and Cl (40Ar, 39Ar, 38Ar, 37Ar and 36Ar). The plateau is defined by the criteria of Fleck et al. (1977) which regards a plateau to be three or more consecutive steps containing 50% or more of the total 39ArK released, and the errors of these steps are overlapped to the level of 2 standard deviations of reliability. When determining a plateau age and an isochrone age in an analysis, the first one is preferred except when the data of the second one indicates excess argon 332
Annex VIII Appendix A
3
(40Ar/36Ar>295.5±0.5) and it has been defined properly (MSWD<3); in such cases, the
isochrone age is preferred.
Annex VIII Appendix A
333
4 Sample UTM (1) Lithology Method and Material Age (Ma ± 2σ) Reference Obs. N E Cabana Ignimbrite Piic RSP-14D 7.563.554 596.534 Bt and hb crystaltuff U-Pb ELA-ICP-MS zircon 4,12 ± 0,08 This work Volcanic sequences from the Upper Miocene - Pliocene MsPvd RSP-13D 7.563.561 596.648 Porphyritic dacite of bt and hb K/Ar in groundmass 2.6 ± 0.4 This work No inf. No inf. No inf. Dacite K/Ar in bt 5.84 ± 0.09 Almendras et al., 2002 Volcanic sequences from the Lower Pleistocene Pliv(a) No inf. No inf. No inf. Bt andesite K/Ar in bt 1.48 ± 0.02 Almendras et al.,2002 AL-197 7.566.976 594.878 Basaltic andesite of ol and px Ar/Ar in groundmass 1.612 ± 0.018 Sellés and Gardeweg, in edition Table 1. Radiometric ages. Sample Unit Material % K Ar rad. (nl/g) % Atm. Ar Age (Ma ± 2σ) RSP-13D MsPvd Groundmass 2.668 0.2687 85.4 2.6 ± 0.4 Table 2. Analytic K-Ar data from this work. 334
Annex VIII Appendix A
5
CORRECTED
RATIOS2
CORRECTED
AGES (Ma)
U
(ppm)1
Th
(ppm)1 Th/U
207Pb/206
Pb
±2
abs
207Pb/23
5U
±2
abs
206Pb/238
U
±2
abs
208Pb/232
Th
±2
abs Rho
206Pb/23
8U
±2

207Pb/23
5U
±2

207Pb/206
Pb ±2
Best
age
(Ma) ±2
Zircon_66_
B651-
RSP14 224 49 0,22 0,4400
0,0430
0,1076 0,0094 0,0019 0,0001 0,0090
0,0055
0,31 12,2 0,8 103,6 8,5 3960,0
110,
0 12,2 0,8
Zircon_67 2174 1226
0,563
94 0,156 0,019 0,0161 0,0041 0,000753
0,0000
81 0,00069
0,000
51
0,1812
6 4,85
0,5
2 16,2 4 2390 140 4,9 0,5
Zircon_68 348 94
0,270
11 0,456 0,035 0,1281 0,0089 0,001989
0,0000
92 0,008
0,0045
0,2662
6 12,81
0,5
9 122,1 8,2 4143 71 12,8 0,6
Zircon_69 715 173
0,241
96 0,325 0,024 0,0629 0,0043 0,001437
0,0000
65 0,0041
0,002
5
0,3035
4 9,26
0,4
2 61,9 4,1 3598 74 9,3 0,4
Zircon_70 322 104
0,322
98 0,479 0,043 0,188 0,023 0,00289
0,0002
3 0,011
0,007
7
0,8337
3 18,6 1,5 173 20 4220 100 18,6 1,5
Zircon_71 636 250
0,393
08 0,245 0,029 0,0444 0,0071 0,00128
0,0000
87 0,002
0,0014
0,2680
6 8,25
0,5
6 44,1 6,7 3180 120 8,3 0,6
Zircon_72 322 146
0,453
42 0,329 0,032 0,0855 0,0066 0,00175
0,0001
1 0,0036
0,0019
0,2079
8 11,26
0,6
9 83,1 6,2 3679 78 11,3 0,7
Zircon_73 418 162
0,387
56 0,388 0,03 0,0855 0,0063 0,00158
0,0000
84 0,0037
0,002
3
0,1025
5 10,17
0,5
4 83,2 5,8 3898 80 10,2 0,5
Zircon_74 2033 187
0,091
98 0,175 0,019 0,0185 0,0021 0,000771
0,0000
26 0,0034
0,0022
-
0,1470
4 4,97
0,1
7 18,6 2,1 2620 130 5,0 0,2
Zircon_75 676 115
0,170
0,249 0,024 0,0448 0,0039 0,001224
0,0000
0,004
0,002 0,1558
7,89
0,3
44,4 3,7 3258 80 7,9 0,4
Annex VIII Appendix A
335
6 12 56 4 9 6 Zircon_76 1692 520 0,30733 0,26 0,02 0,0332 0,0032 0,000901 0,000058 0,00195 0,00095 -0,19977 5,8 0,37 33,2 3,1 3219 63 5,8 0,4 Zircon_77 812 242 0,29803 0,235 0,018 0,0392 0,003 0,001271 0,000063 0,0023 0,0015 0,17833 8,19 0,41 39 2,9 3060 71 8,2 0,4 Zircon_78 865 850 0,98266 0,206 0,03 0,0298 0,0071 0,001059 0,000074 0,00071 0,00067 0,48878 6,83 0,47 29,7 6,8 2890 170 6,8 0,5 Zircon_79 482 179 0,37137 0,289 0,025 0,0544 0,0047 0,001338 0,000074 0,0025 0,0015 0,16352 8,62 0,48 53,7 4,5 3435 94 8,6 0,5 Zircon_80 513 141 0,27485 0,256 0,022 0,0465 0,0035 0,001336 0,000065 0,0026 0,0016 -0,11074 8,61 0,42 46,1 3,4 3247 94 8,6 0,4 Zircon_81 1299 357 0,27483 0,199 0,021 0,0256 0,0037 0,000987 0,000092 0,0016 0,0011 0,49704 6,36 0,59 25,6 3,8 2851 97 6,4 0,6 Zircon_82 317 136 0,42902 0,287 0,041 0,057 0,01 0,00149 0,0001 0,0027 0,002 -0,083259 9,6 0,65 56 9,3 3430 120 9,6 0,7 Zircon_83 344 110 0,31977 0,333 0,041 0,084 0,011 0,0018 0,00012 0,0045 0,0031 -0,11347 11,59 0,76 82 10 3710 130 11,6 0,8 Zircon_84 413 116 0,28087 0,444 0,035 0,143 0,019 0,00236 0,00015 0,0093 0,0068 0,60256 15,2 0,98 135 16 4044 90 15,2 1,0 Zircon_85 488 161 0,32992 0,383 0,037 0,0764 0,0072 0,001541 0,000087 0,0041 0,0026 -0,14085 9,93 0,56 74,7 6,9 3897 69 9,9 0,6 Zircon_86 3832 99 0,02584 0,1028 0,0083 0,0098 0,00084 0,000697 0,000018 0,0042 0,003 0,051579 4,49 0,11 9,9 0,85 1657 92 4,5 0,1 336
Annex VIII Appendix A
7
Zircon_87 519 198
0,381
5 0,373 0,036 0,0879 0,0091 0,00183
0,0001
2 0,0043
0,002
8
-
0,2093
1 11,79
0,7
8 85,4 8,4 3770 100 11,8 0,8
Zircon_88 486 175
0,360
08 0,31 0,027 0,0691 0,0071 0,001646
0,0000
91 0,0036
0,0022
0,0871
51 10,6
0,5
9 68,6 6,6 3572 99 10,6 0,6
Zircon_89 642 199
0,309
97 0,395 0,026 0,0989 0,0056 0,001798
0,0000
85 0,0052
0,0033
0,0806
53 11,58
0,5
5 95,7 5,2 3948 65 11,6 0,6
Zircon_90 654 249
0,380
73 0,275 0,024 0,0599 0,0056 0,001562
0,0000
85 0,0025
0,0018
0,5436
7 10,06
0,5
5 59 5,3 3380 100 10,1 0,6
Zircon_91 404 99
0,245
05 0,333 0,03 0,068 0,0045 0,001482
0,0000
85 0,0041
0,0025
0,0520
87 9,55
0,5
4 66,7 4,4 3638 70 9,6 0,5
Zircon_92 415 305
0,734
94 0,446 0,039 0,112 0,01 0,00177
0,0001
3 0,0028
0,0019
0,3313
3 11,43
0,8
1 107,9 9,4 4179 66 11,4 0,8
Zircon_93 545 337
0,618
35 0,337 0,036 0,0683 0,0078 0,00138
0,0000
8 0,002
0,001
6
0,0012
01 8,89
0,5
2 66,9 7,3 3726 95 8,9 0,5
Zircon_94 610 200
0,327
87 0,316 0,035 0,0577 0,0072 0,00134
0,0000
74 0,003
0,0018
0,2905
4 8,63
0,4
8 56,8 6,8 3580 150 8,6 0,5
Zircon_95 676 275
0,4068
0,372 0,028 0,0898 0,0074 0,00173
0,0000
68 0,004
0,0026
-
0,2122
9 11,14
0,4
4 87,2 6,7 3832 64 11,1 0,4
Zircon_96 426 182
0,427
23 0,495 0,037 0,147 0,017 0,00221
0,0001
5 0,0064
0,004
6
0,0974
95 14,24
0,9
9 140 15 4221 74 14,2 1,0
Zircon_97 628 212
0,337
58 0,331 0,025 0,0729 0,0046 0,001557
0,0000
6 0,0037
0,0021
0,1144
7 10,03
0,3
8 71,4 4,4 3611 70 10,0 0,4
Zircon_98 2563 249
0,097
15 0,194 0,02 0,0194 0,0023 0,000763
0,0000
29 0,0032
0,0022
-
0,3329
3 4,92
0,1
9 19,5 2,3 2750 110 4,9 0,2
Annex VIII Appendix A
337
8 Zircon_99 2452 970 0,3956 0,186 0,018 0,0199 0,003 0,000796 0,000062 0,0011 0,00069 -0,19214 5,13 0,4 20 2,9 2668 99 5,1 0,4 Zircon_100 490 300 0,61 0,3460 0,0280 0,0802 0,0061 0,0017 0,0001 0,0028 0,0020 0,18 10,7 0,4 78,2 5,7 3679,0 63,0 10,7 0,4 Table 3. Analytic U-Pb Data from this work. 338
Annex VIII Appendix A
1
APPENDIX B
PETROGRAPHY
Annex VIII Appendix B
339
2 Sample UTM(1) N E Descriptions Cabana Ignimbrite Piic RSP-14D 7.563.558 596.534 Tuff having a high content of crystals and ash, with 45-50% volume of phenocryst fragments. Of these fresh plagioclase feldspar make up 25-30%. Biotite makes up 5-7% (crystals up to 2.25 mm), amphibole up to 2-3% (crystals up to 0.7 mm) and both are oxidise. Quartz makes up 10% and has embayments and curved fractures and there are a few Fe-Ti oxides and clinopyroxene (<<1%, up to 0.22 mm), in a matrix of dark brown to yellowish brown glass having intersertal and spherulitic (smectite) texture that is typical of devitrification. Volcanic Sequences from the Upper Miocene - Pliocene MsPvd RSP-13D 7.563.561 596.648 Porphyritic dacites of biotite, amphibole and pyroxene, with abundant phenocrysts (30 vol. %) of plagioclase, biotite (5-7%, up to 2 mm), amphibole that are partially or fully oxidised (2-3%, up to 0.94 mm), quartz and Fe-Ti oxides, immersed in groundmass having an intersertal texture of colourless to yellowish interstitial glass with abundant aligned plagioclase microliths and euhedral microcrystals of fresh clinopyroxene (3-5%, 0.05-0.3 mm), as well as granules of Fe-Ti oxides and a few amphiboles. The biotites and amphiboles show oxidation and several of these phenocrysts are completely opaque. The quartz has textures of embayment and curved fractures. Pyroclastic Fall Deposits PlH(pc) RSP-12D 7.563.438 596.708 The pumice fragments are white and consist of vesicular pumice (30 vol. %) and fibrous pumice having abundant pale to dark brown glass with flow texture and granules of Fe-Ti oxides, and few biotite crystals (1 vol. %) of up to 0.75 mm. There are very few fragments of plagioclase phenocrysts and Fe-Ti oxides (<3 vol. %) are found. Silala Ignimbrite Pliis RSP-16D 7.563.146 595.801 Vesicular pumice (40 vol. %) with abundant fragments of phenocrysts (30 vol. %) of fresh plagioclase (20-25 vol. %), fresh clinopyroxene (5-7%) of up to 4 mm, with inclusions of opaque minerals, and Fe-Ti oxide as isolated crystals forming clusters, included in a vitreous mass that is yellowish brown to reddish in colour having an intersertal texture. Table 1. Petrographycs descriptions. 340
Annex VIII Appendix B
3
Geologic unit Sample Optic conditions Photomicrograph
Volcanic Sequences from the Upper Miocene
- Pliocene MsPvd
RSP 13 5x / plane-polarized light
Cabana Ignimbrite Piic RSP 14 5x / plane-polarized light
Cabana Ignimbrite Piic RSP 15 5x / cross-polarized light
Annex VIII Appendix B
341
4 Silala Ignimbrite Pliis RSP 16 5x / plane-polarized light Table 2. Slides petrography. 342
Annex VIII Appendix B
5
CORE DESCRIPTION
Unit Deep (m) Lithology From
(m) to (m) Description
Non-consolidated alluvial deposit
6.2
Coarse
subrounded
sand
0 2.1
Coarse subrounded sand, with subrounded pebbles of 0.5-3 cm in
diameter, with macro organic remains of plants and roots. Anhedral
silica clast to up 3.4 cm.
Boulder gravel 2.1 2.3
Boulder gravel to up 30 cm of diameters, with fragments of black and
porphyritic andesites, of fine grain, 20-15% in volume of crystal and
black andesitic scoria.
Polymictic
boulder gravel
2.3 6.2
Polymictic Boulder gravel of gray and porphyric biotite and amphibole
dacites, reddish very well welded tuff with black fiammes and andesites
of pyroxene of 5-15 cm of diameter.
Silala Ignimbrite
57.8
Welded lithic
tuff
6.2 8.0
Welded lithic tuff (20%) and pumice with pink scoria fragments to up
1,5 cm and pyroxene andesites in pinky fine ash matrix. Medium to
high and fast water absorption.
Welded lithic
tuff
8.0 13.2
Polymictic welded lithic tuff with angular andesite fragments, gray
porphyric dacite with plagioclase and pyroxene in groundmass, reddish
oxidated scoria (pseudo fiammes), some of them elongated and
flattened of pseudoplastic aspect and pumice in very coarse ash matrix
(1.5-2 mm). Greater degree of fracture. High porosity and
transmissivity.
Annex VIII Appendix B
343
6 Welded scoria breccia tuff 13.2 19.2 Welded scoria breccia tuff with fragments of oxidate scoria of 6.4-15 cm of diameter (40-50% in volume) in gray reddish coarse ash matrix. With oblique fracturing in sections of 10-70 cm. Greater degree of fracture. High porosity transmissivity. Welded breccia tuff 19.2 25.4 Pale reddish welded breccia tuff with scoria fragments of 1-5 cm (40% in volume) in medium ash matrix. Minor fractures each 70-100 cm. High porosity and transmissivity. Welded breccia tuff 25.4 33.4 Similar to previous section with matrix size variations from coarse to medium ash. High density of fractures. Breccia tuff with a lot scoria fragments 33.4 46.3 Breccia tuff with many elongated and flattened scoria fragments (0.5-4 cm) and others with minor deformation of pseudoplastic of black colour (10-15 cm) in coarse to medium ash well welded matrix, minor fractured. Rock with lower water porosity and transmissivity. Coarse breccia tuff 46.3 51.4 Coarse breccia tuff (up to 15 cm of diameter) of porphyric and vitreous scoria with 45% in volume of andesitic and dacitic lithics, scoria, 25% fine brown fibrous pumice in very coarse ash matrix weakly welded. High water absorption. RSP-20 sample to thin sections. Debris flow deposit 12.6 Coarse to medium sedimentary breccia 51.4 62.0 Brown to reddish coarse to medium sedimentary breccia non-laminated, supported matrix and disintegrable with matrix (70% in volume) of two granulometry: clay-silt (50% in volume) and very coarse and fine sand (20% in volume) with magnetic scoria fragments of subangular monominerals of plagioclase, biotite, silica scarce and amphibole with gravel (< 5 cm) of major angular clasts (30% in volume) of magnetic scoria lithic of brown to black, dark and porphyric dacitic lithics and porphyric pumice with crystals of plagioclase and amphibole. Moderate 344
Annex VIII Appendix B
7
water absorption. From 58 to 61.3 m have high fracture density. RSP-
19 sample to binocular loupe.
Coarse
sedimendary
breccia
62.0 64.0
Coarse sedimentary breccia with angular and black scoria to up 6 cm,
brown pumice (10% in volume) in matrix of coarse sand silty clay.
Cabana Ignimbrite
53.4
(mínimum)
Medium to
poorly welded
crystal tuff
64.0 75.6
Medium to poorly welded crystal tuff, massive, fine grain, of pale gray
color with biotite, amphibole, plagioclase an silica (35-40% in volume)
with white and fibrous pumice to up 2 cm. Slow water absorption.
Breccia tuff of
blocks
75.6 77.8
Breccia tuff of blocks with lithics fragments of 7-40 cm of diameter of
Gray and porphyric dacites of biotite and amphibole. Medium degree of
fracturing.
Medium
welded pumice
tuff
77.8 81.65
Medium welded pumice tuff of pinky grey color with pumice (60% in
volume) of 0.5-8 cm of diameter and crystal rich (pyroxene, amphibole,
biotite, silica and plagioclase) in ash and crystal matrix. Rock with low
porosity of water and mostly fractured.
Welded tuff of
crystals and
coarse ash
81.65 85.2
Welded tuff of crystal (40% in volume) and coarse ash with white
pumice (20% in volume) of 0.5-1.5 cm of diameter.
Annex VIII Appendix B
345
8 Fault gouge 85.2 85.25 Fault gouge in oblique fault with stretch marks. Gray tuff of crystals and coarse ash 85.25 97.9 Gray tuff of crystal (45% in volume) and coarse ash with fine pumice (0.3-1 cm) and degassing pipe at 85.9 m of 40 cm. Fractured rock between 85.9 and 97.9 m. Welded tuff of crystals and coarse ash 97.9 106.2 Welded tuff of crystals and coarse ash with dark gray pumice (1-6 cm of diameter and 5-10% in volume) with mafic minerals rich (mainly biotite and amphibole). Pumice tuff of crystal and ash 106.2 117.4 Tuff of crystal and ash with pumice (15% in volume and 1.6 cm of diameter) with negative relief (cavity). Medium porosity. Table 3. Core description of borehole CW-CO. 346
Annex VIII Appendix B
1
APPENDIX C
INTERNATIONAL CHRONOSTRATIGRAPHIC CHART
Annex VIII Appendix C
347
2 348
Annex VIII Appendix C
3
APPENDIX D
Polanco, E., 2012. Geología a escala 1:50.000 del área de la Cadena Volcánica Paniri-Toconce, Provincia del Loa, Región de Antofagasta. Energía Andina S.A. 35 pp. (Unpublished report)
(Original in Spanish and English translation)
Annex VIII Appendix D
349
Número correlativo de informe: XXX Jefe de Proyecto: Aldo Giavelli Área: Dirección de Geociencias Preparado por: Edmundo Polanco Fecha:24-8-12 Revisado por: Jorge Clavero Fecha: XX-X-12 Revisado por: Fecha: Revisado por: Fecha: GEOLOGÍA A ESCALA 1:50.000 DEL ÁREA DE LA CADENA VOLCÁNICA PANIRI-TOCONCE, PROVINCIA DEL LOA, REGIÓN DE ANTOFAGASTA 24 DE AGOSTO, 2012
350
Annex VIII Appendix D
Correlative Report Number: XXXHead of Project: Aldo GiavelliArea: Geoscience Management Prepared by: Edmundo PolancoRevised by: Jorge ClaveroRevised by:Revised by:Date: 24-8-12Date : XX-X-12Date :Date :GEOLOGY AT SCALE 1:50,000 OF THE PANIRI-TOCONCE VOLCANIC CHAIN AREA, LOA PROVINCE, REGION OF ANTOFAGASTA AUGUST 24, 2012
Annex VIII Appendix D
351
Geología 1:50.000 del área de la cadena…– Agosto, 2012 1 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 1Localización y objetivo El Proyecto Paniri ubicado en la Provincia de El Loa, Región de Antofagasta, comprende la concesión de exploración de energía geotérmica Paniri de 33 mil hectáreas (11 por 20 km) definidas por dos vértices en coordenadas UTM (Datum PSAD 1956) de N7.568.000 y E567.000 km y N7.557.000 y E597.000 km, respectivamente (Figura 1). Figura 1. Localización del área de concesión de exploración de energía geotérmica Paniri de EASA (rectángulo de color rojo). El área de concesión está delimitada por los ríos Loa y Salado por el oeste y sur, respectivamente, mientras que la parte norte es cortada de este a oeste por el Río San Pedro. Además, las quebradas de Cupo y de Paniri tienen una orientación prácticamente N-S localizadas al oeste y este del Complejo Volcánico Paniri (CVP), respectivamente. 1
352
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20121 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 1Location and Objective The Paniri Project located in the El Loa Province, Antofagasta Region, comprises the Paniri geothermal energy exploration concession spanning 33 thousand hectares (11 per 20 km) defined by two vertices in UTM coordinates (Datum PSAD 1956) of N7.568.000 and E567.000 km, and N7.557.000 and E597.000 km, respectively (Figure 1). Figure 1. Location of the Paniri geothermal energy exploration concession of EASA (in red rectangle)The concession area is delimited by the Loa and Salado Rivers on the west and south, respectively, while the northern part is cut from east to west by the San Pedro River. The Cupo and Paniri ravines basically have a N-S trend direction, and are located to the west and east of the Paniri Volcanic Complex (CVP by its acronym in Spanish), respectively.
Legend
Complejo volcánico = Volcanic complex
Domo = Dome
Ojos = Water hole, Springs
Retén = Police Station
Volcán = Volcano
1
Annex VIII Appendix D
353
Geología 1:50.000 del área de la cadena…– Agosto, 2012 2 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Por otro lado, los altos topográficos del sector corresponden a los volcanes San Pedro y San Pablo (6.145 y 6.092 m s.n.m., respectivamente) que se localizan inmediatamente al NW del área de concesión (Figura 1). Además, el área de interés es atravesada por una cadena volcánica alineada en dirección NW-SE (N130°E) que se extiende por más de 48 km desde los volcanes San Pedro al Toconce. En el área de interés los centros eruptivos forman parte de esta cadena de NW a SE son CVP, domo Chao, Complejo Volcánico del León-Lagunita y volcán Toconce (Figura 1) y se extiende por más de 20 km de longitud. Una característica relevante es la diferencia superior a 500 m del bloque norte con respecto al bloque sur de esta cadena volcánica evidenciando un fuerte control estructural de está alineación en el contexto del régimen de esfuerzos regionales. Por su parte, los centros poblados más importantes cercanos al área de concesión corresponden a Toconce, Paniri, Turi, Ayquina y Cupo que se localizan al sur de la misma a menos de 16 km de distancia. No obstante, los pueblos de Cupo, Paniri y Turi son los pueblos más cercanos al área del proyecto que se encuentran a 2,5; 6,1 y 15,5 km al sur del área, respectivamente. Además, el Retén de Inacaliri se localiza en el límite este del área de concesión y el caserío de Ojo de San Pedro (21 habitantes; INE, 2005) ubicado inmediatamente al norte del área de concesión (a una distancia de más de 3 km). Uno de los objetivos en el marco del Proyecto Paniri fue realizar el levantamiento de la geología de área de estudio a escala 1:50.000, además de un estudio estructural de detalle. La geología del basamento y el análisis y modelo estructural fueron adjudicados y elaborados por la Consultora Tehema. Mientras que, el levantamiento geológico de detalle a escala 1:50.000 del lineamiento NW-SE de los volcanes que constituyen la parte más elevada topográficamente de la zona de interés y que abarcan desde el CVP hasta el volcán Toconce, extendiéndose más hacia el SE del área de concesión, fue elaborado por EASA. El presente informe expone los resultados del trabajo desarrollado en el levantamiento de la geología de detalle del área de la concesión Paniri al este del volcán Cupo e incluye además, los centros volcánicos que se encuentran inmediatamente al sur y SE de esta concesión (Figura 1). De esta forma, incluye desde NW a SE los siguientes volcanes y complejos volcánicos: Complejo Volcánico Paniri, domo Chao, volcán Negro, Complejo Volcánico del León-Lagunita, volcán Toconce y domo Chillahuita (Figura 1). 2Metodología El levantamiento geológico se realizó en 4 etapas, la mayoría de ellas consecutivas: recopilación de antecedentes, geointerpretación, mapeo en terreno y recolección de muestras y, finalmente, procesamiento de la información y elaboración de mapa e informe. No obstante, las últimas dos fueron etapas metodológicas que se retroalimentaron dada la planificación de al menos dos campañas de terreno. 2.1Recopilación de antecedentes Esta etapa consistió en reunir y revisar todo el material disponible del sector de interés, en especial, la Hoja Calama (Lahsen y Marinovic, 1984) y una publicación de un estudio de detalle del domo Chao (de Silva et al., 1994). 2
354
Annex VIII Appendix D
2 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl In addition, the topographic highs in the region correspond to the San Pedro and San Pablo volcanoes (6,145 and 6,092 m above sea level, respectively). These are located immediately NW of the concession area (Figure 1). The area of interest is crossed by a volcanic chain running in a NW-SE direction (N130°E), which covers more than 48 km from the San Pedro volcano to the Toconce volcano. The eruptive centres that comprise this chain from NW to SE are: CVP, the Chao dome, the Del León-Lagunita Volcanic Complex and the Toconce volcano (Figure 1); the chain is more than 20 km long. A relevant characteristic is the distance of over 500 m between the northern and southern blocks of this volcanic chain which demonstrates a strong structural control of this ridge in relation to the regional stress regime.The most significant population centres near the concession area are Toconce, Paniri, Turi, Ayquina, and Cupo which are all located to the south, less than 16 km away. The towns of Cupo, Paniri, and Turi are the nearest to the project area and are 2.5 km, 6.1 km, and 15.5 km south of the area, respectively. The Inacaliri Police Station is at the east end of the concession area, and the hamlet of Ojo de San Pedro (21 inhabitants; INE, 2005) is located immediately north of the concession area (at a distance greater than 3 km).One of the objectives in the framework of the Paniri Project was to conduct a survey of the geology in the studied area at a scale of 1:50,000, as well as a detailed structural study. The bedrock geology and the structural model and analysis were prepared by the consulting firm Tehema. EASA prepared the detailed geological survey, at a scale 1:50,000, of the NW-SE lineament of the volcanoes constituting the highest topographical part of the area and spanning from the CVP to the Toconce volcano through the SE of the concession area.This Report presents the results of the work executed during the detailed geological survey of the Paniri concession area to the east of the Cupo volcano, and also includes the volcanic centres found immediately south and SE of this concession (Figure 1). Thus it encompasses from NW to SE the following volcanoes and volcanic complexes: Paniri Volcanic Complex, Chao dome, Negro volcano, Del León-Lagunita Volcanic Complex, Toconce volcano, and Chillahuita dome (Figure 1).2Methodology The geological survey was done in 4 stages, most of them consecutively: gathering of background information, geo-interpretation, on-site mapping and sample collection, and lastly, information processing and elaboration of map and report. However, the last two were methodological stages that received feedback from at least two field campaigns.2.1Gathering of Background Information This stage consisted of gathering and checking all the material available about the area of interest,especially the “Calama” Sheet (Lahsen and Marinovic, 1984) and a detailed study of the Chao dome (Silva et al., 1994). Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 2012
2
Annex VIII Appendix D
355
Geología 1:50.000 del área de la cadena…– Agosto, 2012 3 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 2.2Geointerpretación La etapa de geointerpretación correspondió a la definición de grandes unidades en la imagen satelital disponible (GeoEye) a escala 1:40.000 y en la imágenes disponibles en el GoogleEarth. 2.3Campañas de terreno y recolección de muestras Se realizaron dos campañas de terreno para la elaboración de la geología del sector de interés y la recolección de muestras, además de una salida de reconocimiento y accesibilidad (entre 20 al 23 de junio del año 2011) al área de estudio y que fue aprovechado para la recolección de muestras (2 para dataciones radiométricas y 3 para análisis químicos de roca total). Las salidas se realizaron entre los días 20 al 26 de agosto y 15 al 21 de octubre del año 2011 con el apoyo logístico de los Sres. Pedro Herrera (Externo) y el Sr. Julio Ríos, respectivamente. En estas salidas se revisaron 57 localidades (ver anexo de base de datos) y se recolectaron 43 muestras de roca (códigos PAE) para la elaboración de cortes transparentes, 35 muestras para análisis de roca total y 30 muestras para dataciones Ar/Ar (Figura 2). 2.4Procesamiento de la información y elaboración del mapa e informe La etapa de procesamiento de la información y elaboración del mapa geológico e informe se realizó exclusivamente en gabinete y consistió en: a)La selección de las muestras de roca para la realización de los análisis químicos de roca total (35) y lasdataciones radiométricas de Ar/Ar (20 que incluye dos muestras de códigos GAS recolectadas por Jorge Clavero en una campaña previa) (Figura 2), enviadas a ACME Laboratories y al Laboratorio del Servicio Nacional de Geología y Minería (SNGM), respectivamente. b)Se realizó mediante la utilización del microscopio petrográfico el análisis y la descripción de los cortestransparentes de las muestras recolectadas (43) que fueron elaborados por el Sr. Rubén Espinosa, además de, la selección y adquisición de microfotografías de cada una de las muestras. c)Se procesaron y diagramaron en distintos gráficos seleccionados los resultados de análisis químicos de lamuestras de roca separándolas por unidades. d)Se completó y actualizó la información en una base de datos (planilla Excel) para todas las localidadesrevisadas (57). e)Apoyado en la caracterización petrográfica y geoquímica de las muestras seleccionadas, así como, en losresultados de las dataciones radiométricas, se elaboró un modelo geológico conceptual y definieron las distintas unidades geológicas (ver mapa anexo a escala 1:50.000) que se describen a continuación. 3
356
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20123 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 2.2Geo-interpretation The geo-interpretation stage involved defining large units from the available satellite images (GeoEye) at scale 1:40,000 and in images available in GoogleEarth.2.3Field Campaigns and Sample Collection Two field campaigns were carried out to prepare the geological map and to collect samples, as well as a reconnaissance and accessibility outing (between June 20 - 23, 2011) to the study area, which was also used to collect samples (2 for radiometric dating and 3 for chemical analysis of the total rock). The field trips took place between August 20 - 26 and October 15 - 21 of 2011 with the logistical support of Messrs. Pedro Herrera (external) and Julio Ríos, respectively. During these trips, 57 sites were inspected (see database annex) and 43 rock samples were gathered (PAE Codes) to prepare transparent slices, 35 samples for total rock analysis, and 30 samples for Ar/Ar datings (Figure 2).2.4Information Processing and Preparation of the Map and Report The stage of information processing and elaboration of the geological map and report was exclusively a desk study. It consisted of:a)Selection of rock samples for total rock chemical analyses (35) and Ar/Ar radiometric dating (20, includingtwo GAS code samples collected by Jorge Clavero during a previous campaign) (Figure 2), which were sent to ACME Laboratories and to the Laboratory of the National Geology and Mining Service (SNGM in Spanish), respectively. b)With the help of a petrographic microscope, the samples were analysed and the transparent slides (formicroscopic examination) of the samples (43), which were prepared by Mr. Rubén Espinosa, were described and photomicrographs prepared.c)the results from the rock sample chemical analyses were processed and diagnostic and analyticaldiagrams were prepared.d)Information from all sites (57) was entered into a database (Excel spreadsheet).e)Based on the petrographic and geochemical characterisation of the selected samples, as well as on theresults of the radiometric datings, a conceptual geological model was developed and the different geological units defined (see annexed map at scale 1:50,000). These are described in the following sections.
3
Annex VIII Appendix D
357
Geología 1:50.000 del área de la cadena…– Agosto, 2012 4 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 2. Distribución de las localidades revisadas en el área de estudio (círculos de color anaranjado). Los círculos de color verde corresponden a muestras con edades, los círculos de color azul son muestras con análisis químicos y los de color rojo son los sólo con corte transparente. 4
358
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20124 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 2. Distribution of the sites inspected in the studied area (orange circles). The green circles correspond to samples with ages; the blue circles, to samples with chemical analysis; and red circles, to samples with transparent slide only.
4
Legend
Análisis químico = Chemical analysis
Corte transparente = Transparent slice
Datación = Dating
Observación = Observation
Annex VIII Appendix D
359
Geología 1:50.000 del área de la cadena…– Agosto, 2012 5 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3 Estratigrafía El levantamiento geológico de terreno se realizó sobre imágenes satelitales disponibles (imágenes satelitales GeoEye y Lansat), además del apoyo en gabinete de las imágenes del software libre de Google Earth en algunos sectores que requirieron más detalle. La escala del tiempo geológico utilizada corresponde a la publicada por la Geogical Society of America del año 2009. 3.1Mioceno La unidad más antigua en el área de estudio asignada al Mioceno corresponde a un único centro volcánico que se localiza inmediatamente al este del Complejo Volcánico Paniri. 3.1.1Volcán Negro (Mioceno) (Mvn) El volcán Negro (4.860 m s.n.m.; González, 1995) (Sierra Negra; de Silva et al., 1994) tiene un aspecto fuertemente erosionado (Figura 3), tiene unos 3,6-4,2 km de diámetro y está constituido por una serie de coladas de lava de composición andesítica silícea (63% en peso de SiO2) y de color negro, gris y rojizo que petrográficamente corresponden a andesitas porfíricas de piroxeno. Además, incluye una serie de depósitos piroclásticos de bloques y ceniza que se reconocen en los flancos de este centro eruptivo (Msnp). Figura 3. El volcán Negro fotografiado desde el oeste (en el flanco este del CVP) constituido por una serie de coladas de lava intensamente erosionadas. En primer plano se observa la parte norte del domo Chao. Antecedentes indican una edad Pliocena-Pleistocena (González, 1995), no obstante, se ha obtenido una edad de 6,223 ± 0,020 Ma (Ar/Ar en masa fundamental; este trabajo) (Tabla 1) en una roca hacia la base de la colada de lava de la parte SW del volcán, lo que permite asignarla al Mioceno Superior coincidente con el alto grado de erosión del edificio volcánico. 5
360
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20125 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3 Stratigraphy The field geological survey of the terrain was based on available satellite images (GeoEye and Lansat satellite images), as well as on desk studies of the images from the free software Google Earth in some areas that required more detail.The geological time scale used was the one published by the Geological Society of America from 2009.3.1Mioceno The oldest unit in the study area assigned to the Miocene corresponds to a single volcanic centre located immediately east of the Paniri Volcanic Complex.3.1.1Negro Volcano (Miocene) (Mvn) The Negro volcano (4,860 m.a.s.l.; González, 1995) (Sierra Negra; by Silva et al., 1994) has a strongly eroded appearance (Figure 3), is about 3.6-4.2 km in diameter, and is made up of a series of siliceous andesite lava flows (63% by weight of SiO2), black, grey, and reddish in colour, that petrographically correspond to porphyritic pyroxene andesite. It includes a series of pyroclastic deposits of blocks and ash that can be identified on the sides of this eruptive centre (Msnp). Figure 3. The Negro volcano photographed from the west (on the east side of the CVP), constituted by a series of highly eroded lava flows. In the foreground, the northern part of the Chao dome.Records indicate a date in the Pliocene - Pleistocene (González, 1995). However, an age of 6.223 ± 0.020 Ma (Ar/Ar in groundmass; this work) (Table 1) was obtained for a rock towards the bottom of the lava flow from the SW part of the volcano, which means it can be attributed to the Upper Miocene which would account for the high level of erosion of the volcanic edifice.
5
Annex VIII Appendix D
361
Geología 1:50.000 del área de la cadena…– Agosto, 2012 6 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Tabla 1. Dataciones radiométricas Ar/Ar realizadas en rocas pertenecientes a los distintos centros eruptivos de la cadena volcánica Paniri-Toconce. Las edades tienen unidades en ka, excepto los resultados con asterisco (*) que son reportados en Ma. CVP: Complejo Volcánico Paniri, CVLL: Complejo Volcánico del León-Lagunita, Anf: anfíbola y MF: masa fundamental. Muestra Centro eruptivo Unidad Época Material Edad Error GAS-16 domo Chillahuita Chillahuita Pleistoceno Superior Anf 370 110 PAE-30 domo Chao Chao 1 Pleistoceno Superior Anf 250 140 PAE-11 domo Chao Chao 2 Pleistoceno Superior Anf 240 70 PAE-1 domo Chao Chao 3 Pleistoceno Superior Anf 350 40 PAE-8 CVP CVP 3 Pleistoceno Inferior-Medio Anf 150 6 PAE-25 CVP CVP 3 Pleistoceno Inferior-Medio MF 163 3 PAE-2 CVP CVP 2 Pleistoceno Inferior-Medio Anf 260 100 PAE-3 CVP CVP 2 Pleistoceno Inferior-Medio RT 325 8 PAE-43 CVP CVP 1 Pleistoceno Inferior-Medio Anf 620 90 PAE-55 CVP CVP 1 Pleistoceno Inferior-Medio Anf 640 140 PAE-9 CVP CVP 1 Pleistoceno Inferior-Medio Anf 1,39* 0,28 PAE-48 CVLL CVLL 2 (del León) Pleistoceno Inferior-Medio MF 275 7 PAE-36 CVLL CVLL 2 (del León) Pleistoceno Inferior-Medio MF 367 18 PAE-44 CVLL CVLL 1 (Lagunita) Pleistoceno Inferior-Medio MF 629 7 PAE-37 CVLL CVLL 1 (Lagunita) Pleistoceno Inferior-Medio MF 664 12 PAE-16 CVLL CVLL 1 (Lagunita) Pleistoceno Inferior-Medio MF 1,054* 0,011 PAE-15 CVLL CVLL 1 (Lagunita) Pleistoceno Inferior-Medio MF 1,635* 0,014 PAE-42 volcán Toconce Toconce Pleistoceno Inferior MF 0,959* 0,005 PAE-12 volcán Negro Negro Mioceno Superior MF 6,223* 0,020 3.2Plioceno-Pleistoceno Inferior Al Plioceno-Pleistoceno Inferior sólo se ha asignado el volcán Linzor sólo por sus relaciones de contacto y el estado de preservación del edificio dado que carece de edades. 3.2.1Volcán Linzor (Plioceno-Pleistoceno Inferior) (PPlvl) Remanentes de bloques que constituyen una colada de lava muy escasamente preservada distribuida en la planicie al oeste del volcán Linzor (5.680 m s.n.m.) y al norte del domo Chillahuita. Observaciones realizadas desde el NE del volcán Toconce apoyada por lo observado en las imágenes satelitales indican que la fuente de estos remanentes de colada de lava corresponde al volcán Linzor. Se le asigna al Plioceno-Pleistoceno por su alto grado de erosión a pesar de que antecedentes indican que una edad pleistocena-holocena para este centro eruptivo (Marinovic y Lahsen, 1984). 6
362
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20126 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Table 1. Ar/Ar radiometric datings executed on rocks belonging to the different eruptive centres of the Paniri-Toconce volcanic chain. The ages are expressed in units of ky BP, except the results with an asterisk (*) that are reported in Ma. CVP: Paniri Volcanic Complex, CVLL: Del León-Lagunita Volcanic Complex, Amph: amphibole, and GM: groundmass.Sample Eruptive centreUnit Epoch Material Age Error GAS-16 Chillahuita dome Chillahuita Upper PleistoceneAmph 370 110 PAE-30 Chao domeChao 1 Amph 250 140 PAE-11 Chao domeChao 2 Amph 240 70 PAE-1 Chao domeChao 3 Upper PleistoceneUpper PleistoceneUpper PleistoceneAmph 350 40 PAE-8 CVP CVP 3 Amph 150 6 PAE-25 CVP CVP 3 GM 163 3 PAE-2 CVP CVP 2 Amph 260 100 PAE-3 CVP CVP 2 RT 325 8 PAE-43 CVP CVP 1 Amph 620 90 PAE-55 CVP CVP 1 Amph 640 140 PAE-9 CVP CVP 1 Lower – Middle PleistoceneLower – Middle PleistoceneLower – Middle Pleistocene Lower – Middle PleistoceneLower – Middle PleistoceneLower – Middle PleistoceneLower – Middle PleistoceneAmph 1,39* 0,28 PAE-48 CVLL CVLL 2 (del León) GM 275 7 PAE-36 CVLL CVLL 2 (del León) GM 367 18 PAE-44 CVLL CVLL 1 (Lagunita) GM 629 7 PAE-37 CVLL CVLL 1 (Lagunita) GM 664 12 PAE-16 CVLL CVLL 1 (Lagunita) GM 1,054* 0,011 PAE-15 CVLL CVLL 1 (Lagunita) Lower – Middle Pleistocene Lower – Middle PleistoceneLower – Middle PleistoceneLower – Middle PleistoceneLower – Middle PleistoceneLower – Middle PleistoceneGM 1,635* 0,014 PAE-42 Toconce volcanoToconce Lower PleistoceneGM0,959* 0,005 PAE-12 Negro volcano Negro Upper MioceneGM 6,223* 0,020 3.2Pliocene - Lower Pleistocene The Linzor volcano was attributed to the Pliocene - Lower Pleistocene because of its stratigraphic contacts and position ratios and the state of preservation of the edifice, as there are no ages available.3.2.1Linzor Volcano (Pliocene - Lower Pleistocene) (PPlvl) Block remnants that constitute a poorly preserved lava flow, are distributed on the plains west of the Linzor volcano (5,680 m.a.s.l.) and north of the Chillahuita dome. Observations made from the NE of the Toconce volcano supported by what was seen in the satellite images indicate that the source of these lava flow remnants was the Linzor volcano. These are assigned to the Pliocene - Pleistocene due to the high level of erosion despite records pointing to a Pleistocene – Holocene date for this eruptive centre (Marinovic & Lahsen, 1984).
6
Annex VIII Appendix D
363
Geología 1:50.000 del área de la cadena…– Agosto, 2012 7 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.3Pleistoceno Inferior El único centro eruptivo que ha sido asignado al Pleistoceno Inferior corresponde al volcán Toconce. 3.3.1Volcán Toconce (Pleistoceno Inferior) (Plvt) El volcán Toconce (5.408 m s.n.m.) es un estratovolcán fuertemente erosionado (Figura 4) que tiene profundos escarpes glaciares hacia el este y norte de su cima. La erosión ha permitido exponer el color rojizo y pardo amarillento de la parte central del edificio volcánico resultado de una intensa alteración hidrotermal (Figura 4). Está conformado por potentes coladas de lava porfíricas de composición dacítica y andesítica que se han construido en al menos dos etapas evolutivas. Se dispone de una edad de referencia para esta unidad de 1,1 ± 0,1 Ma (Baker y Francis, 1977; Marinovic y Lahsen, 1984) prácticamente coincidente con una edad de 0,959 ± 0,005 Ma (Ar/Ar en masa fundamental; este trabajo) (Tabla 1) obtenida en una singular colada de lava dacítica que se distribuye hacia el SW de su fuente y preserva sus estructuras de avance. La singularidad de la colada radica en que el frente y borde la misma presenta depósitos piroclásticos de bloque y ceniza. Figura 4. Volcán del Toconce mirado desde el norte. Destaca el color rojizo y pardo amarillento que evidencia la intensa alteración hidrotermal que ha afectado la parte central del edificio volcánico. 3.4Pleistoceno Inferior a Medio Los centros volcánicos asignados al Pleistoceno Inferior a Medio corresponden a los complejos volcánicos del León-Lagunitas y Paniri en base a nuevos antecedentes de dataciones radiométricos, coincidente con sus relaciones de contacto y el estado de preservación de os edificios y sus productos. 3.4.1Complejo Volcánico del León-Lagunita (Pleistoceno inferior-Medio) (Plimvll) El Complejo Volcánico del León-Lagunita (CVLL) corresponde a un centro de larga evolución dado que está conformado por los volcanes del León (5.753 m s.n.m.) y Lagunita (5.404 m s.n.m.) (Figura 5), localizado entre el volcán Toconce y el domo Chao. 7
364
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20127 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.3Lower Pleistocene The only eruptive centre attributed to the Lower Pleistocene is the Toconce volcano.3.3.1Toconce Volcano (Lower Pleistocene) (Plvt) The Toconce volcano (5,408 m.a.s.l.) is a highly eroded stratovolcano (Figure 4) that has deep glacial escarpments toward the east and north of its peak. The erosion has helped expose the yellowish brown and reddish colours of the central part of the volcanic edifice which is a result of intense hydrothermal alteration (Figure 4). It is formed by large porphyritic lavas of dacitic and andesitic composition that were created in at least two evolutionary stages. There is a reference age for this unit of 1.1 ± 0.1 Ma (Baker and Francis, 1977; Marinovic and Lahsen, 1984) that more or less matches an age of 0.959 ± 0.005 Ma (Ar/Ar in groundmass; this work) (Table 1) obtained from a single dacitic lava flow that spreads to the SW of its source and in which are preserved its flow structures. The singularity of the lava flow is indicated by the front and edges along which are block-and-ash pyroclastic deposits.Figure 4. The Toconce volcano seen from the north, highlighting yellowish brown and reddish colours that reveal an intense hydrothermal alteration which affected the central part of the volcanic edifice.3.4Lower to Middle Pleistocene The volcanic complexes of Del León-Lagunitas and Paniri are attributed to the Lower to Middle Pleistocene based on new radiometric dates that match their stratigraphic contact and the state of preservation of the edifices and their products.3.4.1Del León-Lagunita Volcanic Complex (Lower – Middle Pleistocene) (Plimvll) The Del León-Lagunita Volcanic Complex (CVLL) corresponds to a centre with a long evolution. It is formed by the volcanoes called Del León (5,753 m.a.s.l.) and Lagunita (5,404 m.a.s.l.) (Figure 5), located between the Toconce volcano and the Chao dome.
7
Annex VIII Appendix D
365
Geología 1:50.000 del área de la cadena…– Agosto, 2012 8 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl El volcán Lagunita (Pleistoceno Inferior) corresponde a un estratovolcán más antiguo fuertemente erosionado, de más de 7 km de diámetro en dirección E-W que presenta una intensa alteración hidrotermal representada por la variada coloración de este centro eruptivo (rojizo, gris azuloso, gris verdoso, pardo y blanco amarillento) constituido por coladas de lava andesítico y dacítico (59 y 65-66% en peso de SiO2) y depósitos piroclásticos de bloques y ceniza. Productos de este centro eruptivo cubre a rocas del volcán Negro. Las edades obtenidas para este centro eruptivo corresponden a 1,635 ± 0,014 y 1,054 ± 0,011 Ma (Ar/Ar en masa fundamental; este trabajo) (Tabla 1). Figura 5. Vista desde el norte del Complejo Volcánico del León-Lagunita. El volcán del León (Pleistoceno Medio) es un estratovolcán de más de 6 km de diámetro conformado principalmente por coladas de lava andesíticas y dacíticas (Figura 6) de piroxeno y anfíbola aunque también tiene depósitos piroclásticos de bloques y ceniza asociados. En su cima tiene dos cráteres semicirculares anidados de 260-270 m de diámetro. El estado de conservación de las coladas señala que la construcción de este centro eruptivo se ha formado en al menos 3 etapas evolutivas (Etapa I a III). La recolección y selección de 3 muestras de roca pertenecientes a este volcán han entregado edades que varían de 664 ± 12 a 275 ± 7 ka (Ar/Ar en masa fundamental; este trabajo) lo que permite asignar a este centro eruptivo al Pleistoceno Medio (Tabla 1). Figura 6. Vista del volcán del León mirado desde el flanco norte del volcán Toconce. 8
366
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20128 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl The Lagunita volcano (Lower Pleistocene) corresponds to a highly eroded older stratovolcano, which is over 7 km in diameter and elongated in a E-W direction. It has intense hydrothermal alteration represented by the varied colouration of the eruptive centre (reddish, bluish grey, greenish grey, brown, and yellowish white) and is formed by andesitic and dacitic lava flows (59 and 65-66% by weight of SiO2) and block-and-ash pyroclastic deposits. Products of this eruptive centre cover rocks of the Negro volcano and the ages obtained correspond to 1.635 ± 0.014 and 1.054 ± 0.011 Ma (Ar/Ar in groundmass; this work) (Table 1).Figure 5. View from the north of the Del León-Lagunita Volcanic Complex. The Del León volcano (Middle Pleistocene) is a stratovolcano with a diameter greater than 6 km, formed chiefly by andesitic and dacitic lava flows (Figure 6) of pyroxene and amphibole, though it also has associated block-and-ash pyroclastic deposits. At its peak there are two semi-circular nested craters 260-270 m in diameter. The state of preservation of the lava flows indicates that this eruptive centre was created in at least 3 evolutionary stages (Stage I - III). The gathering and selection of 3 rock samples belonging to this volcano provided ages ranging from 664 ± 12 to 275 ± 7 ky (Ar/Ar in groundmass; this work), so it is attributed to the Middle Pleistocene (Table 1).Figure 6. View of the Del León volcano from the north side of the Toconce volcano.
8
Annex VIII Appendix D
367
Geología 1:50.000 del área de la cadena…– Agosto, 2012 9 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.4.2Complejo Volcánico Paniri (Pleistoceno Inferior-Medio) (Plvp) El Complejo Volcánico Paniri (CVP) (5.945 m s.n.m.) corresponde a un voluminoso centro eruptivo (Figura 7a) de composición andesítica y dacítica (57 y 64-65% en peso de SiO2) de más de 11 por 15 km de diámetro (medidos en la direcciones E-W y N-S, respectivamente) que se ha construido en al menos 3 etapas evolutivas (Figura 7b) como señalan las variaciones en el estado de preservación, las relaciones de contacto y la variación litológica de las coladas de lava que constituyen el edificio volcánico. El CVP dispone de 7 edades cuyos resultados varían entre 1,39 ± 0,28 Ma y 150 ± 6 ka (Ar/Ar en masa fundamental; este trabajo) que permiten acotar bastante bien los tres estados de evolución: 1670-530, 360-160 y 160-144 ka. Figura 7. Complejo Volcánico Paniri. a) Forma cónica típica de estratovolcán visto desde el NW del camino al Retén de Inacaliri. b) Visto desde el este se distinguen dos edificios, el sur más antiguo (CVP I) y el norte y más elevado que cubre al anterior (CVP III). 3.5Pleistoceno Superior Las únicas unidades volcánicas que se han sido asignadas al Pleistoceno Superior corresponden a los domos Chao y Chillahuita, basado en los antecedentes aportados por Tierney et al. (2010) que indican nuevas edades Ar/Ar en biotita de 108 ± 6 a 190 ± 50 ka para los domos Chao, Chillahuita, Chanka, Chascón-Runtu Jarita, y Tocopuri. 9
368
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 20129 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.4.2Paniri Volcanic Complex (Lower – Middle Pleistocene) (Plvp) The Paniri Volcanic Complex (CVP) (5,945 m.a.s.l.) corresponds to a voluminous eruptive centre (Figure 7a) of andesitic and dacitic composition (57 and 64-65% by weight of SiO2), over 11 by 15 km in diameter (measured in E-W and N-S directions, respectively). It was created in at least 3 evolutionary stages (Figure 7b), as indicated by the variations in the state of preservation, geological contacts, and lithological variation of the lava flows that form the volcanic edifice.There are 7 ages for the CVP, ranging between 1.39 ± 0.28 Ma and 150 ± 6 ky (Ar/Ar in groundmass; this work) that help narrow fairly well the three stages of evolution to 1670-530, 360-160, and 160-144 ky.Figure 7. Paniri Volcanic Complex. a) Cone shape typical of stratovolcanoes, view from the NW from the road to the Inacaliri Police Station. b) View from the east. It is possible to see two edifices: the southern one is older (CVP I) and the northern one is higher and covers the previous one (CVP III).3.5 Upper Pleistocene The only volcanic units that have been attributed to the Upper Pleistocene are the Chao and Chillahuita domes, based on the records provided by Tierney et al. (2010) that indicate new ages of Ar/Ar in biotite of 108 ± 6 to 190 ± 50 ky for the domes called Chao, Chillahuita, Chanka, Chascón-Runtu Jarita, and Tocopuri.
9
Annex VIII Appendix D
369
Geología 1:50.000 del área de la cadena…– Agosto, 2012 10 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.5.1Domo Chao (Pleistoceno Superior) (Plvdc) El domo Chao (5.166 m s.n.m.) es una estructura volcánica de forma alongada de más de 12 km de largo por un ancho máximo de unos 7 km (Figura 1) que corresponde a uno de los domos torta (o lava domo) más grande del planeta con un volumen estimado de unos 26 km3 (de Silva et al., 1994). El domo torta se ha construido en a lo menos tres etapas evolutivas (Figura 8) e incluye unos importantes depósitos de flujos piroclásticos de pómez que se distribuyen en forma de abanico hacia el SW de su fuente (Etapa I). Petrográficamente son dacitas (66-67% en peso de SiO2) porfíricas de anfíbola, biotita y piroxeno aunque resultados previos señalan también la ocurrencia de riodacitas (69-70% en peso de SiO2; de Silva et al., 1994). Figura 8. Domo Chao es una lava domo que se ha construido en dos etapas (II y III, respectivamente). Al fondo alcanza aparecer el CVP. Edades K-Ar en biotita permiten asignar al domo Chao al Pleistoceno Superior (de Silva et al., 1994), a pesar de que los errores son del orden de magnitud de las edades. Además, nuevas dataciones Ar/Ar obtuvieron una edad de 111,2 ± 7,5 ka para este domo (Ar/Ar en biotita; Tierney et al., 2010). No obstante, en el presente trabajo se realizaron tres dataciones radiométricas que entregaron edades de 250 ± 140, 240 ± 70 y 350 ± 40 Ma (Ar/Ar en masa fundamental, este trabajo) para las etapas I, II y III, respectivamente (Tabla 1). 3.5.2Domo Chillahuita (Pleistoceno Superior) (Plvdch) El domo Chillahuita (4.680 m s.n.m.) es un domo torta (o lava domo) riodacítico (69% en peso de SiO2; de Silva et al., 1994) porfírico de anfíbola y biotita de unos 4 por 3,6 km (medidos en dirección N-S y E-W, respectivamente) y de un espesor variable entre 150 y 300 metros (Figura 9). Se ha estimado un volumen de 2,6 km3 y originalmente se ha asignado al Pleistoceno-Holoceno (Marinovic y Lahsen, 1984; 4 km3 según de Silva et al., 1994), dado el grado de preservación, similar al domo Chao. Se dispone de una edad de 107,8 ± 6,4 ka (Ar/Ar en biotita; Tierney et al., 2010) que restringen esta unidad al Pleistoceno Superior. Sin embargo, se dispone de una muestra de roca de la parte SE del domo que entrega una edad de 370 ± 110 ka (Ar/Ar en anfíbola, este trabajo) (Tabla 1). 10
370
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201210 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.5.1 Chao Dome (Upper Pleistocene) (Plvdc)The Chao dome (5,166 m.a.s.l.) is a volcanic structure having an elongated shape, which is more than 12 km long and has a maximum width of about 7 km (Figure 1). It is one of the largest cake-shaped domes (lava dome) on Earth, having an estimated volume of approximately 26 km3 (from Silva et al., 1994). The lava dome was formed in at least three evolutionary stages (Figure 8) and includes some significant pumice pyroclastic flow deposits that fan out toward the SW of its source (Stage I). In petrographic terms, these are porphyritic dacites (66-67% by weight of SiO2) of amphibole, biotite, and pyroxene, although previous results also indicate the occurrence of rhyodacites (69-70% by weight of SiO2; from Silva et al., 1994).Figure 8. Chao dome is a lava dome that was formed in two stages (II and III, respectively). In the back, the CVP.K-Ar dating of biotite helps attribute the Chao dome to the Upper Pleistocene (from Silva et al., 1994), despite errors in connection with the ages. A new Ar/Ar date yielded an age of 111.2 ± 7.5 ky (Ar/Ar in biotite; Tierney et al., 2010), however, in this work, three new dates were obtained and these provided ages of 250 ± 140, 240 ± 70, and 350 ± 40 Ma (Ar/Ar in groundmass, this work) for Stages I, II, and III, respectively (Table 1).3.5.2 Chillahuita Dome (Upper Pleistocene) (Plvdch)The Chillahuita dome (4,680 m.a.s.l.) is a porphyritic rhyodacite cake-shaped dome (69% by weight of SiO2; from Silva et al., 1994) of amphibole and biotite. It is about 4 by 3.6 km (measured in a N-S and E-W direction, respectively) and has a width varying between 150 - 300 metres (Figure 9). A volume of 2.6 km3 was estimated, and the dome was initially attributed to the Pleistocene - Holocene (Marinovic & Lahsen, 1984; 4 km3 according to Silva et al., 1994), given the level of preservation, similar to the Chao dome.There is an age of 107.8 ± 6.4 ky (Ar/Ar in biotite; Tierney et al., 2010) that limits this unit to the Upper Pleistocene. However, a rock sample from the SE part of the dome that yields a dating of 370 ± 110 ky (Ar/Ar in amphibole, this work) (Table 1).
Legend
Domo torta = Cake-shaped dome Etapa = Stage
10
Annex VIII Appendix D
371
Geología 1:50.000 del área de la cadena…– Agosto, 2012 11 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.6Pleistoceno Superior-Holoceno 3.6.1Depósito Aluviales Antiguos (PlHa) Los Depósitos Aluviales Antiguos corresponden a acumulaciones no consolidadas que comúnmente tienen forma lobulares y son caóticas, polimícticas, subhorizontales y pobremente seleccionadas con bloques angulosos a subredondeados (< 30 cm de diámetro) en una matriz de ceniza alterada. Figura 9. Vista desde el norte del domo Chillahuita. Es posible observar que tiene un espesor variable que es superior hacia el borde oeste (300 m aproximadamente). 3.6.2Depósitos Glaciales (Pleistoceno-Holoceno) (PlHg) Los Depósitos Glaciales son acumulaciones de material volcánico detrítico pobremente seleccionadas y caóticas de morfologías alargadas que nacen a los pies de escarpes glaciares y terminan en forma de lengua. Comúnmente se reconocen en los flancos sur de los edificios volcánicos. 3.6.3Depósitos Aluvio-Coluviales Antiguos (Pleistoceno-Holoceno) (PlHac) Los Depósitos Aluvio-Coluviales Antiguos corresponden a gravas, arenas y limos no consolidadas comúnmente polimícticas donde dominan los clastos angulosos a subangulosos que se distribuyen en la partes bajas de los valles y las quebradas y las laderas de los mismos. Es una mezcla de material detrítico transportado por gravedad y por el agua proveniente de las lluvias estacionales. 3.6.4Depósitos Piroclásticos de Bloques y Ceniza (Pleistoceno-Holoceno) (PlHpbc) Los Depósitos Piroclásticos de Bloques y Ceniza corresponden a depósitos antiguos de morfología suavisada, escaso alcance y cubiertos por depósitos aluviales y aluvio-coluviales. Son generados principalmente por el colapso de domos o lavas domo y se caracterizan por ser prácticamente monomícticos, constituidos por bloques de material juvenil denso inmersos en una matriz de ceniza que preservan algunos bloques con fracturamiento prismático. 1111
372
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201211 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.6Pleistoceno Superior-Holoceno 3.6.1 Old Alluvial Deposits (PlHa)The Old Alluvial Deposits are unconsolidated accumulations that usually have a lobular shape and are chaotic, polymictic, sub-horizontal, and poorly sorted with angular to sub-rounded blocks (< 30 cm in diameter) within a matrix of altered ash.Figure 9. View from the north of the Chillahuita dome. A variable thickness that is greater toward the west edge can be seen (300 m approximately).3.6.2 Glacial Deposits (Pleistocene - Holocene) (PlHg)The Glacial Deposits are accumulations of detrital volcanic material that are poorly sorted, chaotic, and having elongated morphologies, which originate at the foot of glacial escarpments and end up being tongue-shaped. These are usually identified on the south sides of volcanic edifices.3.6.3 Old Alluvio-Colluvial Deposits (Pleistocene - Holocene) (PlHac)The Old Alluvio-Colluvial Deposits correspond to unconsolidated gravels, sands, and silts, usually polymictic, dominated by angular to sub-angular clasts that are distributed in the lower parts and slopes of the valleys and ravines. It is a mixture of detrital material transported by gravity and by the water from seasonal rains.3.6.4 Block-and-Ash Pyroclastic Deposits (Pleistocene - Holocene) (PlHpbc)The Block-and-Ash Pyroclastic Deposits correspond to old deposits with smoothed morphology, limitted extent, and covered by alluvial and alluvio-colluvial deposits. They are mainly generated by the collapse of domes or lava domes, and are characterised by being practically monomictic, made up of blocks of dense juvenile material immersed in a matrix of ash that preserves some blocks with prismatic fracturing.
11
Annex VIII Appendix D
373
Geología 1:50.000 del área de la cadena…– Agosto, 2012 12 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.6.5Depósito Piroclástico de Ceniza (Pleistoceno Superior-Holoceno) (PlsHp) El Depósito Piroclástico de Ceniza corresponde a acumulaciones de color blanco y tamaño ceniza que se encuentran adosadas a los flancos de varios centros eruptivos. Comúnmente se encuentra formando dunas eólicas cuya distribución, distancia y relación estratigráfica permiten suponer que su fuente correspondería al volcán San Pedro. Originalmente consistiría un depósito piroclástico de caída de ceniza que ha sido retrabajado. 3.7Holoceno 3.7.1Depósitos Aluviales Recientes (Ha) Los Depósitos Aluviales Recientes están constituidos por acumulaciones no consolidadas que se distribuyen en los flancos de los volcanes, en las quebradas y depresiones que bajan de los edificios volcánicos. Son pobremente seleccionados con fragmentos dominantemente subangulos y angulosos de diámetro inferior a 100 cm. Los clastos mayores tienden a estar hacia la parte superior del depósito. 3.7.2Depósitos Coluviales (Holoceno) (Hc) Los Depósitos Coluviales corresponden a acumulaciones de gravas, arenas y limos no consolidadas comúnmente polimícticas, formadas dominantemente por clastos angulosos a subangulosos que se distribuyen en las laderas de los edificios volcánicos y han sido transportadas sólo por gravedad. 3.7.3Depósitos Aluvio-Coluviales Recientes (Holoceno) (Hac) Los Depósitos Aluvio-Coluviales Recientes corresponden a gravas, arenas y limos no consolidadas comúnmente polimícticas donde dominan los clastos angulosos a subangulosos que se distribuyen en las quebradas del área de estudio. 12
374
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201212 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 3.6.5 Pyroclastic Deposits of Ash (Upper Pleistocene - Holocene) (PlsHp)The Pyroclastic Deposits of Ash correspond to white accumulations of ash sized particles that are attached to the sides of several eruptive centres. They are typically found forming wind blown dunes whose distribution, distance, and stratigraphic position lead to the assumption that their source would be likely to be the San Pedro volcano. They are an ash fall pyroclastic deposit that has been re-deposited.3.7 Holocene3.7.1 Recent Alluvial Deposits (Ha)The Recent Alluvial Deposits comprise unconsolidated accumulations that are distributed on the sides of volcanoes, in ravines, and depressions that run down the slopes of volcanic edifices. These are poorly sorted, with predominantly sub-angular and angular fragments with a diameter smaller than 100 cm. The larger clasts tend to be found toward the upper part of the deposit.3.7.2 Colluvial Deposits (Holocene) (Hc)The Colluvial Deposits correspond to unconsolidated accumulations of gravels, sands, and silts, normally polymictic, predominantly formed by angular to sub-angular clasts that are distributed on the slopes of volcanic edifices and have been transported by gravity alone.3.7.3 Recent Alluvio-Colluvial Deposits (Holocene) (Hac)The Recent Alluvio-Colluvial Deposits correspond to unconsolidated gravels, sands, and silts, normally polymictic, with mainly angular to sub-angular clasts that are distributed in the ravines of the studied area.
12
Annex VIII Appendix D
375
Geología 1:50.000 del área de la cadena…– Agosto, 2012 13 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 4Petrografía Se seleccionaron y recolectaron 41 muestras de roca durante las campañas de terreno para la realización de cortes transparentes, los cuales fueron elaborados por el Sr. Rubén Espinosa. Además, se disponía de dos cortes transparentes (muestras GAS-10 y GAS-16) recolectadas en el marco del Proyecto Generativo, una de ellos del domo Chillahuita. La descripción y análisis al microscopio petrográfico de los cortes transparentes tienen por objeto reconocer los minerales y las texturas presentes con el objeto de dar ideas de los procesos que afectaron a los magmas en profundidad. Pero desde el punto de vista más práctico, permite identificar las muestras de roca que son factibles de datar y permiten complementar la definición de las unidades geológicas. Los resultados de la petrografía de las rocas del área de estudio fueron divididos por litología con la idea de resaltar las características comunes tanto en mineralogía como las texturas presentes. De esta forma, se reconocen cuatro grupos bien definidos: Lavas domo y pómez, coladas de lava dacíticas, andesitas silíceas, coladas de lava andesíticas y, finalmente, las inclusiones máficas. 4.1Lavas domo y pómez Las muestras de roca más evolucionadas del área de estudio, dacitas y riodacitas de anfíbola y biotita (Figura 10a, b, c y c), corresponden a las lavas domo (o domos torta) pertenecientes a las etapas 2 y 3 del Domo Chao (muestras PAE-14 y PAE-1), al Domo Chillahuita (muestra GAS-10) y a una lava domo de la primera etapa de construcción del Complejo Volcánico del León-Lagunita (muestra PAE-16). Además de, las pómez vesiculares y densas del Domo Chao (muestras PAE-30 y PAE-31, respectivamente) (Figura 10c). La asociación mineral de estas rocas está constituida por plagioclasa-anfíbola-biotita±cuarzo±piroxeno-óxidos de Fe-Ti. Los domos torta son dacitas de textura porfíricas de grano grueso conformadas por fenocristales de anfíbola y biotita inmersas en vidrio incoloro a color amarillento con textura perlítica (Figura 10a, b y d) y algunas bandas de textura fluidal. En algunos casos, se reconoce desvitrificación en el vidrio caracterizada por textura esferulítica (Figura 10d). La riodacitas a diferencia de las dacitas tienen cuarzo como fenocristal, comúnmente con textura de embahiamiento y fracturas curvas (Figura 10a). El piroxeno no es común pero cuando está presente es escaso. La plagioclasa es la fase mineral más abundante y normalmente no sólo presenta textura seriada, sino también coexisten fenocristales con y sin textura de zonación oscilatoria (Figura 10b) y de cedazo (Figura 10d). Las pómez (muestras PAE-30 y PAE-31) corresponden a dacitas porfíricas de anfíbola y biotita inmersos en vidrio incoloro a pardo oscuro vesicular y denso que se caracterizan por presentar textura fluidal (Figura 10c) y perlítica, respectivamente. Las anfíbolas y, en especial, las biotitas aparecen oxidadas. 13
376
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201213 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 4Petrography 41 rock samples were selected and collected during the field campaigns to make transparent slides, which were made by Mr. Rubén Espinosa. Additionally, two available transparent slides (samples GAS-10 and GAS-16) were gathered in the context of “Proyecto Generativo”, one of them being from the Chillahuita dome.The objective of examination of the slides under a petrographic microscope was to identify the minerals and textures present, so as to shed light on the processes that affected the magmas at depth. Yet, from a more practical standpoint, these allow identifying the rock samples that can be dated and also allow better definition of geological units.The results from the petrography of the rocks from the studied area were divided according to lithology in order to highlight the shared characteristics in mineralogy and the textures present. In this way, five well-defined groups can be identified: Lava domes and pumice; dacitic lava flows, siliceous andesites; andesitic lava flows; and lastly, maphic inclusions.4.1 Lava Domes and PumiceThe more evolved rock samples of the study area, amphibole and biotite dacites and rhyodacites (Figure 10a, b, c and d), correspond to the lava domes (or cake-shaped domes) belonging to Stages 2 and 3 of the Chao dome (samples PAE-14 and PAE-1), the Chillahuita dome (sample GAS-10), and to the first stage of construction of the Del León-Lagunita Volcanic Complex (sample PAE-16). Also, there is the dense vesicular pumice from the Chao dome (samples PAE-30 and PAE-31, respectively) (Figure 10c). The mineral association of these rocks is formed by plagioclase-amphibole-biotite±quartz±pyroxene-oxides of Fe-Ti.The cake-shaped domes are dacites having a coarse-grained porphyritic texture made up of phenocrysts of amphibole and biotite immersed in colourless to yellowish glass with perlitic texture (Figure 10a, b, and d) and some flow banding. In some cases, it is possible to identify devitrification in the glass characterised by a spherulitic texture (Figure 10d). In contrast to dacites, the rhyodacites have quartz as phenocrysts, usually with embayments and curved fractures (Figure 10a). Pyroxene is not common, but when present it is in minimal amounts. Plagioclase is the most abundant mineral phase and normally presents not only serialised texture but there is a coexistence of phenocrysts with and without oscillatory zoning (Figure 10b) and a sieve texture (Figure 10d).The pumice (samples PAE-30 and PAE-31) corresponds to porphyritic dacite of amphibole and biotite immersed in dense vesicular glass, colourless to dark, characterised with flow banding (Figure 10c) and perlitic texture, respectively. The amphiboles and especially the biotites appear to be oxidised.
13
Annex VIII Appendix D
377
Geología 1:50.000 del área de la cadena…– Agosto, 2012 14 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 10. Microfotografías de cortes transparentes seleccionados de muestras de dacitas y riodacitas. a) Fenocristales de biotita y anfíbolas oxidadas y cuarzo con textura de embahiamiento y fracturas curvas inmersos en vidrio con textura perlítica (muestra del Domo Chao, Etapa 3). b) Coexistencia de fenocristales de plagioclasa con y sin textura de zonación oscilatoria, además de fenocristales de anfíbola y plagioclasa (muestra del Domo Chao, Etapa 2). c) Textura fluidal en pómez vesicular de la Etapa 1 del Domo Chao. d) Textura esferulítica y perlítica producto de desvitrificación en dacita del Complejo Volcánico del León-Lagunita (Etapa 1). Además, coexisten plagioclasas con y sin textura de cedazo. 4.2Coladas de lava dacíticas Las coladas de lava dacíticas (Figura 11a, b, c y d) sólo se han reconocido en los complejos volcánico Paniri (Muestras PAE-55 y PAE 43 de la Etapa 1, muestras PAE-25 y PAE-2 de las etapas 2 y 3, respectivamente) y del León-Lagunita (muestras PAE-18 y PAE-48, ambas de la Etapa 2). Las rocas corresponden a dacitas porfíricas y, en algunos casos, glomerofídicas con distintos grado de vesicularidad que están constituidas por plagioclasa-anfíbola±biotita±piroxeno-óxidos de Fe-Ti. La masa fundamental comúnmente está conformada por vidrio incoloro a color pardo pálido a amarillento con microlitos de plagioclasa y gránulos de óxidos de Fe-Ti, dando origen a una textura intersertal y, en algunos casos traquítica y fluidal y excepcionalmente a esferulítica (muestra PAE-25). La ocurrencia de plagioclasas con textura de zonación oscilatoria y de cedazo, está última principalmente en el núcleo y con un borde sobrecrecido es común. También es habitual reconocer algunas inclusiones máficas de microcristales de plagioclasa, piroxeno y 14
378
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201214 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 10. Photomicrographs of selected transparent slides of dacite and rhyodacite samples. a) Phenocrysts of oxidised amphiboles and biotite, and quartz with embayments and curved fractures immersed in glass with perlitic texture (sample from Chao dome, Stage 3). b) Coexistence of plagioclase phenocrysts with and without oscillatory zoning, as well as phenocrysts of amphibole and plagioclase (sample from Chao dome, Stage 2). c) Flow banding in vesicular pumice from Stage 1 of the Chao dome. d) Spherulitic and perlitic texture due to devitrification in dacite from the Del León-Lagunita Volcanic Complex (Stage 1). Additionally, coexistence of plagioclase with and without sieve texture.4.2 Dacitic Lava FlowsThe dacitic lava flows (Figure 11a, b, c, and d) have only been identified in the volcanic complexes of Paniri (samples PAE-55 and PAE 43 from Stage 1, samples PAE-25 and PAE-2 from Stages 2 and 3, respectively) and Del León-Lagunita (samples PAE-18 and PAE-48, both from Stage 2). These rocks correspond to porphyritic dacites and, in some cases, glomerophyric dacite with different levels of vesicularity that are formed by plagioclase-amphibole±biotite±pyroxene-oxides of Fe-Ti. The groundmass is usually made up of glass, colourless to pale brown to yellowish in colour, with plagioclase microlites and granules of Fe-Ti oxides, generating an intersertal texture and, in some cases, trachytic texture and flow banding, and exceptionally spherulitic texture (sample PAE-25). It is normal to see the occurrence of plagioclase with oscillatory zoning and sieve texture, the latter is common mainly in the core and with an overgrown rim. It is also common to identify some maphic inclusions of microcrystals of plagioclase, pyroxene, and 14
Annex VIII Appendix D
379
Geología 1:50.000 del área de la cadena…– Agosto, 2012 15 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl óxidos de Fe-Ti inmersos en vidrio color pardo oscuro. Por lo general, las biotitas y anfíbolas están oxidadas y algunas veces con borde opacítico e incluso en el caso de las anfíbolas algunas totalmente opacíticas. Figura 11. Microfotografías de cortes transparentes de muestras de rocas de los complejos volcánicos Paniri (a y b) y del León-Lagunita (c y d). a) Textura porfírica y vesicular con fenocristales de plagioclasa, anfíbola y piroxeno. Coexisten plagioclasas frescas y con textura de cedazo caracterizado por un núcleo reabsorbido y el borde sobrecrecido. b) Fenocristal de anfíbola con borde opacítico. c) Coexistencia de plagioclasas con y sin textura de zonación oscilatoria. d) Fenocristales esqueletales de anfíbola totalmente opacíticos. 4.3Andesitas silíceas Las andesitas silíceas (Figura 12a, b, c y d) corresponden a muestras de roca pertenecientes a los complejos volcánicos Paniri (muestras PAE-5, PAE-8, PAE-9, PAE-10 y PAE-26 de la Etapa 2 y PAE-43 de la Etapa 1) y del León-Lagunita (muestras PAE-38 y PAE-37 de la Etapa 2 y PAE-44 de la Etapa 1) y del volcán Toconce (PAE-4, PAE-36, PAE-41 y PAE-42). Las andesitas silíceas (o daciandesitas) se pueden dividir en 3 grupos bien definidos cuya principal diferencia radica en los minerales ferromagnesianos hidratados que presentan dentro de su asociación mineral: andesitas silíceas de piroxeno y escasa biotita, andesitas silíceas de piroxeno y escasa anfíbola, andesitas silíceas de piroxeno y escasa anfíbola y biotita. 1515
380
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201215 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Fe-Ti oxides immersed in dark brown glass. In general, the biotites and amphiboles are oxidised and, sometimes, have an opacitic rim and some– in the case of amphiboles - are even completely opacitic. Figure 11. Photomicrographs of transparent slices of rock samples from the volcanic complexes of Paniri (a and b) and Del León-Lagunita (c and d). a) Porphyritic and vesicular texture with phenocrysts of plagioclase, amphibole and pyroxene. Coexistence of fresh plagioclase and with sieve texture characterised by a reabsorbed core and overgrown rim. b) Amphibole phenocryst with opacitic rim. c) Coexistence of plagioclase with and without oscillatory zoning. d) Fully opacitic skeletal amphibole phenocrysts.4.3 Siliceous Andesites The siliceous andesites (Figure 12a, b, c, and d) correspond to rock samples from the volcanic complexes Paniri (samples PAE-5, PAE-8, PAE-9, PAE-10, and PAE-26 from Stage 2, and PAE-43 from Stage 1) and Del León-Lagunita (samples PAE-38 and PAE-37 from Stage 2, and PAE-44 from Stage 1), and from the Toconce volcano (PAE-4, PAE-36, PAE-41, and PAE-42).The siliceous andesites (or daciandesites) can be divided into 3 well-defined groups whose main difference lies in the hydrated ferromagnesian minerals they present within their mineral association: siliceous pyroxene andesite with few biotite, siliceous pyroxene andesite with few amphiboles, siliceous pyroxene andesite with few amphibole and biotite.
15
Annex VIII Appendix D
381
Geología 1:50.000 del área de la cadena…– Agosto, 2012 16 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 12. Microfotografías de cortes transparentes de muestras de andesitas silíceas de los complejos volcánicos Paniri (a, b y c) y del León-Lagunitas (d). a) Andesita silícea de piroxeno±anfíbola±biotita de textura porfírica con fenocristales de piroxeno y plagioclasa con y sin textura de zonación oscilatoria. b) Cúmulo de piroxeno, plagiolcasa y óxidos de Fe-Ti y coexistencia de plagioclasas con y sin textura de cedazo en Andesita silícea de piroxeno±biotita. c) Andesita silícea de piroxeno±anfíbola con fenocristales de plagioclasa y piroxeno inmersos en una masa fundamental con microlitos de plagioclasa y gránulos de óxidos de Fe-Ti. d) Andesita silícea de piroxeno±anfíbola±biotita de masa fundamental de vidrio pardo amarillento de textura intersertal con fenocristales de plagiolcasa, piroxeno y anfíbolas oxidados y borde opacíticos. a)de piroxeno±biotita (muestras PAE-4, PAE-26 y PAE-36)Las andesitas silíceas de piroxeno y escasa biotita son porfíricas (Figura 12b) y vesiculares con fenocristales de plagioclasa, piroxeno, biotita (escasos) y óxidos de Fe-Ti inmersos en masa fundamental de vidrio color pardo oscuro y amarillento con textura intersertal y, en algunos casos, fluidal y perlítica. Algunas plagioclasas presentan textura de zonación oscilatoria y un núcleo con textura de cedazo y borde sobrecrecido (Figura 12b). Comúnmente los fenocristales de biotita aparecen oxidados. b)de piroxeno±anfíbola (muestras PAE-5, PAE-10, PAE-38, PAE-41 y PAE-43)Estas andesitas síliceas son de textura porfírica (Figura 12c) y, en algunos casos, vesicular que se caracterizan por una asociación mineral constituida por plagioclasa-piroxeno±anfíbola-óxidos de Fe-Ti. La masa fundamental presenta texturas variables de perlítica, intersertal (Figura 12c) e intergranular con vidrio color pardo pálido, 16
382
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201216 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 12. Photomicrographs of transparent slides of siliceous andesite samples from the volcanic complexes Paniri (a, b, and c) and Del León-Lagunitas (d). a) Siliceous andesite of pyroxene±amphibole±biotite having porphyritic texture, with phenocrysts of pyroxene and plagioclase with and without oscillatory zoning. b) Cluster of pyroxene, plagioclase and Fe-Ti oxides and coexistence of plagioclase with and without sieve texture in siliceous andesite of pyroxene±biotite. c) Siliceous andesite of pyroxene±amphibole containing phenocrysts of plagioclase and pyroxene immersed in groundmass with plagioclase microlites and granules of Fe-Ti oxides. d) Siliceous andesite of pyroxene±amphibole±biotite from groundmass of yellowish brown glass having intersertal texture with phenocrysts of plagioclase, oxidised pyroxene and amphiboles, and opacitic rims.a)Of Pyroxene±Biotite (Samples PAE-4, PAE-26, and PAE-36)The siliceous andesites of pyroxene with few biotites are porphyritic (Figure 12b) and vesicular containing phenocrysts of plagioclase, pyroxene, biotite (few) and Fe-Ti oxides immersed in groundmass of dark brown and yellowish glass having intersertal texture and, in some cases, flow banding and perlitic texture. Some plagioclase present oscillatory zoning and a core having sieve texture and overgrown rim (Figure 12b). Biotite phenocrysts usually appear oxidised.b)Of Pyroxene±Amphibole (Samples PAE-5, PAE-10, PAE-38, PAE-41, and PAE-43)These siliceous andesites have porphyritic (Figure 12c) and, in some cases, vesicular texture that are characterised by a mineral association constituted by plagioclase-pyroxene±amphibole-Fe-Ti oxides. The groundmass presents variable textures such as perlitic, intersertal (Figure 12c), and inter-granular with pale
16
Annex VIII Appendix D
383
Geología 1:50.000 del área de la cadena…– Agosto, 2012 17 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl amarillento a incoloro cuando está presente. Aparecen algunos fenocristales de plagioclasa con textura de zonación oscilatoria y núcleos con textura de cedazo y borde sobrecrecido, además de, algunos enclaves microcristalinos. c)de piroxeno±anfíbola±biotita (muestras PAE-8, PAE-9, PAE-37, PAE-42 y PAE-44)Las andesitas síliceas de piroxeno y escasas anfíbola y biotita son de textura porfírica (Figura 12a) y, en algunos casos, traquítica y vesicular. La masa fundamental comúnmente de textura intersertal (Figura 12 d), raramente perlítica (PAE-9), con microlitos de plagioclasa y gránulos de óxidos de Fe-Ti diseminados. Algunos cristales de plagioclasa con textura de zonación oscilatoria (Figura 12a) y núcleo con textura de cedazo y borde sobrecrecido. También se reconocen algunas inclusiones máficas microcristalinas. Comunmente ambos minerales hidratados aparecen oxidados (Figura 12d). 4.4Coladas de lava andesíticas Las coladas de lava andesíticas se reconocen en los volcanes Negro (muestras PAE-12 y PAE-21) y Toconce (muestras PAE-46 y PAE-51) y los complejos volcánicos del León-Lagunita (muestras PAE-14 y PAE-15, ambas de la Etapa 1) y Paniri (muestras PAE-6 y PAE-23 de la Etapa 2 y PAE-3, PAE-27 y PAE-29 de la Etapa 1). Las andesitas son de textura porfírica (Figura 13a, b, c y d) y, en algunos casos, vesicular (Figura 13a y b) y/o microvesicular, constituida por una asociación mineral de plagioclasa-piroxeno-óxidos de Fe-Ti. Dominantemente, los cristales de piroxeno están formando cúmulos (Figura 13c y d) aunque también aparece como fenocristales aislados. La masa fundamental presenta textura intersertal dominante con microlitos de plagioclasa y gránulos de óxidos de Fe-Ti diseminados (muy abundantes en muchos casos) y escaso a nulos microcristales de piroxeno inmersos en vidrio color pardo oscuro, pardo pálido y pardo amarillento. A veces, la masa fundamental tiene textura prácticamente intergranular con muy escaso a nulo vidrio. Es común la ocurrencia de plagioclasas con textura de zonación oscilatoria y de cedazo en el núcleo con el borde sobrecrecido (Figura 13a, b y d). También, se reconocen inclusiones máficas de microcristales plagioclasa, piroxeno y óxidos de Fe-Ti. Las andesitas pertenecientes al CVLL presentan además, textura fluidal y traquítica. 4.5Inclusiones máficas Las inclusiones máficas son bastante comunes en las rocas de los centros eruptivos del área del Proyecto Paniri en particular y, en los volcanes de la Zona Volcánica Central en general. Corresponden a acumulaciones microcristalinas de plagioclasa, piroxeno y óxidos de Fe-Ti inmersos en vidrio incoloro a color pardo pálido (Figura 14a, b y c). Las inclusiones, normalmente, se diferencian de los glomérulos de fenocristales (o acumulaciones) (Figura 14d), textura común en las rocas de composición intermedia a básica, por el tamaño de los cristales, el tipo de borde de la inclusión con respecto a la roca circundante y la composición del vidrio intercristalino. 17
384
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201217 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl brown or yellowish to colourless glass when present. There are some plagioclase phenocrysts having oscillatory zoning and cores with sieve texture and overgrown rim, as well as some microcrystalline inclusions.c)Of Pyroxene±Amphibole±Biotite (Samples PAE-8, PAE-9, PAE-37, PAE-42, and PAE-44)The siliceous andesite of pyroxene with few amphiboles and biotites have porphyritic texture (Figure 12a) and, in some cases, trachytic and vesicular texture. The groundmass usually has intersertal texture (Figure 12 d), rarely perlitic (PAE-9), containing plagioclase microlites and scattered granules of Fe-Ti oxides. Some plagioclase crystals having oscillatory zoning (Figure 12a) and a core having sieve texture and overgrown rim. It is also possible to identify some microcrystalline maphic inclusions. Normally both hydrated minerals appear oxidised (Figure 12d).4.4 Andesitic Lava Flows The andesitic lava flows can be identified at the Negro (samples PAE-12 and PAE-21) and Toconce (samples PAE-46 and PAE-51) volcanoes and at the volcanic complexes of Del León-Lagunita (samples PAE-14 and PAE-15, both from Stage 1) and Paniri (samples PAE-6 and PAE-23 from Stage 2, and PAE-3, PAE-27, and PAE-29 from Stage 1).The andesites have a porphyritic texture (Figure 13a, b, c, and d) and, in some cases, vesicular (Figure 13a and b) and/or micro-vesicular texture, formed by a mineral association of plagioclase-pyroxene-Fe-Ti oxides. Predominantly, the pyroxene crystals are forming clusters (Figure 13c and d), although they also appear as isolated phenocrysts. The groundmass presents a predominant intersertal texture with plagioclase microlites and dispersed granules of Fe-Ti oxides (very abundant in many cases) and few to no pyroxene microcrystals immersed in glass that is dark brown, pale brown and yellowish in colour. On occasion the groundmass has a practically intergranular texture with very little to no glass. The occurrence of plagioclase with oscillatory zoning and sieve texture in the core with overgrown rim is common (Figure 13a, b, and d). Maphic inclusions of microcrystals of plagioclase, pyroxene and Fe-Ti oxides can also be identified. The andesites from CVLL also present flow banding and trachytic texture.4.5 Maphic InclusionsMaphic inclusions are quite common in the rocks from the eruptive centres of the Paniri Project area in particular, and from the volcanoes of the Central Volcanic Zone in general. These correspond to microcrystalline accumulations of plagioclase, pyroxene and Fe-Ti oxides immersed in glass that is colourless to pale brown (Figure 14a, b, and c). Normally the inclusions differ from the phenocryst glomerules (or clusters) (Figure 14d), a common texture in the rocks having intermediate to basic composition, due to the size of the crystals, the type of rim of the inclusion with respect to the surrounding rock, and the composition of the inter-crystalline glass.
17
Annex VIII Appendix D
385
Geología 1:50.000 del área de la cadena…– Agosto, 2012 18 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 13. Microfotografías de cortes transparentes de muestras de andesitas del Complejo Volcánico Paniri (a, b y c) y del volcán Toconce (d). a) Textura porfírica y vesicular. Coexisten fenocristales de plagioclasa con y sin textura de cedazo. b) Textura vesicular y porfírica de fenocristales de piroxeno, óxidos de Fe-Ti y plagioclasa, algunas con textura de cedazo. c) Glomérulo de fenocristales de plagioclasa, piroxeno y óxidos de Fe-Ti. d) Glomérulo de fenocristales de piroxeno y plagioclasa coexistiendo con fenocristal de textura de cedazo. 18
386
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201218 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 13. Photomicrographs of transparent slices of andesites from the Paniri Volcanic Complex (a, b, and c) and from the Toconce volcano (d). a) Porphyritic and vesicular texture. Coexistence of plagioclase phenocrysts with and without sieve texture. b) Vesicular and porphyritic texture of phenocrysts of pyroxene, Fe-Ti oxides and plagioclase, some with sieve texture. c) Glomerule of phenocrysts of plagioclase, pyroxene and Fe-Ti oxides. d) Glomerule of phenocrysts of pyroxene and plagioclase coexisting with phenocryst having sieve texture.
18
Annex VIII Appendix D
387
Geología 1:50.000 del área de la cadena…– Agosto, 2012 19 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 14. Microfotografías de inclusiones máficas en muestras de rocas del Proyecto Paniri. a) Detalle de inclusión máfica constituida de microcristales de plagioclasa, piroxenos y óxidos de Fe-Ti en inmersos en vidrio color pardo pálido con abundantes vesículas. b) En el extremo superior derecho se reconoce un enclave máfico con vesículas y microcristales de plagioclasa, piroxeno y óxidos de Fe-Ti en vidrio color pardo pálido. c) Enclave máfico microcristalino de plagiolcasas, piroxenos y óxidos de Fe-Ti. d) Glomérulo de fenocristales de plagiolcasa, piroxeno y óxidos de Fe-Ti. 19
388
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201219 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 14. Photomicrographs of maphic inclusions in rock samples from the Paniri Project. a) Detail of maphic inclusion formed by microcrystals of plagioclase, pyroxenes and Fe-Ti oxides immersed in pale brown glass with abundant vesicles. b) Upper right part: a maphic cluster with vesicles and microcrystals of plagioclase, pyroxene and Fe-Ti oxides in pale brown glass. c) Microcrystalline maphic cluster of plagioclase, pyroxenes and Fe-Ti oxides. d) Glomerule of phenocrysts of plagioclase, pyroxene and Fe-Ti oxides.
19
Annex VIII Appendix D
389
Geología 1:50.000 del área de la cadena…– Agosto, 2012 20 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 5Geoquímica Los análisis químicos de roca total permiten caracterizar y clasificar los productos de los distintos centros eruptivos del área del Proyecto Paniri. Asimismo, es posible proponer algunas hipótesis, mediante la utilización de una serie de diagramas, sobre los procesos magmáticos que han afectado a los magmas que dieron origen a estas rocas en el contexto de un ambiente de margen de subducción activo caracterizado por un potente espesor cortical como es el caso de la Zona Volcánica Central (ZVC: 16-28º S), segmento volcánico en el cual se encuentra nuestra zona de interés. 5.1Selección de muestras Se seleccionaron 33 muestras de roca para los análisis de roca total (elementos mayores y trazas, incluyendo las tierras raras) (Tabla 2) que representan a los 4 centros eruptivos más recientes que constituyen la cadena volcánica NW-SE: Complejo Volcánico Paniri (CVP), Domo Chao, Complejo Volcánico del León-Lagunita y el volcán Toconce. Además, se incluyeron 2 muestras del volcán Negro que por el grado de preservación del edificio volcánico parece ser más antiguo que el resto de los centros eruptivos de la cadena y tiene una posición ligeramente más al norte de la misma. Finalmente, se incluyó una muestra de una roca volcánica perteneciente a la Formación Lomas Negras (Cretácico) caracterizada por representar un volcanismo de afinidad toleítica. 5.2Análisis químicos Las muestras de roca del área de estudio seleccionadas para los análisis químicos de roca total (elementos mayores y trazas, incluyendo las tierras raras) fueron enviadas a ACME Laboratories (Vancouver, Canadá) con oficina en Chile donde sólo se realiza la molienda de las muestras. Los elementos mayores fueron analizados mediante la técnica analítica de ICP-ES (Espectrometría de Emisión con fuente de Plasma Acoplado Inductivamente), mientras que, los elementos trazas fueron analizados mediante la técnica analítica de ICP-MS (Espectrometría de Masas con fuente de Plasma Acoplado Inductivamente) (p.ej., Jarvis, 1988). Por su parte, el contenido de volátiles (contenido de H2O, CO2 y S) fue determinado mediante la técnica de LOI (o pérdida por calcinación) (Lechler y Desilets, 1987). Finalmente, el contenido de C y S en la muestras fue analizado mediante el método LECO donde se quema (oxida) el C y S (p. ej., Mular et al., 2002) y se mide al gas resultante mediante un detector molecular infrarrojo. 5.3Elementos mayores Los resultados de los análisis químicos de elementos mayores de las muestras de roca seleccionadas del área del Proyecto Paniri indican que corresponden a una serie subalcalina (límite de los campos según Irvine y Baragar, 1971) de composición intermedia y evolucionada con contenidos de sílice variable entre 57 a 68% en peso (Tabla 2) (Figura 15). Además, el contenido de K2O varía entre 2,1 y 4,4% en peso, valores típicos de las series calco-alcalinas de alto contenido de potasio (límites de los campos según Rickwood, 1989) (Figura 16). Por su parte, la muestra de la colada de lava de la Formación Lomas Negras se caracteriza por una afinidad alcalina (traquiandesita basáltica, Figura 15), shoshonítica (Figura 16) y un alto contenido de K2O (5% en peso). 20
390
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201220 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 5Geochemistry The total rock chemical analyses help characterise and classify the products of the different eruptive centres. By using a series of diagrams it is also possible to set forth hypotheses about the processes that have affected the magmas which gave rise to these rocks in the context of an active subduction-border environment characterised by a large crust thickness, as is the case of the Central Volcanic Zone (ZVC: 16-28º S), where our area of interest is located.5.1 Selection of Samples33 rock samples were selected for the total rock analyses (major and trace elements, including rare earth elements) (Table 2) that represent the 4 more recent eruptive centres forming the NW-SE volcanic chain: the Paniri Volcanic Complex (CVP), Chao dome, Del León-Lagunita Volcanic Complex, and the Toconce volcano. In addition, 2 samples from the Negro volcano were included because, due to the level of preservation of the volcanic edifice, it seems to be older than the rest of the eruptive centres of the chain and it is positioned slightly further north of the chain. Lastly, one sample of a volcanic rock from the Lomas Negras Formation (Cretaceous) characterised by representing volcanism of tholeitic affinity was included.5.2 Chemical AnalysesThe rock samples were sent to ACME Laboratories (Vancouver, Canada) who have offices in Chile where sample grinding is performed.The major elements were analysed by ICP-ES analytical techniques (Inductively Coupled Plasma Emission Spectrometry), while the trace elements were analysed by ICP-MS analytical techniques (Inductively Coupled Plasma Mass Spectrometry) (e.g. Jarvis, 1988). Furthermore, the content of volatiles ( H2O, CO2, and S) was determined by the LOI technique (Loss On Ignition) (Lechler and Desilets, 1987). Lastly, the content of C and S in the samples was analysed through the LECO method in which the C and S content is burned (oxidised) (e.g. Mular et al., 2002) and the resulting gas is measured with an infrared molecular detector.5.3 Major ElementsThe chemical analyses results for the major elements of the selected rock samples from the Paniri Project area indicate that they correspond to a sub-alkaline series (field limit according to Irvine and Baragar, 1971) of an intermediate evolved composition having silica contents ranging between 57 - 68% by weight (Table 2)(Figure 15). Moreover, the K2O content varies between 2.1 and 4.4% by weight, which are typical values for the high potassium calc-alkaline series (field limits according to Rickwood, 1989) (Figure 16). In its turn, the lava flow sample from the Lomas Negras Formation is characterised by an alkaline affinity (basaltic trachyandesite, Figure 15), shoshonitic (Figure 16) and having a high K2O content (5% by weight).
20
Annex VIII Appendix D
391
Geología 1:50.000 del área de la cadena…– Agosto, 2012 21 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Tabla 2. Resultados de los análisis de roca total (elemento mayores y trazas en % en peso y ppm, respectivamente) de las muestras de roca del Proyecto Paniri. CVP: Complejo Volcánico Paniri, CVLL: Complejo Volcánico del León-Lagunita y Fm. LN: Formación Lomas Negras. Unidad Chao 3 Chao 2 Chao 1 Chao 1 CVP 3 CVP 2 CVP 2 CVP 2 CVP 2 (%) PAE-1 PAE-11 PAE-30 PAE-31 PAE-2 PAE-23 PAE-25 PAE-5 PAE-6 SiO2 67,15 66,69 67,72 66,32 64,55 64,28 64,33 65,00 57,20 Al2O3 15,19 15,02 14,61 15,69 15,59 15,78 16,01 15,97 16,93 Fe2O3 3,64 3,62 2,99 3,80 4,39 4,47 4,53 3,95 6,91 MgO 1,69 1,65 1,35 1,70 2,22 1,98 2,07 1,17 4,00 CaO 3,54 3,56 3,02 3,78 3,93 4,36 4,11 3,43 6,79 Na2O 3,30 3,28 3,26 3,45 3,58 3,68 3,65 3,70 3,53 K2O 3,64 3,80 4,03 3,60 3,30 3,06 3,52 3,55 2,13 TiO2 0,48 0,48 0,43 0,50 0,56 0,58 0,62 0,56 0,85 P2O5 0,11 0,14 0,11 0,11 0,14 0,19 0,17 0,20 0,22 MnO 0,06 0,06 0,05 0,06 0,07 0,08 0,07 0,06 0,10 LOI 1,0 1,5 2,3 0,8 1,5 1,3 0,7 2,2 1,1 Total 98,80 98,30 97,57 99,01 98,33 98,46 99,08 97,59 98,66 (ppm) Cr 41 34 27 34 55 14 27 48 Ni 6,4 4,7 5,9 2,8 3,9 1,4 6,8 Sc 7 7 6 8 9 10 9 6 18 Ba 596 570 616 605 698 723 692 760 591 Co 8,3 6,8 5,2 7,5 10,7 8,4 10 5,2 18,1 Cs 12,4 12,6 12,8 12 8,8 5,6 10 6,6 2,1 Ga 18,4 16,2 15,5 17,4 18,8 17,4 17,8 18,4 17,8 Hf 4,5 4,8 4,2 5,1 4,7 5,2 6 5,9 3,6 Nb 9,7 10,4 9,2 9,1 9,3 9 10,9 11,1 7,6 Pb 2,1 1,2 1 1,5 2,8 0,8 0,8 1,7 0,8 Rb 166,9 157,2 161,7 153 157,0 118,5 163,7 138,5 57,5 Sr 364,6 307,4 303,1 369,1 433,1 412,8 405,4 432,2 547,5 Ta 1,2 1,2 1,2 1,1 0,9 0,9 0,9 0,9 0,5 Th 28,2 24,8 25,5 23,7 23 13,9 22,5 15,6 5,6 U 11,2 9,6 9,5 8,9 7,4 4,5 7,1 4,5 1,5 V 73 70 53 75 93 87 98 72 158 Zn 52 34 18 27 63 29 27 33 28 Zr 145,6 147,9 120,4 133,8 167,9 166 182,7 180,1 134,1 Y 14,7 13 10,6 12,9 15,4 16,7 16,9 13,4 16 La 33,7 29,9 30,6 30,9 32,6 25,6 32,8 31,7 20,5 Ce 64,7 58,1 58,2 59 69,4 50,8 63,6 63,8 41,5 Pr 6,94 6,32 6,3 6,47 7,06 5,98 7,53 7,11 5,06 Nd 24,2 23,7 22,6 22,1 29 22,4 28,4 26,8 20,2 Sm 4,33 3,93 3,67 4,07 4,54 4,08 4,89 4,48 4,2 Eu 0,76 0,72 0,68 0,85 0,9 0,84 0,97 0,9 1,05 Gd 3,19 3 2,73 2,96 3,5 3,13 3,97 3,47 3,6 Tb 0,49 0,41 0,37 0,43 0,51 0,49 0,55 0,48 0,52 Dy 2,66 2,4 1,96 2,25 2,62 2,7 3,22 2,51 2,78 Ho 0,5 0,42 0,36 0,41 0,61 0,52 0,6 0,47 0,56 Er 1,38 1,05 1,01 1,06 1,62 1,49 1,58 1,29 1,42 Tm 0,22 0,19 0,17 0,19 0,22 0,25 0,26 0,21 0,22 Yb 1,53 1,19 1,14 1,21 1,51 1,42 1,74 1,15 1,5 Lu 0,24 0,18 0,19 0,19 0,23 0,24 0,26 0,2 0,22 21
392
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201221 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Table 2. Results of the total rock analyses (major and trace elements in % by weight and ppm, respectively) for the rock samples from the Paniri Project. CVP: Paniri Volcanic Complex, CVLL: Del León-Lagunita Volcanic Complex and Fm. LN: Lomas Negras Formation.Unit Chao 3 Chao 2 Chao 1 Chao 1 CVP 3 CVP 2 CVP 2 CVP 2 CVP 2 (%) PAE-1 PAE-11 PAE-30 PAE-31 PAE-2 PAE-23 PAE-25 PAE-5 PAE-6 SiO2 67,15 66,69 67,72 66,32 64,55 64,28 64,33 65,00 57,20 Al2O3 15,19 15,02 14,61 15,69 15,59 15,78 16,01 15,97 16,93 Fe2O3 3,64 3,62 2,99 3,80 4,39 4,47 4,53 3,95 6,91 MgO 1,69 1,65 1,35 1,70 2,22 1,98 2,07 1,17 4,00 CaO 3,54 3,56 3,02 3,78 3,93 4,36 4,11 3,43 6,79 Na2O 3,30 3,28 3,26 3,45 3,58 3,68 3,65 3,70 3,53 K2O 3,64 3,80 4,03 3,60 3,30 3,06 3,52 3,55 2,13 TiO2 0,48 0,48 0,43 0,50 0,56 0,58 0,62 0,56 0,85 P2O5 0,11 0,14 0,11 0,11 0,14 0,19 0,17 0,20 0,22 MnO 0,06 0,06 0,05 0,06 0,07 0,08 0,07 0,06 0,10 LOI 1,0 1,5 2,3 0,8 1,5 1,3 0,7 2,2 1,1 Total 98,80 98,30 97,57 99,01 98,33 98,46 99,08 97,59 98,66 (ppm) Cr 41 34 27 34 55 14 27 48 Ni 6,4 4,7 5,9 2,8 3,9 1,4 6,8 Sc 7 7 6 8 9 10 9 6 18 Ba 596 570 616 605 698 723 692 760 591 Co 8,3 6,8 5,2 7,5 10,7 8,4 10 5,2 18,1 Cs 12,4 12,6 12,8 12 8,8 5,6 10 6,6 2,1 Ga 18,4 16,2 15,5 17,4 18,8 17,4 17,8 18,4 17,8 Hf 4,5 4,8 4,2 5,1 4,7 5,2 6 5,9 3,6 Nb 9,7 10,4 9,2 9,1 9,3 9 10,9 11,1 7,6 Pb 2,1 1,2 1 1,5 2,8 0,8 0,8 1,7 0,8 Rb 166,9 157,2 161,7 153 157,0 118,5 163,7 138,5 57,5 Sr 364,6 307,4 303,1 369,1 433,1 412,8 405,4 432,2 547,5 Ta 1,2 1,2 1,2 1,1 0,9 0,9 0,9 0,9 0,5 Th 28,2 24,8 25,5 23,7 23 13,9 22,5 15,6 5,6 U 11,2 9,6 9,5 8,9 7,4 4,5 7,1 4,5 1,5 V 73 70 53 75 93 87 98 72 158 Zn 52 34 18 27 63 29 27 33 28 Zr 145,6 147,9 120,4 133,8 167,9 166 182,7 180,1 134,1 Y 14,7 13 10,6 12,9 15,4 16,7 16,9 13,4 16 La 33,7 29,9 30,6 30,9 32,6 25,6 32,8 31,7 20,5 Ce 64,7 58,1 58,2 59 69,4 50,8 63,6 63,8 41,5 Pr 6,94 6,32 6,3 6,47 7,06 5,98 7,53 7,11 5,06 Nd 24,2 23,7 22,6 22,1 29 22,4 28,4 26,8 20,2 Sm 4,33 3,93 3,67 4,07 4,54 4,08 4,89 4,48 4,2 Eu 0,76 0,72 0,68 0,85 0,9 0,84 0,97 0,9 1,05 Gd 3,19 3 2,73 2,96 3,5 3,13 3,97 3,47 3,6 Tb 0,49 0,41 0,37 0,43 0,51 0,49 0,55 0,48 0,52 Dy 2,66 2,4 1,96 2,25 2,62 2,7 3,22 2,51 2,78 Ho 0,5 0,42 0,36 0,41 0,61 0,52 0,6 0,47 0,56 Er 1,38 1,05 1,01 1,06 1,62 1,49 1,58 1,29 1,42 Tm 0,22 0,19 0,17 0,19 0,22 0,25 0,26 0,21 0,22 Yb 1,53 1,19 1,14 1,21 1,51 1,42 1,74 1,15 1,5 Lu 0,24 0,18 0,19 0,19 0,23 0,24 0,26 0,2 0,22 21
Annex VIII Appendix D
393
Geología 1:50.000 del área de la cadena…– Agosto, 2012 22 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Tabla 2. Continuación. Unidad CVP 2 CVP 2 CVP 2 CVP 1 CVP 1 CVP 1 CVP 1 CVP 1 CVLL 2 (%) PAE-9 PAE-8 PAE-10 PAE-3 PAE-54 PAE-55 PAE-56 PAE-43 PAE-37 SiO2 64,84 63,95 63,71 62,03 63,85 67,70 61,08 62,52 60,10 Al2O3 16,12 15,78 16,50 16,84 16,25 15,53 17,32 16,41 16,36 Fe2O3 4,18 4,51 4,55 5,36 4,44 2,96 5,95 4,83 6,48 MgO 1,47 2,13 1,68 2,17 2,02 0,96 1,92 2,05 4,21 CaO 3,88 4,29 4,18 5,07 4,32 2,99 5,01 4,36 5,85 Na2O 3,94 3,45 3,83 3,71 3,72 3,65 4,09 3,26 3,15 K2O 3,20 3,49 3,04 2,59 2,83 3,66 2,86 3,62 2,59 TiO2 0,57 0,61 0,61 0,74 0,64 0,46 0,91 0,62 0,81 P2O5 0,19 0,20 0,18 0,18 0,17 0,13 0,26 0,18 0,17 MnO 0,06 0,07 0,07 0,08 0,07 0,05 0,09 0,07 0,1 LOI 1,4 1,3 1,4 1,0 1,5 1,7 0,3 1,9 -0,1 Total 98,45 98,48 98,35 98,77 98,31 98,09 99,49 97,92 99,82 (ppm) Cr 34 27 21 157 Ni 1,8 4 2,1 43 Sc 8 9 9 11 9 5 14 8 15 Ba 739 680 697 721 742 797 687 708 642 Co 8,2 9,2 9,9 11,5 10,7 4,4 10,3 9,7 17,9 Cs 4,9 8,8 4,6 4,1 3,8 6,1 4,2 5,4 6,2 Ga 17,7 17,6 17,2 20 19,5 17,9 19,1 18,9 20,4 Hf 5,5 5,5 4,7 5,2 4,2 5,2 5,2 4,8 4,8 Nb 8,8 10,9 8,5 9,4 8 8,7 10,9 10,4 9,7 Pb 0,7 0,9 1 2 0,7 0,5 0,9 1,1 4 Rb 119,6 153,2 109,9 93,3 105,3 144,9 104,2 125,7 104,6 Sr 426,5 419,4 428,6 548,4 505,4 404,3 477 492,1 465 Ta 0,8 1 0,9 0,7 0,5 0,8 0,8 0,9 0,8 Th 13,1 22 11,9 10,8 11,5 17,1 12,6 15,6 12,7 U 4,1 6,9 3,6 3,5 3,9 5,1 3,9 3,2 3,8 V 83 90 93 127 102 61 121 94 153 Zn 28 26 25 42 28 20 41 42 52 Zr 166,1 181,5 156,2 167,5 174,4 192,4 203,8 181,3 165,5 Y 15,6 16,9 15 17,2 15,1 13,5 26,4 14,6 17,5 La 26,7 31,7 25,1 27,6 29 32,3 31,1 33,3 29,6 Ce 54,5 63,8 50 58,3 61,2 67,6 67,5 65,3 61,8 Pr 6,15 7,23 5,81 6,25 6,57 6,94 7,49 7,27 6,87 Nd 24,3 27,3 21,7 20,5 23,3 25,4 29,1 26,4 26,5 Sm 4,43 4,74 3,97 4,58 4,55 4,12 5,75 4,33 4,82 Eu 0,91 0,95 0,87 1 0,87 0,8 1,19 0,96 1,13 Gd 3,47 3,81 3,42 3,94 3,86 3,3 5,14 3,51 4,14 Tb 0,5 0,54 0,5 0,62 0,52 0,46 0,84 0,52 0,62 Dy 2,75 3,1 2,62 2,97 2,38 2,58 4,68 2,77 2,98 Ho 0,56 0,56 0,51 0,65 0,49 0,46 0,91 0,47 0,69 Er 1,41 1,44 1,47 1,7 1,46 1,42 2,6 1,5 1,56 Tm 0,25 0,24 0,23 0,26 0,23 0,18 0,44 0,21 0,27 Yb 1,57 1,61 1,54 1,63 1,27 1,23 2,59 1,41 1,51 Lu 0,23 0,24 0,24 0,26 0,22 0,23 0,42 0,19 0,21 22
394
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201222 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Table 2. ContinuedUnit CVP 2 CVP 2 CVP 2 CVP 1 CVP 1 CVP 1 CVP 1 CVP 1 CVLL 2 (%) PAE-9 PAE-8 PAE-10 PAE-3 PAE-54 PAE-55 PAE-56 PAE-43 PAE-37 SiO2 64,84 63,95 63,71 62,03 63,85 67,70 61,08 62,52 60,10 Al2O3 16,12 15,78 16,50 16,84 16,25 15,53 17,32 16,41 16,36 Fe2O3 4,18 4,51 4,55 5,36 4,44 2,96 5,95 4,83 6,48 MgO 1,47 2,13 1,68 2,17 2,02 0,96 1,92 2,05 4,21 CaO 3,88 4,29 4,18 5,07 4,32 2,99 5,01 4,36 5,85 Na2O 3,94 3,45 3,83 3,71 3,72 3,65 4,09 3,26 3,15 K2O 3,20 3,49 3,04 2,59 2,83 3,66 2,86 3,62 2,59 TiO2 0,57 0,61 0,61 0,74 0,64 0,46 0,91 0,62 0,81 P2O5 0,19 0,20 0,18 0,18 0,17 0,13 0,26 0,18 0,17 MnO 0,06 0,07 0,07 0,08 0,07 0,05 0,09 0,07 0,1 LOI 1,4 1,3 1,4 1,0 1,5 1,7 0,3 1,9 -0,1 Total 98,45 98,48 98,35 98,77 98,31 98,09 99,49 97,92 99,82 (ppm) Cr 34 27 21 157 Ni 1,8 4 2,1 43 Sc 8 9 9 11 9 5 14 8 15 Ba 739 680 697 721 742 797 687 708 642 Co 8,2 9,2 9,9 11,5 10,7 4,4 10,3 9,7 17,9 Cs 4,9 8,8 4,6 4,1 3,8 6,1 4,2 5,4 6,2 Ga 17,7 17,6 17,2 20 19,5 17,9 19,1 18,9 20,4 Hf 5,5 5,5 4,7 5,2 4,2 5,2 5,2 4,8 4,8 Nb 8,8 10,9 8,5 9,4 8 8,7 10,9 10,4 9,7 Pb 0,7 0,9 1 2 0,7 0,5 0,9 1,1 4 Rb 119,6 153,2 109,9 93,3 105,3 144,9 104,2 125,7 104,6 Sr 426,5 419,4 428,6 548,4 505,4 404,3 477 492,1 465 Ta 0,8 1 0,9 0,7 0,5 0,8 0,8 0,9 0,8 Th 13,1 22 11,9 10,8 11,5 17,1 12,6 15,6 12,7 U 4,1 6,9 3,6 3,5 3,9 5,1 3,9 3,2 3,8 V 83 90 93 127 102 61 121 94 153 Zn 28 26 25 42 28 20 41 42 52 Zr 166,1 181,5 156,2 167,5 174,4 192,4 203,8 181,3 165,5 Y 15,6 16,9 15 17,2 15,1 13,5 26,4 14,6 17,5 La 26,7 31,7 25,1 27,6 29 32,3 31,1 33,3 29,6 Ce 54,5 63,8 50 58,3 61,2 67,6 67,5 65,3 61,8 Pr 6,15 7,23 5,81 6,25 6,57 6,94 7,49 7,27 6,87 Nd 24,3 27,3 21,7 20,5 23,3 25,4 29,1 26,4 26,5 Sm 4,43 4,74 3,97 4,58 4,55 4,12 5,75 4,33 4,82 Eu 0,91 0,95 0,87 1 0,87 0,8 1,19 0,96 1,13 Gd 3,47 3,81 3,42 3,94 3,86 3,3 5,14 3,51 4,14 Tb 0,5 0,54 0,5 0,62 0,52 0,46 0,84 0,52 0,62 Dy 2,75 3,1 2,62 2,97 2,38 2,58 4,68 2,77 2,98 Ho 0,56 0,56 0,51 0,65 0,49 0,46 0,91 0,47 0,69 Er 1,41 1,44 1,47 1,7 1,46 1,42 2,6 1,5 1,56 Tm 0,25 0,24 0,23 0,26 0,23 0,18 0,44 0,21 0,27 Yb 1,57 1,61 1,54 1,63 1,27 1,23 2,59 1,41 1,51 Lu 0,23 0,24 0,24 0,26 0,22 0,23 0,42 0,19 0,21 22
Annex VIII Appendix D
395
Geología 1:50.000 del área de la cadena…– Agosto, 2012 23 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Tabla 2. Continuación. Unidad CVLL 2 CVLL 2 CVLL 2 CVLL 1 CVLL 1 CVLL 1 CVLL 1 CVLL 1 CVLL 1 (%) PAE-38 PAE-47 PAE-48 PAE-14 PAE-15 PAE-16 PAE-35 PAE-44 PAE-45 SiO2 59,42 63,88 61,03 59,23 64,45 65,70 59,19 63,10 59,09 Al2O3 16,03 16,15 16,38 17,07 16,79 16,46 16,50 15,92 16,80 Fe2O3 6,47 5,08 6,07 6,44 4,03 3,22 6,26 4,64 6,24 MgO 4,02 2,59 3,50 3,64 1,10 0,84 3,54 2,67 3,56 CaO 6,35 4,62 5,83 6,35 3,88 2,50 5,92 4,57 6,04 Na2O 3,02 3,15 3,18 3,36 4,03 4,00 2,97 3,14 3,16 K2O 2,58 3,16 2,67 2,20 3,34 3,88 2,73 3,39 2,59 TiO2 0,72 0,67 0,74 0,84 0,69 0,53 0,82 0,69 0,83 P2O5 0,16 0,14 0,18 0,21 0,21 0,18 0,19 0,17 0,21 MnO 0,1 0,08 0,09 0,09 0,07 0,08 0,09 0,07 0,09 LOI 0,9 0,3 0,1 0,3 1,2 2,4 1,5 1,4 1,1 Total 98,87 99,52 99,67 99,43 98,59 97,39 98,21 98,36 98,61 (ppm) Cr 62 41 68 89 68 82 68 Ni 31 20 5,6 1,1 1,4 27 Sc 18 12 16 16 8 7 15 10 15 Ba 547 611 592 621 816 949 677 706 683 Co 19,6 11,6 14,7 14,5 4,6 2,8 15,8 10,9 15 Cs 6,1 6,6 5,5 1,9 7,3 5,7 3,3 8,6 3,2 Ga 18,3 17,8 19,3 18,5 19 17,4 19,2 20 20 Hf 3,6 5,2 4,6 4,5 6,1 8,1 4,5 5,9 4 Nb 8,3 10 9,6 7,9 12,9 15,5 8,9 10,8 7,5 Pb 1,8 4 2,6 3,4 1,2 3,5 1,2 1,5 1,4 Rb 101,5 133,8 106,0 69,5 118,6 128,8 81,1 136,7 78,3 Sr 449,7 403,8 453,9 480,5 417,8 318,1 521,3 469,1 526,7 Ta 0,9 1,3 0,7 0,5 1 1 0,6 0,8 0,6 Th 14,6 20,4 16,3 7 12,9 14,2 8,3 18 9 U 4,8 7,4 5,8 1,4 4,8 4 2,6 5,8 2,5 V 152 115 140 145 51 33 149 112 147 Zn 39 27 30 35 23 33 38 27 36 Zr 138 160,1 153,7 149,8 215,3 281,3 156,6 192,9 158,2 Y 19,1 16,9 18,3 15,8 21 30,1 19,2 16 19,2 La 27,5 32,1 28,7 25,5 32,6 39,5 26,1 33,8 26,8 Ce 57,1 67,7 60,2 50,1 65,9 81,5 56,9 70,3 56,9 Pr 6,31 7,18 6,57 6,12 7,9 9,71 6,61 7,62 6,54 Nd 23,1 21,5 25,2 22,6 30,8 36,4 23,6 28,7 24 Sm 4,46 4,59 4,61 4,51 5,77 7 4,84 5,25 5,1 Eu 1,06 0,94 0,96 1,11 1,25 1,41 1,1 0,98 1,12 Gd 4,02 4,27 3,82 3,97 4,78 5,89 4,32 4,06 4,37 Tb 0,61 0,6 0,59 0,55 0,73 0,89 0,66 0,59 0,66 Dy 3,21 3,37 3,45 3,03 3,75 5,27 3,6 3,21 3,51 Ho 0,62 0,55 0,59 0,56 0,74 0,96 0,68 0,58 0,73 Er 1,99 1,65 1,8 1,44 1,8 2,74 1,7 1,55 1,75 Tm 0,28 0,3 0,28 0,22 0,29 0,42 0,3 0,23 0,26 Yb 1,75 1,47 1,73 1,18 1,6 2,59 1,66 1,65 1,6 Lu 0,27 0,26 0,28 0,2 0,27 0,39 0,27 0,21 0,23 23
396
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201223 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Table 2. ContinuedUnit CVLL 2 CVLL 2 CVLL 2 CVLL 1 CVLL 1 CVLL 1 CVLL 1 CVLL 1 CVLL 1 (%) PAE-38 PAE-47 PAE-48 PAE-14 PAE-15 PAE-16 PAE-35 PAE-44 PAE-45 SiO2 59,42 63,88 61,03 59,23 64,45 65,70 59,19 63,10 59,09 Al2O3 16,03 16,15 16,38 17,07 16,79 16,46 16,50 15,92 16,80 Fe2O3 6,47 5,08 6,07 6,44 4,03 3,22 6,26 4,64 6,24 MgO 4,02 2,59 3,50 3,64 1,10 0,84 3,54 2,67 3,56 CaO 6,35 4,62 5,83 6,35 3,88 2,50 5,92 4,57 6,04 Na2O 3,02 3,15 3,18 3,36 4,03 4,00 2,97 3,14 3,16 K2O 2,58 3,16 2,67 2,20 3,34 3,88 2,73 3,39 2,59 TiO2 0,72 0,67 0,74 0,84 0,69 0,53 0,82 0,69 0,83 P2O5 0,16 0,14 0,18 0,21 0,21 0,18 0,19 0,17 0,21 MnO 0,1 0,08 0,09 0,09 0,07 0,08 0,09 0,07 0,09 LOI 0,9 0,3 0,1 0,3 1,2 2,4 1,5 1,4 1,1 Total 98,87 99,52 99,67 99,43 98,59 97,39 98,21 98,36 98,61 (ppm) Cr 62 41 68 89 68 82 68 Ni 31 20 5,6 1,1 1,4 27 Sc 18 12 16 16 8 7 15 10 15 Ba 547 611 592 621 816 949 677 706 683 Co 19,6 11,6 14,7 14,5 4,6 2,8 15,8 10,9 15 Cs 6,1 6,6 5,5 1,9 7,3 5,7 3,3 8,6 3,2 Ga 18,3 17,8 19,3 18,5 19 17,4 19,2 20 20 Hf 3,6 5,2 4,6 4,5 6,1 8,1 4,5 5,9 4 Nb 8,3 10 9,6 7,9 12,9 15,5 8,9 10,8 7,5 Pb 1,8 4 2,6 3,4 1,2 3,5 1,2 1,5 1,4 Rb 101,5 133,8 106,0 69,5 118,6 128,8 81,1 136,7 78,3 Sr 449,7 403,8 453,9 480,5 417,8 318,1 521,3 469,1 526,7 Ta 0,9 1,3 0,7 0,5 1 1 0,6 0,8 0,6 Th 14,6 20,4 16,3 7 12,9 14,2 8,3 18 9 U 4,8 7,4 5,8 1,4 4,8 4 2,6 5,8 2,5 V 152 115 140 145 51 33 149 112 147 Zn 39 27 30 35 23 33 38 27 36 Zr 138 160,1 153,7 149,8 215,3 281,3 156,6 192,9 158,2 Y 19,1 16,9 18,3 15,8 21 30,1 19,2 16 19,2 La 27,5 32,1 28,7 25,5 32,6 39,5 26,1 33,8 26,8 Ce 57,1 67,7 60,2 50,1 65,9 81,5 56,9 70,3 56,9 Pr 6,31 7,18 6,57 6,12 7,9 9,71 6,61 7,62 6,54 Nd 23,1 21,5 25,2 22,6 30,8 36,4 23,6 28,7 24 Sm 4,46 4,59 4,61 4,51 5,77 7 4,84 5,25 5,1 Eu 1,06 0,94 0,96 1,11 1,25 1,41 1,1 0,98 1,12 Gd 4,02 4,27 3,82 3,97 4,78 5,89 4,32 4,06 4,37 Tb 0,61 0,6 0,59 0,55 0,73 0,89 0,66 0,59 0,66 Dy 3,21 3,37 3,45 3,03 3,75 5,27 3,6 3,21 3,51 Ho 0,62 0,55 0,59 0,56 0,74 0,96 0,68 0,58 0,73 Er 1,99 1,65 1,8 1,44 1,8 2,74 1,7 1,55 1,75 Tm 0,28 0,3 0,28 0,22 0,29 0,42 0,3 0,23 0,26 Yb 1,75 1,47 1,73 1,18 1,6 2,59 1,66 1,65 1,6 Lu 0,27 0,26 0,28 0,2 0,27 0,39 0,27 0,21 0,23 23
Annex VIII Appendix D
397
Geología 1:50.000 del área de la cadena…– Agosto, 2012 24 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Tabla 2. Continuación. Unidad Toconce Toconce Toconce Toconce Toconce Toconce Negro Negro Fm. LN (%) PAE-41 PAE-42 PAE-36 PAE-39 PAE-46 PAE-51 PAE-21 PAE-12 PAPA-46 SiO2 65,23 66,64 61,85 57,86 57,57 57,22 61,76 61,50 48,92 Al2O3 15,81 14,57 16,41 17,34 16,62 17,38 16,64 17,06 12,81 Fe2O3 4,35 3,44 5,34 6,79 7,11 6,72 5,73 5,62 10,16 MgO 2,00 1,51 2,76 3,83 4,79 4,28 1,97 1,94 6,02 CaO 4,08 3,06 5,13 7,00 6,73 6,81 5,07 5,14 8,50 Na2O 3,38 2,91 3,06 3,07 3,26 2,90 3,65 3,98 2,40 K2O 3,58 4,41 3,02 2,17 2,09 2,17 2,77 2,45 5,02 TiO2 0,63 0,49 0,76 0,91 0,82 0,82 0,66 0,67 0,95 P2O5 0,15 0,11 0,17 0,21 0,18 0,18 0,23 0,21 1,06 MnO 0,06 0,06 0,08 0,1 0,11 0,11 0,10 0,10 0,14 LOI 0,5 2,6 1,2 0,5 0,5 1,2 1,2 1,1 3,6 Total 99,27 97,20 98,58 99,28 99,28 98,59 98,58 98,67 95,98 (ppm) Cr 48 41 41 48 123 75 103 Ni 26 49 0,9 0,6 33 Sc 10 8 11 18 19 22 10 11 30 Ba 725 642 675 657 574 560 649 595 1058 Co 9 6,4 12,2 18,2 21,9 17,2 10,3 9,4 33,2 Cs 10,3 20,3 6,2 2,4 2,4 3,6 2,3 1,7 1,9 Ga 18,4 17,9 19,9 19,9 18,2 19,9 18,6 18,1 15,5 Hf 6 7 5,4 4 4 4,3 4,6 4,7 3,8 Nb 12,6 13,9 8,7 8,8 7,3 8,6 8,1 8,4 8,2 Pb 4,4 1 1,8 1,4 2,2 2,3 0,7 0,5 22,2 Rb 169,9 228,4 115,4 60,7 68,4 69,2 81,4 75,9 185,0 Sr 408,9 265,3 484,2 563,8 528,6 535,8 493,8 429,8 1052,1 Ta 0,9 1,4 0,7 0,6 0,5 0,7 0,6 0,6 0,6 Th 21,1 26,2 15,9 6,1 7,3 9,9 8,1 7,1 4,7 U 6,5 9,6 5,2 1,3 2,1 3,2 1,9 1,8 1,4 V 103 82 122 186 155 173 92 94 250 Zn 34 17 33 34 31 26 41 38 97 Zr 208,7 220,6 170,4 148 139,1 144,5 160,8 155,6 147,2 Y 18,5 21,8 17,6 20,7 19,5 22,3 20,3 20,2 20 La 35,6 36,5 31,3 24,2 21,9 26,5 25,9 23,1 22,8 Ce 71,9 75,4 66 52,8 48,4 55,4 48,6 48,2 47 Pr 8,1 8,64 7,42 6,2 5,5 6,21 6,21 5,79 5,54 Nd 28,5 31,8 26,5 21,8 22,6 25,2 24,1 22,8 23,6 Sm 5,11 5,68 5,36 4,76 4,62 4,5 4,42 4,23 4,96 Eu 0,98 0,87 1,1 1,19 1,03 1,05 1,06 0,99 1,46 Gd 4,52 4,4 4,32 4,17 4,15 4,11 3,66 3,77 4,68 Tb 0,66 0,65 0,64 0,69 0,65 0,68 0,61 0,61 0,72 Dy 3,18 3,57 3,35 4,07 3,19 3,88 3,41 3,39 3,73 Ho 0,66 0,76 0,57 0,82 0,68 0,82 0,65 0,68 0,79 Er 1,82 2 1,71 1,93 1,98 2,1 1,97 1,82 1,9 Tm 0,26 0,31 0,24 0,31 0,3 0,38 0,31 0,31 0,31 Yb 1,56 1,88 1,48 1,66 1,54 2,11 2,11 1,94 1,63 Lu 0,25 0,31 0,22 0,26 0,22 0,32 0,28 0,27 0,3 24
398
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201224 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Table 2. ContinuedUnit Toconce Toconce Toconce Toconce Toconce Toconce Negro Negro Fm. LN (%) PAE-41 PAE-42 PAE-36 PAE-39 PAE-46 PAE-51 PAE-21 PAE-12 PAPA-46 SiO2 65,23 66,64 61,85 57,86 57,57 57,22 61,76 61,50 48,92 Al2O3 15,81 14,57 16,41 17,34 16,62 17,38 16,64 17,06 12,81 Fe2O3 4,35 3,44 5,34 6,79 7,11 6,72 5,73 5,62 10,16 MgO 2,00 1,51 2,76 3,83 4,79 4,28 1,97 1,94 6,02 CaO 4,08 3,06 5,13 7,00 6,73 6,81 5,07 5,14 8,50 Na2O 3,38 2,91 3,06 3,07 3,26 2,90 3,65 3,98 2,40 K2O 3,58 4,41 3,02 2,17 2,09 2,17 2,77 2,45 5,02 TiO2 0,63 0,49 0,76 0,91 0,82 0,82 0,66 0,67 0,95 P2O5 0,15 0,11 0,17 0,21 0,18 0,18 0,23 0,21 1,06 MnO 0,06 0,06 0,08 0,1 0,11 0,11 0,10 0,10 0,14 LOI 0,5 2,6 1,2 0,5 0,5 1,2 1,2 1,1 3,6 Total 99,27 97,20 98,58 99,28 99,28 98,59 98,58 98,67 95,98 (ppm) Cr 48 41 41 48 123 75 103 Ni 26 49 0,9 0,6 33 Sc 10 8 11 18 19 22 10 11 30 Ba 725 642 675 657 574 560 649 595 1058 Co 9 6,4 12,2 18,2 21,9 17,2 10,3 9,4 33,2 Cs 10,3 20,3 6,2 2,4 2,4 3,6 2,3 1,7 1,9 Ga 18,4 17,9 19,9 19,9 18,2 19,9 18,6 18,1 15,5 Hf 6 7 5,4 4 4 4,3 4,6 4,7 3,8 Nb 12,6 13,9 8,7 8,8 7,3 8,6 8,1 8,4 8,2 Pb 4,4 1 1,8 1,4 2,2 2,3 0,7 0,5 22,2 Rb 169,9 228,4 115,4 60,7 68,4 69,2 81,4 75,9 185,0 Sr 408,9 265,3 484,2 563,8 528,6 535,8 493,8 429,8 1052,1 Ta 0,9 1,4 0,7 0,6 0,5 0,7 0,6 0,6 0,6 Th 21,1 26,2 15,9 6,1 7,3 9,9 8,1 7,1 4,7 U 6,5 9,6 5,2 1,3 2,1 3,2 1,9 1,8 1,4 V 103 82 122 186 155 173 92 94 250 Zn 34 17 33 34 31 26 41 38 97 Zr 208,7 220,6 170,4 148 139,1 144,5 160,8 155,6 147,2 Y 18,5 21,8 17,6 20,7 19,5 22,3 20,3 20,2 20 La 35,6 36,5 31,3 24,2 21,9 26,5 25,9 23,1 22,8 Ce 71,9 75,4 66 52,8 48,4 55,4 48,6 48,2 47 Pr 8,1 8,64 7,42 6,2 5,5 6,21 6,21 5,79 5,54 Nd 28,5 31,8 26,5 21,8 22,6 25,2 24,1 22,8 23,6 Sm 5,11 5,68 5,36 4,76 4,62 4,5 4,42 4,23 4,96 Eu 0,98 0,87 1,1 1,19 1,03 1,05 1,06 0,99 1,46 Gd 4,52 4,4 4,32 4,17 4,15 4,11 3,66 3,77 4,68 Tb 0,66 0,65 0,64 0,69 0,65 0,68 0,61 0,61 0,72 Dy 3,18 3,57 3,35 4,07 3,19 3,88 3,41 3,39 3,73 Ho 0,66 0,76 0,57 0,82 0,68 0,82 0,65 0,68 0,79 Er 1,82 2 1,71 1,93 1,98 2,1 1,97 1,82 1,9 Tm 0,26 0,31 0,24 0,31 0,3 0,38 0,31 0,31 0,31 Yb 1,56 1,88 1,48 1,66 1,54 2,11 2,11 1,94 1,63 Lu 0,25 0,31 0,22 0,26 0,22 0,32 0,28 0,27 0,3 24
Annex VIII Appendix D
399
Geología 1:50.000 del área de la cadena…– Agosto, 2012 25 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 15. Diagrama TAS (sílice contra álcalis totales: K2O+Na2O) (Le Bas et al., 1986; Le Maitre et al., 1989) de las muestras de roca del Proyecto Paniri seleccionadas y analizadas. Como referencia se ha diagramado la muestra de la Formación Lomas Negras que corresponde a una traquiandesita basáltica de afinidad alcalina. Figura 16. Diagrama de sílice contra K2O (Peccerilo y Taylor, 1976) de las muestras analizadas del Proyecto Paniri. Simbología idéntica a la Figura 16. La muestra de la Formación Lomas Negras presenta un tan alto contenido de potasio que corresponde a una serie shoshonítica. 25
400
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201225 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 15. TAS diagram (total alkali vs. silica: K2O+Na2O) (Le Bas et al., 1986; Le Maitre et al., 1989) of the selected and analysed rock samples from the Paniri Project. By way of reference, the sample from the Lomas Negras Formation that corresponds to a basaltic trachyandesite with alkaline affinity, has been diagrammed.Figure 16. Diagram of silica vs. K2O (Peccerilo & Taylor, 1976) of the analysed samples from the Paniri Project. The Legend is identical to that of Figure 16. The sample from the Lomas Negras Formation presents such a high content of potassium that it corresponds to a shoshonitic series.
25
Annex VIII Appendix D
401
Geología 1:50.000 del área de la cadena…– Agosto, 2012 26 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 5.3.1Clasificación de rocas Las rocas seleccionadas de los centros volcánicos del Proyecto Paniri se caracterizan por un amplio intervalo de variación en su contenido de sílice (57 a 68% en peso) conformadas predominantemente por dacitas y andesitas (Figura 15). A continuación se describen los resultados en cada uno de los centros eruptivos ordenados de NE a SW en la alineación volcánica: Complejo Volcánico Paniri, domo Chao, Complejo Volcánico del León-Lagunita y el volcán Toconce. Además, al final se ha incluido al volcán Negro que corresponde a un estratovolcán del Mioceno Superior. a)Complejo Volcánico PaniriEste centro eruptivo de al menos 3 etapas de construcción está conformado dominantemente por dacitas e incluye dos andesitas y 1 traquiandesita (Figura 15). En términos de la evolución de su composición, la Etapa 1 está caracterizada por andesitas silíceas (una de ellas la traquinadesita) y dacitas poco evolucionadas (de <65% en peso de SiO2) y una dacita muy evolucionada (cerca del límite dacita-riolita). La Etapa 2 está representada por dacitas intermedias y una andesita poco evolucionada (cercana a los 57% en peso de SiO2). Finalmente, la Etapa 3 representada con sólo un análisis tiene una composición de dacita intermedia (65% en peso de de SiO2), similar a las dominantes de la segunda etapa evolutiva de este complejo volcánico. b)Domo ChaoEl domo torta Chao se caracteriza por resultados bastante similares prácticamente puntuales (66-68% en peso de SiO2) y, por lo tanto, todas las muestras analizadas de este domo corresponden a dacitas evolucionadas (Figura 15), una de las cuales es muy próxima al límite de las dacitas-riolitas, representando prácticamente a las rocas más evolucionadas de esta cadena volcánica. c)Complejo Volcánico del León-LagunitaEste complejo volcánico se ha construido en dos etapas evolutivas y está formado por dos grupos de composiciones, una de andesitas intermedias (alrededor de 60% en peso de SiO2) y una más evolucionada de dacitas y traquidacitas de entre 64 y 68% en peso de SiO2 (Figura 15). El centro eruptivo más antiguo, el volcán Lagunita, representaría un volcanismo bimodal (59 y 63-65% en peso de SiO2), mientras que, la segunda etapa evolutiva, el volcán del León, también representa un volcanismo bimodal (59-61 y 64% en peso de SiO2) que se traslapa en parte con el de la primera etapa caracterizado por andesitas silíceas y dacitas poco evolucionadas (cercanas al límite de los campos que dividen a las andesitas y las dacitas, 63% en peso de SiO2). d)Volcán ToconceEste centro eruptivo tiene tres grupos de composiciones bien definidos que se relaciona con las etapas evolutivas del edificio volcánico (Figura 15). Las rocas más antiguas de este volcán (Etapa 1) son andesitas poco evolucionadas (57-59% en peso de SiO2), mientras que, la Etapa 2 está representada por una andesita silícea (62% en peso de SiO2) y, finalmente, la Etapa 3 corresponde a dacitas que son las rocas más evolucionadas de este centro eruptivo (65-67% en peso de SiO2) y que sólo se han reconocido en la parte sur del edificio. Además, una de ellas, junto con muestras pertenecientes del CVP y el domo Chao, corresponden a las rocas más evolucionadas cercanas al límite que separa las dacitas y las riolitas. 26
402
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201226 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 5.3.1 Rock ClassificationThe selected rocks from the volcanic centres of the Paniri Project are characterised by a wide range of variation in its silica content (57 to 68% by weight), predominantly formed by dacites and andesites (Figure 15). There follows a description of the results for each of the eruptive centres arranged from NE to SW in the volcanic ridge: Paniri Volcanic Complex, Chao dome, Del León-Lagunita Volcanic Complex, and Toconce volcano. The Negro volcano, a stratovolcano from the Upper Miocene, was included at the end.a)Paniri Volcanic ComplexThis eruptive centre, formed in at least 3 evolutionary stages, is mainly constituted by dacites and includes 2 andesites and 1 trachyandesite (Figure 15). In terms of the evolution of its composition, Stage 1 is characterised by siliceous andesites (one being the trachyandesite) and little evolved dacites (of <65% by weight of SiO2) and one highly evolved dacite (close to the dacite-rhyolite limit). Stage 2 is represented by intermediate dacites and one little evolved andesite (nearly 57% by weight of SiO2). Lastly, Stage 3 is represented by a single analysis. It has an intermediate dacite composition (65% by weight of SiO2), similar to those prevailing in the second evolutionary stage of this volcanic complex.b)Chao DomeThe cake-shaped Chao dome is characterised by rather similar results (66-68% by weight of SiO2) and, thus, all the samples analysed from this dome correspond to evolved dacites (Figure 15), one of which is very close to the dacite-rhyolite limit, representing essentially the most evolved rocks of this volcanic chain.c)Del León-Lagunita Volcanic ComplexThis volcanic complex was formed in two evolutionary stages and is made up of two composition groups: one of intermediate andesites (approximately 60% by weight of SiO2) and one more evolved of dacites and trachydacites (between 64 and 68% by weight of SiO2) (Figure 15). The older eruptive centre, the Lagunita volcano, represents bimodal volcanism (59 and 63-65% by weight of SiO2), while the second evolutionary stage, the Del León volcano, also represents bimodal volcanism (59-61 and 64% by weight of SiO2) that partially overlaps with that of the first stage, and is characterised by siliceous andesites and poorly evolved dacites (close to the field limits that divide andesites and dacites, 63% by weight of SiO2).d)Toconce VolcanoThis eruptive centre has three well-defined composition groups relating to the evolutionary stages of the volcanic edifice (Figure 15). The oldest rocks of this volcano (Stage 1) are poorly evolved andesites (57-59% by weight of SiO2), while Stage 2 is represented by one siliceous andesite (62% by weight of SiO2). Lastly, Stage 3 corresponds to dacites that are the most evolved rocks in this eruptive centre (65-67% by weight of SiO2) and that have only been identified in the south part of the edifice. In addition, one of them – together with samples from the CVP and the Chao dome – corresponds to the most evolved rocks close to the limit which separates dacites and rhyolites.
26
Annex VIII Appendix D
403
Geología 1:50.000 del área de la cadena…– Agosto, 2012 27 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl e)Volcán NegroEste estratovolcán antiguo y erosionado se caracteriza por una composición prácticamente puntual de andesitas silíceas (62% en peso de SiO2). 5.4Tierras raras Los patrones de las tierras raras (REE) de las muestras de roca seleccionadas normalizados a condrita (valores de Sun y McDonough, 1989) son bastante similares y prácticamente paralelos, caracterizado por un enriquecimiento en las REE ligeras (LREE) y un progresivo empobrecimiento en las REE pesadas (HREE) con una evidente anomalía negativa de Eu (Figura 17). Los patrones idénticos de las REE señalan un origen común para los magmas que dieron origen a las rocas, mientras que, los patrones paralelos de REE indican que las rocas valores de REE más altos son el resultado fraccionamiento mineral como proceso dominante, en particular la anomalía negativa de Eu indica que el fraccionamiento de la plagioclasa es dominante en el proceso de diferenciación magmática. Figura 17. Diagrama de REE normalizado a contrita (valores de Sun y McDonough, 1989) de muestras de roca seleccionadas del Proyecto Paniri. Domo Chao (color azul), los complejos volcánicos Paniri (color verde) y del León-Lagunita (color rojo) y el volcán Toconce (color violeta). 27
404
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201227 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl e)Negro VolcanoThis ancient eroded stratovolcano is characterised by a composition that is practically specific to siliceous andesites (62% by weight of SiO2).5.4 Rare Earth ElementsThe patterns of rare earth elements (REE) of the rock samples, normalised to chondrite (values from Sun & McDonough, 1989), are quite similar and practically parallel. They are characterised by an enrichment of the light REE (LREE) and a progressive depletion of the heavy REE (HREE) with an evident negative anomaly in Eu (Figure 17).The identical REE patterns indicate a common origin for the magmas that gave rise to the rocks, whereas the parallel REE patterns indicate that the rocks with higher REE values resulted from mineral fractionation as a predominant process. The negative anomaly of Eu in particular indicates that the fractionation of the plagioclase prevails in the magmatic differentiation process.Figure 17. Diagram of REE normalised to chondrite (values from Sun & McDonough, 1989) of selected rock samples from the Paniri Project. Chao dome (blue), the volcanic complexes of Paniri (green) and Del León-Lagunita (red), and the Toconce volcano (violet).
27
Annex VIII Appendix D
405
Geología 1:50.000 del área de la cadena…– Agosto, 2012 28 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 5.5Diagrama de elementos incompatibles Los patrones de los elementos incompatibles normalizado al MORB (Pearce, 1983) de las muestras de roca seleccionadas de los centros volcánicos del sector del Proyecto Paniri son altamente similares, caracterizado por un enriquecimiento en elementos incompatibles móviles (Sr, K, Rb y Ba) y un paulatino empobrecimiento en los elementos incompatibles inmóviles (Ta a Yb) con anomalías negativas en Nb, P y Ti (Figura 18). Figura 18. Diagrama de elementos incompatibles normalizado al MORB (Pearce, 1983) de las muestras de roca seleccionadas los centros volcánicos del Proyecto Paniri. Es común que el tipo de patrón de elementos incompatibles sea interpretado como típico de magmas de márgenes convergentes donde los magmas se originan por la incorporación de fluidos derivados de la deshidratación de fondo oceánico subducido (enriquecido en elementos incompatibles móviles) y por la fusión del manto o cuña astenosférica (empobrecimiento en elementos incompatibles inmóviles) (Pearce, 1983). La razón constante de las concentraciones de La e Yb (LREE y HREE, respectivamente) a medida que evolucionan los magmas que dan origen a las rocas indica que el principal proceso en la diferenciación magmática corresponde al fraccionamiento mineral. Resulta evidente que esto no se cumple en el caso de la muestras seleccionadas de los centros eruptivos del Proyecto Paniri (Figura 19), por lo tanto, es posible identificar la existencia de otro proceso diferente como responsable de la evolución magmática. 28
406
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201228 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 5.5 Diagram of Incompatible ElementsThe patterns of MORB-normalised incompatible elements (Pearce, 1983) of the selected rock samples from the volcanic centres in the Paniri Project sector are very similar, characterised by an enrichment in mobile incompatible elements (Sr, K, Rb, and Ba) and a gradual depletion in immobile incompatible elements (Ta to Yb) with negative anomalies in Nb, P, and Ti (Figure 18). Figure 18. Diagram of MORB-normalised incompatible elements (Pearce, 1983) of the selected rock samples from the volcanic centres of the Paniri Project.It is common for the type of incompatible elements’ pattern to be interpreted as typical of magmas of convergent margins where the magmas originate due to the incorporation of fluids derived from the dehydration of the subducted ocean floor (enriched in mobile incompatible elements) and due to the melting of the mantle or asthenosphere wedge (depletion in immobile incompatible elements) (Pearce, 1983).The constant concentrations of La and Yb (LREE and HREE, respectively) as the magmas that give rise to the rock evolve, indicate that the main process in magmatic differentiation corresponds to mineral fractionation. It is evident that this does not happen in the case of the selected samples from the eruptive centres of the Paniri Project (Figure 19), therefore it is possible to identify the existence of a different process responsible for the magmatic evolution.
28
Annex VIII Appendix D
407
Geología 1:50.000 del área de la cadena…– Agosto, 2012 29 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 19. Diagrama de variación de sílice contra la razón de las concentraciones de las tierras raras La e Yb de las muestras analizadas de los centros eruptivos del área del Proyecto Paniri. Simbología idéntica a la Figura 16. 2929
408
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201229 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 19. Diagram of variation in silica versus the ratio of concentrations of rare earth elements La and Yb in the analysed samples from the eruptive centres in the Paniri Project area. The legend is identical to that of Figure 16.
29
Annex VIII Appendix D
409
Geología 1:50.000 del área de la cadena…– Agosto, 2012 30 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 6Discusión y conclusiones La cadena volcánica Paniri-Toconce de orientación NW-SE (N130°E) se extiende por más de 20 km y está comprendida entre el Complejo Volcánico Paniri (CVP) y el volcán Toconce. Esta cadena volcánica se caracteriza por una actividad principalmente efusiva durante el Pleistoceno, dominada por dacitas de anfíbola y biotita (63-68% en peso de SiO2) y, subordinadamente por andesitas de piroxeno±anfíbola±biotita (57-63% en peso de SiO2), a excepción de volcán Negro que corresponde a estratovolcán andesítico (62% en peso de SiO2) del Mioceno Superior (6 Ma) que solo está relacionado espacialmente a esta estructura volcánica, al situarse inmediatamente al norte de la misma. La etapa inicial de construcción de esta cadena volcánica ocurre en la parte SE de la misma dando origen a coladas de lava del estratovolcán Lagunita (Figura 20). Posteriormente, ocurre una migración de la actividad hacia la parte NW de la cadena dando inicio a la construcción del edificio más antiguo del CVP (ca. 1,4 Ma). Alrededor de unos 1 Ma la actividad comienza el CVLL (volcán Lagunita) y prácticamente sincrónica con emisión de coladas de lava pertenecientes al volcán Toconce. Después, poco antes de unos 600 ka ocurre volcanismo sincrónico en los complejos volcánicos Paniri (CVP II) y del León-Lagunita. Finalmente, hace unos 400 ka ocurre un volcanismo simultáneo en al menos tres puntos de la cadena volcánica: el estratovolcán del León, el domo Chao y el CVP. Una característica que destaca en el volcanismo tardío de la cadena volcánica Paniri-Toconce es la migración hacia al NW de las fuentes (Figura 20). El volcanismo más joven asociado a esta cadena volcánica corresponde a rocas pertenecientes al CVP que sellarían la actividad volcánica de este centro eruptivo de una edad de 150-163 ka (Figura 20), no obstante, alrededor de unos 100 ka atrás se habría producido la efusión de las etapas tardías del domo Chao (muy probablemente la etapa Chao III) como indican los resultados de Tierney et al. (2010). En particular, el CVP se habría construido durante, al menos, tres etapas (1400-600, 400-200 y 163-150 ka) (Figura 20), de las cuales la primera de ellas es dominada por andesitas y andesitas silíceas de piroxeno (61-64% en peso de SiO2) y es subordinada por dacitas de anfíbola y biotita (68% en peso de SiO2). En la segunda etapa la fuente migra hacia el NW y las rocas evolucionan principalmente a andesitas silíceas y dacitas de piroxeno±anfíbola±biotita (64-65% en peso de SiO2) a excepción de coladas andesíticas (57% en peso de SiO2) que se encuentran estratigráficamente en la parte media de esta etapa. En la tercera y última etapa de este complejo volcánico (CVP III) no hay una variación composicional significativa (65% en peso de sílice) en las dos coladas de lava dacíticas que se distribuyen hacia el NE del cono que sella la actividad volcánica de este complejo, no obstante, existe una leve migración de la fuente hacia el SE. Por su parte, la evolución de los dos estratovolcanes que conforman el Complejo Volcánico del León-Lagunita es distinta. Por un lado, el volcán Lagunita (CVLL 1, 1600-600 ka) (Figura 20), el más antiguo de los dos, corresponde a una secuencia de coladas de lava andesíticas de piroxeno (59% en peso de SiO2) intercaladas por andesitas silíceas y dacitas de piroxeno±anfíbola±biotita (63-66% en peso de SiO2). Mientras que, el volcán del León (390-265 ka) (Figura 20) tiene andesitas y andesitas silíceas de piroxeno±anfíbola±biotita (59-60% en peso de SiO2) en el flanco SW y andesitas silíceas y dacitas de anfíbola±biotita±piroxeno (61 y 64% en peso de SiO2) en el flanco sur. Estas últimas rocas parecen ser la litología dominante en el resto del edificio volcánico. 30
410
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201230 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 6 Discussion and ConclusionsThe Paniri-Toconce volcanic chain, having a NW-SE trend (N130°E) spans more than 20 km and runs between the Paniri Volcanic Complex (CVP) and the Toconce volcano. This volcanic chain is characterised by a mainly effusive activity during the Pleistocene, dominated by dacites of amphibole and biotite (63-68% by weight of SiO2) and, subordinately by andesites of pyroxene±amphibole±biotite (57-63% by weight of SiO2). The exception is the Negro volcano which is an andesitic stratovolcano (62% by weight of SiO2) from the Upper Miocene (6 Ma) and is only related spatially to this volcanic structure, as it is located immediately north of said structure.The initial stage of this volcanic chain took place in its SE part, originating the lava flows of the Lagunita stratovolcano (Figure 20). Later the activity migrated to the NW part of the chain, commencing the formation of the oldest edifice of the CVP (ca. 1.4 Ma). Around 1 Ma, the activity begins at the CVLL (Lagunita volcano) and is practically synchronous with the emission of lava flows from the Toconce volcano. Then, just before 600 ky there is synchronous volcanism at the volcanic complexes of Paniri (CVP II) and Del León-Lagunita. Lastly, about 400 ky ago there was simultaneous volcanism in at least three points along the volcanic chain; the Del León stratovolcano, the Chao dome, and the CVP. One characteristic that stands out during the late volcanism of the Paniri-Toconce volcanic chain is the migration to the NW of the sources (Figure 20).The younger volcanism associated to this volcanic chain corresponds to rocks belonging to the CVP that would end the volcanic activity of this eruptive centre, dated to approximately 150-163 ky (Figure 20). However, about 100 ky ago the effusion of the later stages of the Chao dome (very likely the Stage Chao III) would have taken place, as indicated by the results of Tierney et al. (2010).In particular, the CVP would have been created over, at least, three stages (1400-600, 400-200, and 163-150 ky) (Figure 20). The first stage was dominated by andesites and siliceous pyroxene andesites (61-64% by weight of SiO2) and subordinated dacites of amphibole and biotite (68% by weight of SiO2). During the second stage, the source migrated toward the NW and the rocks evolved mainly into siliceous andesites and dacites of pyroxene±amphibole±biotite (64-65% by weight of SiO2), except for andesitic lava flows (57% by weight of SiO2) that are stratigraphically found in the middle part of this stage. In the third and last stage (CVP III) there was no significant compositional variation (65% by weight of silica) in the two dacitic lava flows that spread toward the NE of the cone ending the volcanic activity of this complex: however there was a slight migration of the source toward the SE.The evolution of the two stratovolcanoes that form the Del León-Lagunita Volcanic Complex was different. On the one hand, the Lagunita volcano (CVLL 1, 1600-600 ky) (Figure 20), the older of the two, corresponds to a sequence of pyroxene andesite lava flows (59% by weight of SiO2) interspersed with siliceous andesites and dacites of pyroxene±amphibole±biotite (63-66% by weight of SiO2). On the other hand, the Del León volcano (390-265 ky) (Figure 20) contains andesites and siliceous andesites of pyroxene±amphibole±biotite (59-60% by weight of SiO2) on the SW side, and siliceous andesites and dacites of amphibole±biotite±pyroxene (61 and 64% by weight of SiO2) in the south side. The latter rocks seem to be the prevailing lithology in the rest of the volcanic edifice.
30
Annex VIII Appendix D
411
Geología 1:50.000 del área de la cadena…– Agosto, 2012 31 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figura 20. Diagrama de síntesis de la evolución temporal y geoquímica de la cadena volcánica Paniri-Toconce. CVP: Complejo Volcánico Paniri, CVLL: Complejo Volcánico del León-Lagunita. El cuadrado de color azul corresponde a datación Ar/Ar en anfíbola, círculo de color rojo en MF y rombo de color verde en roca total. Las fechas de color negro señalan recarga de magma más primitivo (y más caliente) a la cámara magmática. En el borde derecho las cifras en color negro indican los tiempos de reposo, mientras que en color anaranjado se representan los periodos eruptivos. Las rocas más jóvenes (350-240 ka y alrededor de 100 ka según Tierney et al., 2010) que sellarían el volcanismo en esta área y que constituyen además, los productos más evolucionados de esta cadena volcánica, corresponden a las dacitas y riodacitas de anfíbola y biotita del domo Chao (66-68% en peso de SiO2). La 31
412
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201231 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Figure 20. Summary diagram of the time and geochemical evolution of the Paniri-Toconce volcanic chain. CVP: Paniri Volcanic Complex, CVLL: Del León-Lagunita Volcanic Complex. The blue square corresponds to the Ar/Ar dating in amphibole; the red circle to the GM; and green rhombus to total rock. The dates in black indicate the more primitive (and hotter) magma recharge to the magma chamber. In the right margin, the figures in black indicate the resting times, while the eruptive periods is shown in orange. The younger rocks (350-240 ky and approximately 100 ky according to Tierney et al., 2010) that would finish the volcanism in this area and also constitute the more evolved products of this volcanic chain, correspond to the dacites and rhyodacites of amphibole and biotite of the Chao dome (66-68% by weight of SiO2). The
31
Annex VIII Appendix D
413
Geología 1:50.000 del área de la cadena…– Agosto, 2012 32 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl petrografía, la geoquímica y la edad de estas rocas son prácticamente idénticas a las pertenecientes al domo Chillahuita que se localiza inmediatamente al NE del volcán Toconce. Además, existen rocas más antiguas pero similarmente evolucionadas como las pertenecientes a los domos anteriores, las cuales corresponden a dacitas de anfíbola±biotita (66-68% en peso de SiO2) pertenecientes a los complejos volcánicos Paniri (CVP I) y del León-Lagunita (CVLL I) y el volcán Toconce (etapa III). Inmediatamente al NE del domo Chao, el volcán Negro, un estratovolcán erosionado del Mioceno Superior (unos 6 Ma), está constituido por andesitas porfídicas de piroxeno de una composición intermedia (62% en peso de SiO2) de entre los productos menos y más evolucionados de la cadena volcánica NW-SE (57 y 68% en peso de SiO2, respectivamente) y señalarían el volcanismo más antiguo asociado a esta área. Otra idea a destacar que se desprende del modelo elaborado de la evolución de la cadena volcánica Paniri-Toconce (Figura 20) es que los periodos eruptivos son relativamente más cortos comparados con los tiempos de reposo o de inactividad volcánica, resultado coherente con lo formulado en otros volcanes de la Zona Volcánica Central caracterizados por erupciones episódicas separadas por tiempos de quietud prolongados (por ejemplo, Clavero et al., 2004; Polanco et al., 2009; Correa et al., 2012). La coexistencia de varias texturas de desequilibrio en las rocas estudiadas, como por ejemplo, la coexistencia de plagioclasas con y sin textura de cedazo, la textura de zonación oscilatoria, la presencia de inclusiones máficas, el cuarzo con textura embahiamiento, señalan no sólo una dinámica continua de los magmas que dieron origen a las rocas de los distintos centros eruptivos de la cadena volcánica NW-SE, sino también que el proceso de mezcla de magma ha jugado un rol preponderante en la evolución de estas rocas, pero además, ha sido un mecanismo recurrente donde pulsos nuevos de magma más primitivo y caliente invaden la zona de almacenamiento de magma, rompiendo el equilibrio y generando una convección magmática más intensa. Este proceso se infiere también por la química de sus productos en al menos dos etapas de la evolución de la cadena volcánica Paniri-Toconce (Figura 20), aproximadamente a unos 600 y 300 ka en los complejos volcánicos del León-Lagunita y Paniri, respectivamente. Además, la razón variable de los contenidos de elementos trazas incompatibles Yb y La parece ser coherente con la ocurrencia de otro proceso de diferenciación magmático, además de la cristalización fraccionada. Los diagramas de variación de Harker de elementos mayores de las muestras de la zona de estudio (Anexo 3)evidencian una alta dispersión lo que por un lado puede ser correlacionado con el alto contenido defenocristales de las rocas pero también puede indicar que las rocas no son co-genéticas. A pesar de lo anterior, las variaciones de algunos elementos en estos diagramas indicarían el fraccionamiento de algunas fases minerales que si se han reconocido. Así por ejemplo, la disminución del contenido de Ti y V a medida que aumenta el contenido de sílice señalan el fraccionamiento de los óxidos de Fe-Ti. De la misma manera, la disminución del Cr a medida que evolucionan las rocas indica el fraccionamiento de clinopiroxeno. Además, el punto de inflexión en el diagrama de Harker del contenido de Ba (alrededor de 64% en peso de SiO2) puede ser el reflejo del fraccionamiento de la anfíbola y/o la biotita. Ahora bien, la disminución de los contenidos de Ni y Co a medida que aumenta el contenido de sílice indicarían el fraccionamiento de olivino, mineral que no ha sido reconocido en la rocas estudiadas. El patrón de elementos trazas incompatibles en el diagrama de multi-elementos de Pearce (1983) de las muestras de roca analizadas del Proyecto Paniri son interpretados como típicos de rocas calco-alcalinas de zona de subducción, donde la fuente de las rocas es una mezcla de fluidos provenientes del manto astenosférico (elementos trazas incompatibles inmóviles) y de la deshidratación de los sedimentos incorporados en la Placa de 32
414
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201232 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl petrography, geochemistry, and age of these rocks are practically identical to those from the Chillahuita dome that is located immediately NE of the Toconce volcano. In addition, there are older rocks that evolved similarly to those from the previous domes, which correspond to dacites of amphibole±biotite (66-68% by weight of SiO2) belonging to the volcanic complexes of Paniri (CVP I) and Del León-Lagunita (CVLL I), and to the Toconce volcano (Stage III).Immediately NE of the Chao dome lies the Negro volcano, an eroded stratovolcano from the Upper Miocene (about 6 Ma). It is constituted by porphiric pyroxene andesites of an intermediate composition (62% by weight of SiO2) among the less and more evolved products of the NW-SE volcanic chain (57 and 68% by weight of SiO2, respectively): these rocks represent the oldest volcanism associated with this area.Another conclusion that follows from the proposed model of the evolution of the Paniri-Toconce volcanic chain (Figure 20) is that the eruptive periods are relatively short, when compared to the resting times or periods of volcanic inactivity. This result is consistent with what has been found in relation to other volcanoes of the Central Volcanic Zone characterised by episodic eruptions separated by extended periods of inactivity (e.g., Clavero et al., 2004; Polanco et al., 2009; Correa et al., 2012).The coexistence of several textures of imbalance in the studied rocks, such as the coexistence of plagioclase with and without sieve texture, the oscillatory banding, the presence of maphic inclusions, quartz with embayment, not only indicate the continuous dynamism of the magma that gave rise to the rocks of the different eruptive centres of the NW-SE volcanic chain. The above also indicates that the magma mixing process played a major role in the evolution of these rocks, and that it also was a recurring mechanism where new pulses of more primitive and hotter magma invaded the magma storage area, breaking the balance and generating more intense magma convection. This process is also inferred by the chemistry of its products in at least two stages of the evolution of the Paniri-Toconce volcanic chain (Figure 20), approximately at 600 and 300 ky at the volcanic complexes of Del León-Lagunita and Paniri, respectively. Furthermore, the varying ratio in the content of incompatible trace elements Yb and La seems to be consistent with the occurrence of another magmatic differentiation process, as well as the fractioned crystallisation.The Harker variation diagrams of major elements of the samples from the studied areas (Annex 3) reveal a high dispersion, which can be correlated with the high phenocryst content in the rocks but can also indicate that the rocks are not co-genetic. Despite this, the variations in some elements in these diagrams would indicate the fractionation of some mineral phases that have been identified. For example, the decrease in Ti and V content as the silica content increases point to the fractionation of the Fe-Ti oxides. Similarly, the decrease in Cr content as the rocks evolve indicates the fractionation of clinopyroxene. In addition, the inflection point in the Harker diagram of the Ba content (around 64% by weight of SiO2) may be a reflection of the fragmentation of the amphibole and/or biotite. The decrease in Ni and Co content as the silica content increases would indicate the fractionation of olivine, a mineral that has not been identified in the rocks analysed.The pattern of incompatible trace elements in the Pearce multi-element diagram (1983) of the analysed rock samples have been interpreted as typical for calc-alkaline rocks from subduction zones, where the rock source is a mixture of fluids coming from the asthenospheric mantle (immobile incompatible trace elements) and from the dehydration of sediments incorporated in the Nazca Plate
32
Annex VIII Appendix D
415
Geología 1:50.000 del área de la cadena…– Agosto, 2012 33 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Nazca (elementos trazas incompatibles móviles). En particular, los altos valores en los elementos incompatibles móviles (Sr, K, Rb y Ba) son indicativos de contaminación cortical, lo que es consistente con el potente espesor cortical que deben atravesar los magmas en la Zona Volcánica Central, segmento volcánico en el cual se localiza este proyecto y que ha sido extensamente descrito por estudios en otros volcanes de este segmento o provincia volcánica (por ejemplo, Stern, 2004; Stern et al., 2007). Por otro lado, el patrón de las tierras raras de las muestras de roca analizadas es indicativo de la ocurrencia de procesos de cristalización fraccionada en la evolución magmática de las rocas, jugando un importante rol el fraccionamiento de la plagioclasa como señala la anomalía negativa de Eu. Además, el fraccionamiento de otras fases minerales (clinopiroxeno, anfíbola, biotita y óxidos de Fe-Ti) se puede inferir a partir de algunas variaciones de elementos trazas en los diagramas de Harker. En conclusión, las rocas de los centros eruptivos de la cadena volocánica NW-SE que se extiende desde el CVP al volcán Toconce corresponden a dacitas y andesitas calco-alcalinas de alto contenido de potasio. Las rocas más primitivas de las cadenas corresponden a andesitas de piroxeno pertenecientes al volcán Toconce y la etapa 2 del CVP. Por el contrario, las rocas más evolucionadas son dacitas de anfíbola±biotita de los domos Chao y Chillahuita, volcán Toconce y a los complejos volcánicos Paniri y del León-Lagunita. Las evidencias de terreno, petrográficas y geoquímicas señalan procesos de cristalización fraccionada, mezcla de magmas y contaminación y asimilación cortical en la evolución de los magmas que dieron origen a las rocas. En particular, los mecanismos de inyección de magmas más primitivos y más calientes a las zonas de acumulación es recurrente como indican la ocurrencia de andesitas de piroxeno en distintas etapas de la evolución de diferentes centros eruptivos pero en especial por los numerosos y abundantes enclaves máficos que se encuentran en las rocas estudiadas. 33
416
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201233 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl (mobile incompatible trace elements). In particular, the high values of mobile incompatible elements (Sr, K, Rb, and Ba) are indicative of crust contamination, which is consistent with the very large crustal thickness through which the magmas in the Central Volcanic Zone must pass in which this Project is located and that has been described extensively by studies of other volcanoes in this volcanic province (for example, Stern, 2004; Stern et al., 2007).In addition, the pattern of the rare earth elements of the analysed rock samples is indicative of fractioned crystallisation processes occurring in the magmatic evolution of the rocks, the plagioclase fractionation playing an important role as indicated by the negative anomaly of Eu. Moreover, the fractionation of other mineral phases (clinopyroxene, amphibole, biotite, and Fe-Ti oxides) can be inferred from some variations of trace elements in the Harker diagrams.In conclusion, the rocks from the eruptive centres of the NW-SE volcanic chain that spans from the CVP to the Toconce volcano correspond to high-potassium calc-alkaline andesites and dacites. The more primitive rocks from the chain correspond to pyroxene andesites belonging to the Toconce volcano and to Stage 2 of the CVP. In contrast, the more evolved rocks are dacites of amphibole±biotite from the Chao and Chillahuita domes, the Toconce volcano, and the volcanic complexes of Paniri and Del León-Lagunita. The field, petrographic, and geochemical evidence points to fractioned crystallisation processes, magma mixture, and crust contamination and assimilation during the evolution of the magmas that gave rise to the rocks. In particular, the mechanisms that injected more primitive and hotter magma into the accumulation zones were recurrent, as indicated by the occurrence of pyroxene andesites in various stages of the evolution of the different eruptive centres – but especially by the numerous and abundant maphic inclusions found in the studied rocks.
33
Annex VIII Appendix D
417
Geología 1:50.000 del área de la cadena…– Agosto, 2012 34 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 7Referencia Baker M.C.W., Francis P.W. 1978. Upper Cenozoic volcanism in the central Andes; age and volumes. Earth Planetary Science Letter, 41(2):175-187. Clavero, J.E., Sparks, R.S.J., Pringle, M.S., Polanco, E., Gardeweg, M. (2004) Evolution of Taapaca volcanic Complex, Central Andes of Northern Chile. Journal of the Geological Society, vol. 161, 603–618. Correa N., Morata D., Clavero J., Arcos R. 2012. Evolución geológica, geocronológica y petrológica del Complejo Volcánico Quimsachata-Aroma, Altiplano de Iquique, Chile. En: Actas digitales del XIII Congreso Geológico de Chile, Sesión Temática No. 4 de Volcanología y Geotermia, 459-461. Antofagasta. de Silva L.S., Self S., Francis P.W., Drake R.E., Ramirez C.F. 1994. Effusive silic volcanism in the Central Andes: The Chao dacite and other young lavas of the Altiplano-Puna Volcanic Complex. Journal of Geophysical Research, 29(B9):17805-17825. González O. 1995. Volcanes de Chile. Instituto Geográfico Militar, Santiago, 640 p. Irvine T.N., Baragar W.R.A. 1971. A guide to chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences, 8, 523-548. Jarvis K. 1988. Inductively coupled plasma mass spectrometry: a new technique for the rapid or ultra-trace level determination of the rare-earth elements in geological material. Chemical Geology, 68, 31-39. Le Bas M.J., Le Maitre R.W., Streckeisen A., Zanettin B. 1986. A chemical classification of volcanic rocks based on the total alkali–silica diagram. Journal of Petrology, 27(3): 745-750. Lechler P.J., Desilets M.O. 1987. A review of the use of loss on ignition as a measurement of total volatiles in whole rock analysis. Chemical Geology, 63(3-4): 341-344. Marinovic N., Lahsen A. 1984. Hoja Calama, Región de Antofagasta, escala 1:250.000. Carta Geológica de Chile, No. 58. Instituto de Investigaciones Geológicas, 140 p. Mular A.L., Halbe D.N., Barratt D.J. 2002. Mineral processing plant design, practice, and control proceedings, 2422 p. Pearce, J.A. 1983. The role of subcontinental lithosphere in magma genesis at active continental margins. In C.J. Hawkesworth, M.J. Norry (eds.), Continental basalts and mantle xenoliths. Nantwich, Shiva, 230-249. Peccerillo R., Taylor S.R. 1976. Geochemistry of Eocene calc-alkaline volcanic rocks from the Kastamonu area, Northern Turkey. Contribution of Mineralogy and Petrology, 58, 63-81. Polanco E., Clavero J., Gimeno D., Fernández-Turiel J.L. 2009. Procesos de mezcla de magmas y/o autoconvección en el Complejo Volcánico Taapaca (18ºS), Andes Centrales: evidencias texturales y de química mineral. En: Actas digitales del XII Congreso Geológico de Chile, Simposio No. 7 de Volcanología Física: del Ascenso Magmático a los Procesos Eruptivos y su Interacción con el Entorno (S7_024), Santiago. Rickwood P.C. 1989. Boundary lines within petrologic diagrams which use oxides of major and minor elements. Lithos, 22, 247-263. Stern C.R. 2004. Active Andean volcanism: its geologic and tectonic setting. Revista Geológica de Chile, 31, 161-208. Stern C.R., Moreno H., López-Escobar L., Clavero J., Lara L., Naranjo J.A., Parada, M.A., Skewes, M.A. 2007. Chilean volcanoes. In: Moreno T., Gibbons W. (Eds.), Geology of Chile, The Geological Society of London, London, 147-178. Sun S.S., McDonough W.F. 1989. Chemical and isotopic systematic of oceanic basalts: implications for mantle compositions and processes. In: Saunders A.D., Norry M.J. (Eds.), Magmatism in oceanic basins. Geological Society of London. Special Publication, 42, 313-345. 34
418
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201234 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl 7References Baker M.C.W., Francis P.W. 1978. Upper Cenozoic volcanism in the central Andes; age and volumes. Earth Planetary Science Letter, 41(2):175-187. Clavero, J.E., Sparks, R.S.J., Pringle, M.S., Polanco, E., Gardeweg, M. (2004) Evolution of Taapaca volcanic Complex, Central Andes of Northern Chile. Journal of the Geological Society, vol. 161, 603–618. Correa N., Morata D., Clavero J., Arcos R. 2012. Evolución geológica, geocronológica y petrológica del Complejo Volcánico Quimsachata-Aroma, Altiplano de Iquique, Chile. En: Actas digitales del XIII Congreso Geológico de Chile, Sesión Temática No. 4 de Volcanología y Geotermia, 459-461. Antofagasta. de Silva L.S., Self S., Francis P.W., Drake R.E., Ramirez C.F. 1994. Effusive silic volcanism in the Central Andes: The Chao dacite and other young lavas of the Altiplano-Puna Volcanic Complex. Journal of Geophysical Research, 29(B9):17805-17825. González O. 1995. Volcanes de Chile. Instituto Geográfico Militar, Santiago, 640 p. Irvine T.N., Baragar W.R.A. 1971. A guide to chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences, 8, 523-548. Jarvis K. 1988. Inductively coupled plasma mass spectrometry: a new technique for the rapid or ultra-trace level determination of the rare-earth elements in geological material. Chemical Geology, 68, 31-39. Le Bas M.J., Le Maitre R.W., Streckeisen A., Zanettin B. 1986. A chemical classification of volcanic rocks based on the total alkali–silica diagram. Journal of Petrology, 27(3): 745-750. Lechler P.J., Desilets M.O. 1987. A review of the use of loss on ignition as a measurement of total volatiles in whole rock analysis. Chemical Geology, 63(3-4): 341-344. Marinovic N., Lahsen A. 1984. Hoja Calama, Región de Antofagasta, escala 1:250.000. Carta Geológica de Chile, No. 58. Instituto de Investigaciones Geológicas, 140 p. Mular A.L., Halbe D.N., Barratt D.J. 2002. Mineral processing plant design, practice, and control proceedings, 2422 p. Pearce, J.A. 1983. The role of subcontinental lithosphere in magma genesis at active continental margins. In C.J. Hawkesworth, M.J. Norry (eds.), Continental basalts and mantle xenoliths. Nantwich, Shiva, 230-249. Peccerillo R., Taylor S.R. 1976. Geochemistry of Eocene calc-alkaline volcanic rocks from the Kastamonu area, Northern Turkey. Contribution of Mineralogy and Petrology, 58, 63-81. Polanco E., Clavero J., Gimeno D., Fernández-Turiel J.L. 2009. Procesos de mezcla de magmas y/o autoconvección en el Complejo Volcánico Taapaca (18ºS), Andes Centrales: evidencias texturales y de química mineral. En: Actas digitales del XII Congreso Geológico de Chile, Simposio No. 7 de Volcanología Física: del Ascenso Magmático a los Procesos Eruptivos y su Interacción con el Entorno (S7_024), Santiago. Rickwood P.C. 1989. Boundary lines within petrologic diagrams which use oxides of major and minor elements. Lithos, 22, 247-263. Stern C.R. 2004. Active Andean volcanism: its geologic and tectonic setting. Revista Geológica de Chile, 31, 161-208. Stern C.R., Moreno H., López-Escobar L., Clavero J., Lara L., Naranjo J.A., Parada, M.A., Skewes, M.A. 2007. Chilean volcanoes. In: Moreno T., Gibbons W. (Eds.), Geology of Chile, The Geological Society of London, London, 147-178. Sun S.S., McDonough W.F. 1989. Chemical and isotopic systematic of oceanic basalts: implications for mantle compositions and processes. In: Saunders A.D., Norry M.J. (Eds.), Magmatism in oceanic basins. Geological Society of London. Special Publication, 42, 313-345. 34
Annex VIII Appendix D
419
Geología 1:50.000 del área de la cadena…– Agosto, 2012 35 Cerro El Plomo 5630, Piso 14 Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Tierney C., de Silva S.L., Schmitt A.K., Jicha B., Singer B.S. 2010. U-series in zircon and 40Ar/39Ar geochronology reveal the most recent stage of a supervolcanic cycle in the Altiplano-Puna Volcanic Complex, Central Andes. American Geophysical Union (AGU), Fall Meeting, Abstract V41B-2277. San Francisco, California. 35
420
Annex VIII Appendix D
Geology at Scale 1:50,000 of the Paniri-Toconce Volcanic Chain Area …– Aug. 201235 Cerro El Plomo 5630, 14th Floor Las Condes, Santiago de Chile Tel: 798-7170 www.energiandina.cl Tierney C., de Silva S.L., Schmitt A.K., Jicha B., Singer B.S. 2010. U-series in zircon and 40Ar/39Ar geochronology reveal the most recent stage of a supervolcanic cycle in the Altiplano-Puna Volcanic Complex, Central Andes. American Geophysical Union (AGU), Fall Meeting, Abstract V41B-2277. San Francisco, California. 35
Annex VIII Appendix D
421
422
Annex VIII
APPENDIX E
SERNAGEOMIN, Geology of the Silala River Area, 2017
(Original in English)
(Large-sized version of this map can be found in Volume 6)
Annex VIII Appendix E
423
424
Annex VIII Appendix E
Annex VIII Appendix E
425
426
Annex VIII
APPENDIX F
SERNAGEOMIN, Simplified Compilation of the Geology in the Silala River Area, 2017
(Original in English)
(Large-sized version of this map can be found in Volume 6)
Annex VIII Appendix F
427
428
Annex VIII Appendix F
Annex VIII Appendix F
429
430
Annex VIII
APPENDIX G
Sellés, D. and Gardeweg, M., Geology of the Area Ascotán - Inacaliri Hill. SERNAGEOMIN, Geological Map of Chile,
Basic Geology Series, Santiago (in edition)
(Full sized map can be found in Volume 6)
Annex VIII Appendix G
431
IN EDITIONbb“LEGENDCENOZOIC QUATERNARYHOLOCENELOWERSTRATOVOLCANOES Products of volcanic products (
d)Domes and lava flows extent no greater than (fp)Block-and-ash deposits, blocks having prismatic Volcanoes and Lavas Recent-looking volcanic breccia. Correspond mainly cone and lava flow called Pabellón.(d) Domes and lava flows, del Bajo) and hornblende (a) Coalesced sequences (fd) Detritus flow deposits (e) Scoria cone of La Poruña, (“aa”, basaltic Poruña scoria cone.(Mud flow deposits.(Rubble breccia at the Volcanoes and Lavas flows, lava domes Azufre, Aguilucho, and Pyroxene andesite (Avalanche of the (Avalanche of the (Avalanche of the (Avalanche of the Volcanoes of the Lower system, predominantly interspersed tuff and breccia. the Cañapa hill, the domes and Aguilucho volcanoes.Reddish and black Dacite domes with (Volcanic avalanche Volcanoes of the Lower Volcanic systems, predominantly interspersed tuff and breccia. Ascotán-Barrancane volcanoes Reddish and black Dacite domes with Volcanoes of the Lower edifices having composed of andesitic hornblende and biotite Reddish and black Volcanoes and Volcanic Domes, lava domes, lavas systems. The centres the southern foot of the an andesitic composition always present) biotite Dacitic and rhyolitic (Deposits of volcanic of San Pedro and Sifón, Sub-volcanic body Volcanic and Volcanoclastic Discontinuous sequence very thick breccia, with Discrete outcrops, small Station, in Bayos hillocks, Relicts of dacitic to (Deposits of polymictic volume unwelded tuffs, Strata of the Continental, folded volcano-conglomerates, and coarse having an intermediate PlsvPlmvPlivPivDEPOSITS HanHlHfHaPlHpcPlHsPlHlPlHrmPlHePlacPlHacMsipMrmMsispPlgHgrHvUPPERPLEISTOCENE MIDDLEPlacvLOWERPladPsvMsvMmvMidPliiaPPl(Observed / inferred geological contact Intra-formational contactIndeterminate observed faultInferred / covered faultNormal fault observed / inferred Reverse fault observed / inferredCovered reverse faultPhoto-lineamentObserved / inferred topographic caldera margin Observed / inferred volcanic crater Craterless volcanic eruption centreLava flow and direction of flowLevee Glacial cirque Morainic crestAlignment of volcanic eruption centresMass wasting escarpmentTopographic escarpmentPaleo-coastline Direction and dip angle of strataMetallic deposits and prospectsRock and industrial ore depositsCorrelative deposit number in Tables 9 and 10, text Geological profile lineAA'!!!EE¡¹11í®®®(a)(d)(((((((((((((((((((((d)(a)(av1)(e)(b)((((((((((((((((d)(a)(av)(d)(fp)(a)(d)(av)(sv)5/ab)(l)(av4)6.5005.5004.5003.5002.500A’ A5.5004.5003.5002.5001.5006.5001.500B5.5004.5003.5002.5001.5006.5006.5005.5004.5003.5002.500B’’ 1.500Plmv (av2) PlacMsipMsisMrmPlacvMidMsv(sv)Cerrodel DiabloPlmv (av1) PivPlHacPlHsSalar de AscotánPlivPlsv(e)PlsvHv(d)MsispMsv(av)MmvPlsv(a)Vn. CarasillaDomo AscotánVn. AraralVn. San PedroVn. CentralCordón InacaliriPlacMsvPivPliv Mmv (d)Pliv(av)PliiaPlacMmvMsisMsipPlHacPliv(d)PlgPlg(1)PlHacPliiaMsisMsvPsvB’ MsisMmvBreccia((((((((((((((((((((((((((((Inacaliri10569PLIOCENE MIOCENE UPPERUPPERNEOGENE PivPivLacustrine PlHsPlHsPlHlPlHlPlHrmPlHrmPlHePlHePlacPlacPlHacPlHacPlgPlgPlacvPlacvPsvPsvLOWERMsvMsvMmvMmvMIDDLE(a)(d)(((((((((((((((((((((((((((((((a)(d)(av)(sv)(d)(b)(a)PliPli)PlgPlHarr11g11,612±0,01,094±0,0770770444557555555n. CerroVn. CeVn. CePivPivPlHsPlHsSalar AscotánSalar AscotánVn. CarasillaVn. AscotánDomo AscotánPivPivPliv Pliv PliiaPliiaPlacPlacMmvMmvMsisMsisMsipMsipLEGENDSandstoneTuffSandstone/conglomerateAndesitic lavaDacitic domeGravelFault, indicates normal movement directionVertical scale 1:100.000Horizontal scale 1:100.000Vertical scale 1:100.000Horizontal scale 1:100.000Vulcanic complexm.a.s.l.m.a.s.l.m.a.s.l.m.a.s.l.SCHEMATIC PROFILESSOUTH ASTRONÓMICOGEOLOGÍA DEL ÁREA ASCOTAN - CERRO INACALIRISERVICIO NACIONAL DE GEOLOGÍA Y MINERÍAESCALA 1:100.000MAPA LEYENDACENOZOICOCUATERNARIOPLIOCENOMIOCENOHOLOCENOSUPERIORSUPERIORINFERIORNEÓGENOPLEISTOCENOESTRATOVOLCANES volcánicos volcánicos Domos y coladas domo con un alcance no superior Depósitos de bloques claro, conabundantesbloquescondiaclasamiento y lavas del Secuencias Cerro Pabellón. y dacíticos de hornblenda ab) Lavas "aa" andesítico La Poruña(l) Depósitos de lahares(b) (d) av1) Avalancha av2) Avalancha av3) Avalancha av4) Avalancha Apacheta, coladas distales de los volcanes Azufre Domos, son de composición andesítica y no siempre presentes (d) Domos dacíticos y av) Depósitos de avalanchas San Pedro y Sifón así (sv) Cuerpo subvolcánico Secuencias volcánicas Secuencia discontínua gruesas, con bloques discretos, pequeños y en Cerritos Bayos, al (d) Relictos de domos b) Depósitos de brechas soldadas de pequeño Estratos del Cerro del Secuencia volcano-sedimentaria conglomerados finos exclusivamente volcánicos, PlsvPlmvPlivPivDEPÓSITOS aluvialHanHlHfHaPlHpcPlHsPlHlPlHrmPlHePlacPlHacMsipMrmMsispPlgHgrHvSUPERIORMEDIOPlacvINFERIORPladPsvINFERIORMsvMmvMidMEDIOPliiaContacto geológico observado/inferidoContacto intraformacionalFalla observada indeterminadaFalla inferida/cubiertaFalla normal observada/inferidaFalla inversa observada/inferidaFalla inversa cubiertaFotolineamientoMargen topográfico de caldera observado/inferidaCráter volcánico observado/inferidoCentro de emisión volcánico sin cráterColada de lava y dirección de flujoLevéeCirco glacialCresta de morrenaAlineamiento de centros de emisión volcánicosEscarpe remoción en masaEscarpe topográficoPaleolínea de costaRumbo y manteo de estratosYacimientos y prospectos metálicosYacimientos de rocas y minerales industrialesNúmero correlativo de yacimiento en tablas 9 y al textoTraza de perfil geológicoAA'((®(av2)(av3)(d)(fd)(d)(b)"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""EEEEEEEEEEEEEEEEEEE!!!!!!EEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE!!!!!!!!EEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE((((((!!!!!!))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))®®®®))))))))))))¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¹////¬""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""PivPlHsPliv(av)PlivPlgPivPsvPsvPsvMsvHaHv(fp)PlacPlsvPlHacHiePlacPlivPiv(a)PlHpcHaMsv(Plmv(d)PlivPlsvPlHacPlHacPlmvPlHacPlHacPlsvPlgPlmvMsvPlHacPlacPlmvPlg1Plsv(a)PlmvPlmvPlHacPiv(a)PlHacPlHacPsvMsispPlgPlHpcMsvMsvMsv(d)Plg1MsvPlHacHaPliv(PlacPladPlmvHaPlmv(d)PlivPlsv(d)Msv(d)PlacPsv(a)Pliv(d)Psv(a)PlHacPlmv(av1)HiePlgMsv(d)PivPlHacPlmvPlsv(d)PliiaMsvPlHpcMmv(d)Mmv(d)Psv(a)PlacMsv(d)Msv(d)PlgPlgPlivHaPlHl(d)PlHacPlgPivHaPlacPlgPlHacMidPliv(a)Plmv(av2)PladPliv(a)Plsv(d)PlgHv(d)PlHacPliv(d)Plsv(fd)MsvPlHpcMsv(d)MsvPlgPiv(a)PlgMmvHaPlgHaMsipPlHacPlivPlHacPsv(a)Pliv(av)Plg1Plmv(av1)Plsv(d)PlgPlHacHlPlHacPlgPlacMsvPlacHaHfPsvPlHacPsvPsv(d)PlgPlHacHfPsvPlacPlHpcPlgPlHacPlmvPlHacPPl(pc)PlgMsispPlgMmv(d)Msv(d)Plg1PlgPPl(pc)Psv(d)HaPlHacPsvPlHacPlHacPlmvPlHpcPladHaPlgPlHacPlHacPlgPlHacMsipPlmvPsv(a)PlgPlivMsv(av)PlgPlacPlacPlacMmv(d)MsvPlacvPsv(d)PlmvPlHpcPlacMsipMsipMsv(d)HaMmvPlHacPlgPlHacPlHacPlHpcPlHacPlgPlHacPsvPlHpcPliv(d)PlHacPsv(d)Pliv(d)PlgPlgPlacPlacPlg1PlmvPlgPlmv(av1)PlHpcPlHpcPlHacPlHpcPlgPlgPlgPPl(pc)PsvPlHpcPlHlHaPlHpcMsv(d)Msisp(c)PlgMsipMmvPPl(pc)PsvPlmv(d)Pliv(av)Msv(d)MsipPlHacPlivPliv(av)PlgPlmvHaPlHpcPlmv(av3)HlPlHpcPladHfPlHpcPlHacPlHpcPlgMsipMmsrmHfPlmvHaPlHpcMmv(b)Plsv(d)PlHacPlg1Psv(av)Plsv(fd)PlHacPlgHfPliv(a)PlgPlHpcMsv(PlHpcHaPlHpcPlacPlgPlgPlmvMsv(av)Msv(d)Plsv(l)PlacHaPlgMsisPPl(pc)HaPlHacPlg1PlgPlHpcPivPlHpcPlgPlHrmPlsv(b)Psv(av)PlHpcMsisPlHacHfHaMsvPlHpcPliiaMsvPsvHaPlHacPlHacMsisPlHpcPlHacPlgHiePPl(pc)HaPlHpcPsvPlHpcHgrPlgMmvPlHpcPlHpcPlHacMsvPlHrmHfPlHpcPlsvHaHgrPliv(av)Msv(av)PlacPlHacPlivMmv(d)PlHacHgrPlHacPlHpcHaPlHacPliv(d)PlHacPsvHfPlHl(d)PlHpcHanHgrPlacHieHfPliv(av)PlacHaPlHpcPlHacPlHl(d)Plmv(av2)Msv(d)MsipPlHl(d)MsvHlPlmvHaPlsv(ab)HgrPlacMsisPlHacPlHpcHanHfPlHacMsv(d)PlHacPlacPlHacPlHl(d)Plsv(e)HaMsisPlsv(b)PlHlPsvPsvPlsv(e)HgrHfPlg1HgrPlivPlHacPlHacPsvMsipHgrPlHlMsvPlHacHgrHaMsvPPl(pc)PlHacPsv(a)HfPsvHgrPlsv(l)HfMsvPlHacHfPlacPlHpcHgrHfMsisPlivPlHacHaPlmv(l)HanHfHanPlHacHgrMsv(d)Pliv(fp)PlHpcHgrPlgHgrPlmvHgrPPl(pc)HgrPlHl(HgrHgrPliv(d)HgrPliv(av)HaHanPlHpcHgrPlHacPlHacPlHpcMmvHgrHgrHgrPliiaHfHaHgrHgrMsv(d)PlHpcPliv(av)HgrHgrHaHgrMsisPlHacPliiaMsv(sv)PlacMsisHgrHgrHanMsisHfPlHacPliv(av)HgrHgrMsisPlHlMsvHgrHgrMsipMsipHgrHgrPlg1HgrPlg1Plg1Plg1Plg1PlgPlg1Plg1Plg1PlgPlg1Plg1Plg1Plg1Plg1Plg1Plg1Plg1Plg1Plg1Plg1PlgPlgPliiaPlHpcMsv(d)PlHacPlivPlgMsv(d)Msisp(c)Mmv(b)Mmv(b)PlsvHiePlg1Plmv(av4)Plmv(av4)Plg1PlgPlg1PlHacPlsv(a)PlHac11,0±0,5 (*) (clt) 316,5 ± 0,4 10,331±0,033 100,16±0,05 106,51± 0,04 17,18±0,13 18,11 ± 0,06 110,3±0,5 (*) 310,9±0,5 (*) 31,92±0,03 10,1198±0,0054 910,65±0,14 40,605±0,013 10,7±0,2 115,4±0,5 116,7±0,3 117,5±0,6 111,739±0,016 10,91±0,14 111,43±0,20 10,097±0,014 100,09±0,08 100,068±0,027 100,096±0,036 107,8 ± 0,4 (*) 20,13±0,04 100,14±0,04 100,166±0,016 100,114±0,037 118,6±0,2 103,81±0,3 53,36±0,13 53,15±0,15 50,1586±0,0031 1011,30±0,03 1037,550±0,425 1015,510±0,125 105,563±0,015 100,105±0,025 83,9±0,2 61,1±0,2 (*) 22,75±0,04 12,67±0,02 13,73±0,08 15,101±0,034 114,39±0,35 10,652±0,012 10,560±0,040 12,614±0,024 10,05±0,01 71,612±0,018 11,109 ±0,023 12,54±0,23 17,30±0,35 1<1 16,852±0,076 18,1±0,7 114,668±0,088 11,024±0,033 10,517±0,023 11,094±0,016 113,04±0,18 10,166±0,012 100,139±0,033 106,6±0,1 125,80±0,02 123,65±0,30 5H>!;!!!>HH>>>"H"!HHH!H;;;H;;"H"HHQRS>"";H"HHHHHHHH>HH;JJ-!>HH!HHHJJ!ahahahahahahahahahah32198765432119171422212016181513121110de45004750CHILECHILESalarSalarPAMPA APACHETAPedroCorralQuebrada SantanaBOLIVIABOLIVIAEL LUNARQuebradaLAGUNITAAguiluchoEL VEINTEBarrancaneCo. ChancaPortezueloLa CachimbaCo. LailaiCUEVA NEGRALA ESTANCIACo. PabellónCo. PabellónCo. ColoradoCUEVA CHICADel EstanquePASO DEL INCACo. AguiluchoCUEVA GRANDELOS MORRILLOSPAMPA PERDIZCo. del DiabloPAMPA LAMPAYACebollar ViejoCo. del AzufrePAMPA LAGUNITACo. Tres PuntasEstación AscotánESTANCIA MORRILLOSEl Ojo de San PedroCo. Cebollar ViejoLA ABRA DE SAN PABLOFALDAS DE BARRANCANEPAMPA LA CACHIMBAS a l a r d e A s c o t á nCERRITOS BAYOSQuebrada CuevitasQuebradaLandaQuebradaCebollaroLaRepresaS a l a r d e A s c o t á nHITO S/N LXIXPort. PalpanaCo.de las CuevasPort. El PabellónCo. CebollarEstación CebollarRINCON RÍO BLANCOQuebradaSotojunoCerro AraralHITO S/N L-XX PORTEZUELODE ASCOTÁN O DEL JARDÍNCo. Ascotán o del JardínOjos de AscotánCo. CarasillaCo. PabelloncitoCo. PocaumoCo. PolapiPaulinaCo. NegroQuebradaRíoBlancoQuebradaPabellónCo. PabellónQuebradadelCerroNegroQuebradaLaCejaQuebradaLaPaulinaQuebradaGuallarumoQuebradaLasCuevasQuebradaLaTurberaLa TurberaRepresaCASA QUEMADAQuebradaCasaQuemadaQuebradaCuestillaQuebradaCasaQuemadaQuebradaPolapiEst. PolapiAvestruzCo. ChinchillaQuebradadelIncaHITO S/N L-XX PORTEZUELODEL INCA O BARRANCANECERROS DEL INCAO BARRANCANEQuebradadelHitoQuebradaBarrancaneCllo. PlomoQuebradaAzufreradelBajoQuebradaChac-IncaCERROSDELINCAOBARRANCANEQuebradaLaPerdízHITO S/N LXXIICo. ApachetaQuebradaAguiluchoLay LayCampamentoNuevoQuebradaElEncantoQuebradaLaLavaRÍOSANPEDRORÍOSANPEDROQuebradaSanPedroQuebradaElOnceCo. RedondoPAMPA BLANCAQuebradaTreintayUnoCorralCERROSDECOLANACORDÓNDEINACALIRICORDÓNDEINACALIRIAA'B'B''B2121B-145303016103726397038163718380339054073391439653969405540383909415837183820422041934419399740063993400939204186414537164156381137543937409837474125421141094219397441954142376041834112402439813855412639484023407039843815402739614153394837803973401143163965405837954095396039704018398940604097409839464272398737194103382044614781414237954367425751164463419452963719372547774810482650145688396352625418419157164967373053315253559851665530556240105403372046755291379638065342549650684563380654055205374343725190566844255524560843985475482544465783373437283725594940205260422752124260519244085165523045044214513742824022373041383724462038034030470754705478514351584813497745803730475548664182544049754703511543204016514750405260524451374980411752164621504050104469373048333978451845154028457145984655456545584062431037354788379845804808388146494094401046614916406438564875387950044410420741324164490437154171395449633676493640255068393043574312528747654061386642103968510636184465494638843905457236444685422148185120455444405516402350544840451743095582584645993554554556753887556147174769377838614748381238685453392643954697430956244247544245803802499255485440507152924345375256724536380937545573405635804560503040554995563148214923411040884548410755133985475747544829439047234905377555814956473045813830436046214530471239304327475842765335375543553734453144404970523048214523493338605283472652744830419345064466522038545157486848354510468050855192494054304918552036095493442747114649533447604921527848054539378242646092597447394556529857524660594861454479346247704487604750225076477743974756422058694278374540655140499549644764478953734791405045084337480849244064444252654198493047645380381642644598467942143930391052124922355242545373533039025361538151633887367040305476542638845397385952643949373054463654386451173866395349965373362154085342409451475385350540475422486137755360362337703948400853305020383935713814558456283822369438483848418243804239440938614407386236323872467246623876Vn. AraralVn. AscotánDomoAscotánDomoPabelloncito bajoVn.CarasillaVn.PolapiVn.CebollarVn. CerroLas CuevasVn. CerroInacaliriComplejo VolcánicoCordón InacaliriDomosCachimbaVn. ApachetaDomo PabellónVn. AguiluchoDomos Chac IncaVn. AzufreDomos ChancaVN. SAN PABLOVN. CENTRALVN. SAN PEDROVn. BarrancaneVn. Cañapa68°7570km22°55'758050'759045'760040'761035'7620km21°30'7570km22°55'758050'759045'760040'761035'7620km21°30'68°30'25'560km20'57010'15'580590600km5'590600km68°5'58010'15'560km57020'25'68°30'PERFILES ESQUEMÁTICOSm s.n.m.6.5005.5004.5003.5002.500A’Am s.n.m.5.5004.5003.5002.5001.5006.5001.500Bm s.n.m.5.5004.5003.5002.5001.5006.500m s.n.m.6.5005.5004.5003.5002.500B’’ Hv(fp2)PlmvPlmvPlHacMsv(d)CentralComplejo VolcánicoCordón (d)PlHacPliiaMsisMsvPsvEscala vertical 1:100.000Escala horizontal 1:100.000Escala vertical 100.000Escala horizontal 1:100.000B’ MsisMmvSIMBOLOGÍADomo dacíticoBrechaArenisca(((Arenisca/conglomeradoLava andesíticaTobaGravaFalla, indica sentido de desplazamiento normal(((Inacaliri10569EN EDICIÓNEN EDICIÓNPLIOCENOEN EDICIÓNPLIOCENOMIOCENOEN EDICIÓNMIOCENOEN EDICIÓNSUPERIOREN EDICIÓNSUPERIORSUPERIOREN EDICIÓNNEÓGENOEN EDICIÓNSecuencias volcánicas EN y tobas intercaladas. Corresponden EDICIÓNy Pedro, el cono de escoria EDICIÓNPedro, Domos y coladas domo EDICIÓNDomos Secuencias de flujos Depósitos de flujos EDICIÓNDepósitos Cono de escorias olivinoEN EDICIÓNCono EDICIÓNLavas Brechas de escombro domosEN EDICIÓNBrechas y lavas del MedioEN EDICIÓNVolcanes de lava, lavas losEN EDICIÓNColadas San Pablo, Azufre, EDICIÓNvolcanes Domos andesítico hornblendaEN de la PolapiEN EDICIÓNAvalancha del volcán PedroEN del volcán PaniriEN del volcán AguiluchoEN del Pleistoceno InferiorEN volcánicos predominantemente EDICIÓNAparatos además de brechas y EDICIÓNademás EDICIÓNApacheta, EDICIÓNde (a) EDICIÓN(Secuencias rojizas (d) Domos dacíticos de anfíbolaEN av) Avalancha volcánica AzufreEN del Plioceno volcánicos predominantemente además de brechas y alineamiento de volcanes EDICIÓNalineamiento (a) Secuencias rojizas (d) Domos dacíticos de del Plioceno InferiorEN volcánicos EDICIÓNEdificios predominantemente por EDICIÓNpredominantemente hornblenda y biotita y EDICIÓNhornblenda (a) Secuencias rojizas Volcanes y secuencias SuperiorEN lavas domo, EDICIÓNDomos, variablemente desmantelados. EDICIÓNvariablemente Lay y Colorado, al pie EDICIÓNLay EDICIÓNson flujoEN EDICIÓNSan EDICIÓNCuerpo PivEN EDICIÓNPivGlaciares EDICIÓNGlaciares EDICIÓNGravas EDICIÓNperenne, caídaEN EDICIÓNprincipalmente salinosEN EDICIÓN lacustresEN cuencasEN EDICIÓNrepresadas masaEN constituidosEN EDICIÓNpor eólicosEN coluvialesEN EDICIÓNcuyos EDICIÓNen glacialesEN abundanteEN EDICIÓNmatriz EDICIÓNamorrillada EDICIÓNmorrenas jóvenesEN EDICIÓNIgnimbrita EDICIÓNca. EDICIÓNDepósito EDICIÓNreciente PleistocenoEN quebradasEN EDICIÓNaluviales EDICIÓNinactivas EDICIÓNalgunos PlHsEN EDICIÓNPlHsPlHlEN EDICIÓNPlHlPlHrmEN EDICIÓNPlHrmPlHeEN EDICIÓNPlHePlacEN EDICIÓNPlacPlHacEN EDICIÓNPlHacPlgEN EDICIÓNPlgPlacvEN EDICIÓNPlacvEN EDICIÓNPsvEN EDICIÓNPsvINFERIOREN EDICIÓNINFERIORMsvEN EDICIÓNMsvMmvEN EDICIÓNMmvMEDIOEN EDICIÓNMEDIOEN EDICIÓN((((((((((EDICIÓNHieEN EDICIÓN""EDICIÓN""EDICIÓN"""""""""""""""""""EDICIÓNEEN EDICIÓNEEEN EDICIÓN)))EDICIÓN))))))EDICIÓN))))EDICIÓN)EDICIÓN¡EDICIÓN¡¡¡EDICIÓN¡"¡"EDICIÓN"¡"¡¡EDICIÓN"¡""¡"EDICIÓN¡)))¡)))EDICIÓN)))¡))))¡)EDICIÓN)¡)EDICIÓN"¡"E¡EEN EDICIÓNE¡EE¡E¡EDICIÓN¡¡¡¡EDICIÓN¡¡¡¡¡¡EDICIÓN¡¡¡"¡"¡"¡"EDICIÓN"¡"¡"¡""Msv(d)EDICIÓNMsv(Pliv(EDICIÓNPliv())))))))PlacEN EDICIÓNPlacMsvEN EDICIÓNMsvMsvEN EDICIÓNMsvPlacEN EDICIÓNPlacPlgEN EDICIÓNPlgPlgEN EDICIÓNPlgPlHacEN EDICIÓNPlgMsv(d)Msv(HaEN EDICIÓNHaHaEN EDICIÓNHaEN EDICIÓNPlg1EN EDICIÓNPlg1PlgEN EDICIÓNPlg"Plg"PlHacEN EDICIÓNPlHacPlHacEN EDICIÓNPlHacHaEN EDICIÓNHaHgrEN EDICIÓNHgrHgrEN EDICIÓNHgrPlg1EN EDICIÓNPlg1PlHl(d)EDICIÓNPlHl(HgrEN EDICIÓNPlg1Plg1EN EDICIÓNPlg1PlHacEN EDICIÓNPlHacEN EDICIÓN1,612±11,612±EDICIÓN1,094±1EN 11,094±1HEN EDICIÓNHHEN EDICIÓNHEN EDICIÓN16EN EDICIÓN1616EN EDICIÓN16CEN EDICIÓNCEEN EDICIÓNEREN EDICIÓNRREN EDICIÓNROEN EDICIÓNOSEN EDICIÓNSDEN EDICIÓNDEEN EDICIÓNECEN EDICIÓNCOEN EDICIÓNOLEN EDICIÓNL¡L¡AEN EDICIÓNA¡A¡A¡¡NEN EDICIÓNNNEN EDICIÓNN¡N¡N¡¡EDICIÓNAAEN EDICIÓNACEN EDICIÓNOOEN EDICIÓNOREN EDICIÓNRDEN EDICIÓNDDEN EDICIÓNDÓEN EDICIÓNÓÓEN EDICIÓNÓNEN EDICIÓNNDEN EDICIÓNEIEN EDICIÓNINEN EDICIÓNNAEN EDICIÓNCCEN EDICIÓNCAEN EDICIÓNALEN EDICIÓNLIEN EDICIÓNIREN EDICIÓNRIEN EDICIÓNICEN EDICIÓNIIEN EDICIÓNLLEN EDICIÓNIB''EDICIÓNB''4539EN EDICIÓN45394539EN EDICIÓN45394770EN EDICIÓN47704770EN EDICIÓN4770E4770EEN EDICIÓNE4770EE4770EEN EDICIÓNE4770E4777EN EDICIÓN47774756EN EDICIÓN47564756EN EDICIÓN4756E4756EEN EDICIÓNE4756EE4756EEN EDICIÓNE4756E5373EN EDICIÓN53735373EN EDICIÓN5373¡5373¡5373¡¡4808EN EDICIÓN48084808EN EDICIÓN48085265EN EDICIÓN52655265EN EDICIÓN5265E5265EEN EDICIÓNE5265EE5265EEN EDICIÓNE5265E¡5380¡5380¡¡5212EN EDICIÓN52125212EN EDICIÓN52125373EN EDICIÓN53735330EN EDICIÓN53305361EN EDICIÓN53615381EN EDICIÓN53815381EN EDICIÓN53815163EN EDICIÓN51635476EN EDICIÓN54765476EN EDICIÓN5476¡5476¡5476¡¡5476¡)))¡)))5476)))¡)))EDICIÓN)))¡)))5476)))¡))))¡)5476)¡)5426EN EDICIÓN54265426EN EDICIÓN54265397EN EDICIÓN53975397EN EDICIÓN53975264EN EDICIÓN52645264EN EDICIÓN5264¡5264¡5264¡¡5446EN EDICIÓN54465117EN EDICIÓN51175117EN EDICIÓN51174996EN EDICIÓN49965373EN EDICIÓN53735408EN EDICIÓN54085408EN EDICIÓN54085342EN EDICIÓN53425342EN EDICIÓN53425147EN EDICIÓN51475147EN EDICIÓN5147¡5147¡5147¡¡5147¡¡¡¡5147¡¡¡EDICIÓN¡¡¡5147¡¡¡¡¡¡5385EN EDICIÓN53855385EN EDICIÓN53855422EN EDICIÓN54225422EN EDICIÓN54224861EN EDICIÓN48615360EN EDICIÓN53605330EN EDICIÓN5330¡5020¡5020¡¡5584EN EDICIÓN5584"5584"EDICIÓN55845584EN EDICIÓN55845628EN EDICIÓN56285628EN EDICIÓN5628"5628"5628""5628EN EDICIÓN56284182EN EDICIÓN41824182EN EDICIÓN41823861EN EDICIÓN38614672EN EDICIÓN46724662EN EDICIÓN46623876EN EDICIÓN3876EN EDICIÓNVn. CerroEN Cerro"Vn. CerroInacaliriEN EDICIÓNInacaliriInacaliriEN EDICIÓNInacaliri5330Inacaliri5330EN EDICIÓN5330Inacaliri5330InacaliriEN EDICIÓNInacaliriEN EDICIÓN55'7580EN EDICIÓN758050'EDICIÓN50'10'EDICIÓN10'590EN EDICIÓN5905'EDICIÓN5'ESQUEMÁTICOSEN EDICIÓNPERFILES EDICIÓNPivEN EDICIÓNPlHsEN EDICIÓNSalar AscotánEN CarasillaEN EDICIÓNVn. EDICIÓNDomo EDICIÓNPliv EDICIÓNMmv EDICIÓNPliv(EDICIÓNPliiaEN EDICIÓNPlacEN EDICIÓNMmvEN EDICIÓNMsisEN EDICIÓNMsipEN EDICIÓNNational Geology and Mining ServiceScale 1:100.000GEOLOGY OF THE AREA ASCOTÁN – INACALIRI HILL
432
Annex VIII Appendix G
IN EDITIONbb“AGREEMENT BETWEEN THE REPUBLIC OF CHILE AND THE REPUBLIC OF ARGENTINA TO SPECIFY THE ROUTE OF THE BOUNDARY FROM MOUNT FITZ ROY UP TO DAUDET HILL”. (Buenos Aires, 16 December 1998)GEOLOGY OF THE AREAASCOTÁN – INACALIRI HILLREGION OF ANTOFAGASTADaniel Sellés M.Moyra Gardeweg P.BASIC GEOLOGY SERIESGEOLOGICAL MAP OF CHILENo. xxxScale 1:100.000201XNATIONAL SUB-DIRECTORATE OF GEOLOGYISSN xxxx-xxxx90°CHILEAN ANTARTICTERRITORY53°SOUTH POLEDistance between contour lines: 50 mScale 1:100.0001000 m 0 1 2 3 4 5 km Magnetic declination (January 2015)Magnetic northAstronomic north LOCATION MAPBibliographical Reference Sellés M., D.1; Gardeweg P., M.1, N.1,2016. Geología del área Pampa Lirima-Cancosa, Región de Tarapacá [Geology of the Pampa Lirima-Cancosa area, region of Tarapacá]. National Geology and Mining Service (SERNAGEOMIN), Geological Map of Chile, Basis Geology Series, XXX:XX p., 1 map, scale 1:100,000. Santiago.1 Aurum Consultores-Servicios Geológicos y Mineros Ltda. [email protected]; [email protected] 0717-7283 Registration No.271.810© Servicio Nacional de Geología y Minería [National Geology and Mining Services], Av. Santa María 0104PO. Box 10465, Santiago, Chile. National Director (Deputy):Mario Pereira A.National Sub-director of Geology (Deputy): Omar Cortés C.All rights reserved, its reproduction is forbidden.EditionHead of Edition Committee: Renate Wall Z.Edition Committee: Rodrigo Carrasco O., Aníbal Gajardo C., Jorge Muñoz B., Andrew Tomlinson Editors: Constantino Mpodozis M., Jorge Muñoz B.Head of Publications: Soraya Amar N.Head of Geological Information Systems Unit (USIG, in Spanish): Constanza Casanova de L.Rules usedGeological Time Scale: Gradstein, F.M.; Ogg, J.G.; Schmitz, M.D.; Ogg, G.M. (Editors) 2012Topographic BasisMaps at scale 1:50,000, Ascotán, Cerro Araral (hill), Volcanes San Pedro & San Pablo (volcanoes), Cerro Inacaliri aka del Cajón (hill), from Instituto Geográfico Militar [Military Geographic Institute] (Chile), modified.Geodesic ReferenceUniversal Transversal Mercator (UTM) Projection, Zone 19S, SIRGASScientific and Technical SupportPetrographic studies: the AuthorsFeasibilities of radiometric datings: Eugenia Fonseca P. and the AuthorsK-Ar radiometric determinations: Marcelo Yáñez B. and César Vásquez B. 40Ar/39Ar Carlos Pérez de Arce R.Geochronology lab of SERNAGEOMINU-Pb LA-ICP-MS Radiometric determinations: Dirk Frei, Stellenbosch University, South Africa, and Víctor Valencia, Washington State University, USAU-Pb SHIRMP Radiometric determinations: Mark Fanning, Australian National University, AustraliaDigital production: Luis Delcorto A., Aurum Consultores Ltda.Financial SupportSector Funds from SERNAGEOMIN, National Geology Plan“The circulation thereof is authorized, in terms of the maps and quotes that this work contains, referring or related to the international boundaries and borders of the national territory, by Resolution N° 126 of October 13, 2016 of the National Department of State Borders and Boundaries (DIFROL).The edition and circulation of maps, geographic maps or other printed matter and documents that refer to or are in connection with the boundaries and borders of Chile, do not in any way implicate the Chilean State in conformity with Article N° 2, letter g) of Statutory Decree N° 83 of 1979 of the Ministry of Foreign Affairs”.LEGENDCENOZOIC QUATERNARYHOLOCENELOWERSTRATOVOLCANOES AND VOLCANIC SEQUENCESVolcanic Products of the HoloceneHolocene volcanic products from the San Pedro volcano.(d) Domes and lava flows of hornblende and biotite dacites with subordinate pyroxenes, reduced in volume, extent no greater than 2.5 km from the peak of the present-day edifice of the San Pedro volcano.(fp) Block-and-ash deposits, massive, poorly sorted, unwelded, light grey supported matrix, with abundant blocks having prismatic jointing caused by the collapsing of flow domes and domes.Volcanoes and Lavas from the Upper Pleistocene Recent-looking volcanic sequences, comprising andesitic to dacitic lavas and domes, with interstratified tuffs and breccia. Correspond mainly to dacitic lavas of hornblende and pyroxene from the San Pedro volcano, the scoria cone and lava flow called La Poruña, and the dacitic domes called Pabellón, Chac Inca, Chanka and Cerro Pabellón.(d) Domes and lava flows, biotite and hornblende dacite composition (Pabellón, Chac Inca, Chanca, Azufrera del Bajo) and hornblende and pyroxene dacite composition (San Pedro volcano).(a) Coalesced sequences of scoria flows, tephra and thin lava flows, reddish to black, andesitic.(fd) Detritus flow deposits associated with Chanka domes.(e) Scoria cone of La Poruña, basaltic andesite with olivine, and andesitic composition with pyroxene and olivine.(“aa”, basaltic andesite with olivine, and andesite with pyroxene and olivine associated with the La Poruña scoria cone.(Mud flow deposits.(Rubble breccia at the base of dome talus.Volcanoes and Lavas from the Middle PleistoceneLava flows, lava domes and domes having moderately preserved surface features, which form the San Pablo, Azufre, Aguilucho, and Paniri volcanoes, and the early stage of the Chanca domes.Pyroxene andesite domes and dacitic domes with biotite and hornblende.(Avalanche of the Polapi Station.(Avalanche of the San Pedro volcano.(Avalanche of the Paniri volcano.(Avalanche of the Aguilucho volcano.Volcanoes of the Lower PleistoceneVolcanic system, predominantly andesitic to dacitic, formed by lavas, domes and lava domes, as well as interspersed tuff and breccia. This unit includes the La Cueva and Apacheta volcanoes, distal lava flows from the Cañapa hill, the domes called Ascotán, Pabelloncito Bajo and Cachimba, and early lavas from the Azufre and Aguilucho volcanoes.Reddish and black sequences of maphic composition, formed by lavas, agglomerates, and andesitic tuffs.Dacite domes with amphibole.(Volcanic avalanche of the Azufre volcano.Volcanoes of the Lower [sic] Pliocene Volcanic systems, predominantly andesitic to dacitic, formed by lavas, domes, and lava domes, as well as interspersed tuff and breccia. This unit includes the Palpana volcanoes and the alignment of N-NW Araral-Ascotán-Barrancane volcanoes that define the international boundary.Reddish and black sequences of maphic composition, formed by agglomerates, and andesitic tuffs.Dacite domes with amphibole.Volcanoes of the Lower PlioceneVolcanic edifices having NW-SE alignment over lavas and domes from the Upper Miocene. Predominantly composed of andesitic lava flows with ortho and clinopyroxene ± olivine or amphibole, and dacites with hornblende and biotite ± ortho-pyroxene. These include the Cebollar, Polapi, and Carasilla volcanoes.Reddish and black sequences of maphic composition, formed by lavas, agglomerates and andesitic tuffs. Volcanoes and Volcanic and Volcanoclastic Sequences from the Upper Miocene (ca. 11.3-5.4 Ma) Domes, lava domes, lavas and autoclastic breccia that represent the remnants of variably collapsed volcanic systems. The centres aligned in a NW-SE direction stand out in the hills of Colana, Lay Lay and Colorado, at the southern foot of the Inacaliri ridge. Includes the north side of the voluminous Carcanal hill. The lavas have an andesitic composition with phenocrysts of clinopyroxene, ortho-pyroxene and amphibole, and lesser (not always present) biotite and olivine.Dacitic and rhyolitic domes with amphibole and biotite, locally with flow banding.(Deposits of volcanic avalanches and block-and-ash that are found interspersed between the ignimbrites of San Pedro and Sifón, as well as over the Polapi Ignimbrite.Sub-volcanic body of the Diablo hill. White dacitic dike with biotite and amphibole, with quartz cat’s eye.Volcanic and Volcanoclastic Sequences from the Middle Miocene (ca. 15.6-12.1 Ma)Discontinuous sequence of andesitic lavas with amphiboles, dacitic domes with amphibole and biotite and very thick breccia, with blocks of up to 5-6 m, of diverse origin (collapse, mud flows, and block-and-ash). Discrete outcrops, small and having rounded surfaces in the ravines that drain the basin of the Polapi Station, in Bayos hillocks, east of Salar de Ascotán (salt flat), and southwest of the San Pedro volcano.Relicts of dacitic to rhyolitic domes with biotite and amphibole in an NW-SW alignment.(Deposits of polymictic volcanic breccia, of pyroclastic breccia of block-and-ash, subordinated small-volume unwelded tuffs, and intercalations of sands and epiclastic conglomerates.Strata of the (Diablo hill) (Lower Miocene)Continental, folded volcano-sedimentary sequence, constituted by fine epiclastic breccia, fine to medium conglomerates, and coarse to fine red sandstones, locally with calcareous cement. Exclusively volcanic clasts, having an intermediate to siliceous composition.PlsvPlmvPlivPivDEPOSITS AND STRATIFIED ROCKSAnthropogenic Deposits (Holocene)Deposits and accumulations of arid materials and industrial ores resulting from human activity.Ephemeral Lacustrine Deposits (Holocene)Deposits of terrigenous and evaporite silts deposited in basins that contain ephemeral lakes during wet periods.Active Fluvial Deposits (Holocene)Gravels, sandy gravels, and unconsolidated sands, well to moderately sorted, associated to permanent or intermittent river courses. Locally, these include marsh deposits and evaporites.Active Alluvial Deposits (Holocene) Poorly consolidated deposits of gravels, sands, and polymictic silts, with varying grades of sorting, which develop alluvial plains of variable extent and fans having low to moderate slopes.Rock Glaciers (Holocene)Gravels composed of rock fragments and unconsolidated fine material, with cement or perennial ice core, deposited at the base of talus, the majority having superficial striations that indicate movement.Fall Pyroclastic Deposits (Upper Pleistocene – Holocene)Deposits of salt crusts having varying degrees wind re-deposition that fill the basin of Ascotán Salt Flats, consisting mainly of gypsum, halite, and ulexite.Saline Deposits (Upper Pleistocene – Holocene)Deposits of salt crusts having varying degrees wind re-deposition that fill the basin of the Ascotán Salt Flats, consisting mainly of gypsum, halite, and ulexite.Deposits of Mass Wasting (Upper Pleistocene – Holocene)Deposits of detritus flows having limited extent, unconsolidated, chaotic, poorly sorted, consisting of gravels with angular to sub-angular fragments and a matrix of fine gravel, sand, and silt.Aeolian Deposits (Upper Pleistocene – Holocene)Deposits of terrigenous sands and evaporite that partially cover the lower topography to the east of the Ascotán Salt Flats.Alluvial and Colluvial Deposits (Upper Pleistocene – Holocene)Unconsolidated sedimentary sequences, moderately sorted, polymictic and matrix-supported, whose deposits generally constitute vast alluvial plains and alluvial cones with little transport on the upper slopes. They present recurrent intercalations of fall pyroclastic deposits.El Encanto Ignimbrite (Holocene, ca. 11 ky)Deposit of pumice pyroclastic flow, dacitic, small in volume, unwelded, associated with the most recent eruption of the San Pedro Volcano.Alluvial and Colluvial Deposits of the PleistoceneSequences of medium to poorly sorted gravels and sandy gravels that constitute fossil alluvial plains and old alluvial and colluvial fans, with abundant incisions from both inactive and present-day ravines.Avalanche of Cuero de Vaca (Medium Pleistocene, ca. 0.6 Ma)Chaotic clastic deposit, very coarse, with blocks of up to several metres of dark clinopyroxene lavas, some with “jigsaw” texture or prismatic jointing that also incorporates large blocks several metres in size, internally deformed and fractured, of the Polapi Ignimbrite. It crops out along the upper course of the Polapi brook, with a characteristic appearance with large spots or areas of light and dark colours.Avalanches of Debris from the PleistoceneMassif chaotic deposits with blocks up to metres in size in a sandy-silty matrix. These correspond to partial gravitational collapses of volcanic systems, that took place after the volcanic activity.Aguilucho Ignimbrite (Lower Pleistocene, ca. 1 Ma)Tuffs of ash and reddish lapilli and pyroclastic breccia that crops out in the Aguilucho ravine, to the north of the Inacaliri graben and in the high parts of the Apacheta volcano, with a maximum thickness of 20 m. They are locally rich in juvenile fragments with an incomplete mixture of magmas (banded scoria). They include dense andesitic lithics, scoriaceous bombs, amphibole pumice and biotite and banded scoria.Fall Pyroclastic Deposits (Upper Pliocene – Pleistocene) Fall pyroclastic sequences that alternate dark layers of scoria with light pumice layers, found on the hillsides in the northwest sector of the map. Probably the product of the activity of the Palpana and/or Chela volcanoes.Polapi Ignimbrite (Upper Miocene, ca. 8.2 Ma)Rhyolitic tuff with biotite, unwelded, very rich in pumice and lithic fragments. Made up of more than one flow unit, with local vertical variations, but generally having a massif appearance, up to 45 m thick. It crops out in ravines and rolling hills of the Polapi Station basin. It fills gorges carved in the Sifón Ignimbrite.Mass Wasting and Alluvial Deposits from the Upper MioceneClastic sequence dominated by a deposit of detritus avalanche comprised exclusively of fragments from the clastic sequence of El Diablo hill, probably caused by a non volcanic gravitational landslide. Towards the top there are immature alluvial sequences that resulted from re-deposition of the avalanche deposit.Sifón Ignimbrite Upper Miocene, ca. 8.6-8.2 Ma)Dacitic tuff, moderately welded, rich in crystals of biotite, plagioclase and quartz, with a few localised fragments of pumice and lithics. Massif, homogenous appearance, with incipient columnar jointing. It crops out along the San Pedro River and in the basin of the Polapi Station, being up to 20 m thick and filling a very irregular topography..San Pedro Ignimbrite Upper Miocene, ca. 10-12 Ma)Unwelded moderately consolidated dacitic tuff, very rich in lithics and pumice fragments. Crystals of plagioclase and less biotite and amphibole. Massif, without internal structures, characterised by erosion tafone. It crops out exclusively along the San Pedro River, having a maximum width of 40 m, no exposed base. With alluvial coverage. HanHlHfHaPlHpcPlHsPlHlPlHrmPlHePlacPlHacMsipMrmMsispPlgHgrHvUPPERPLEISTOCENE MIDDLEPlacvLOWERPladPsvMsvMmvMidPliiaPPl(pc)MsisSalar PuntaNegra Región de AtacamaRegión de Tarapacá70º00'68º00'22º00'24º00'26º00'RíoLoaTocopillaNCalamaOCÉANOPACÍFICOBOLIVIAREGIÓN DE ANTOFAGASTASalar de AtacamaSheet on Ollagüe, scale 1:250.000 Map of Ascotán – Inacaliri hill, scale 1:100.000ARGENTINAANTOFAGASTA0100 km5TaltalPaposo56º43º32º19ºARGENTINA*"1998 AGREEMENT"OCÉANOPACÍFICO72º68ºkm090º53ºSantiagoDIAGRAM FOR LOCATION PURPOSES ON SHEET ON OLLAGÜE SHEET ON ASCOTÁN-INACALIRI HILL, scale 1:100.000 Maps from the IGM Catalogue, scale 1:50.000QUEHUITACHELACHITIGUAVOLCANOMIÑOHILLLLOCASHILLPALPANASAN MARTÍNOR CARCOTE SALT FLATSASCOTANHILLJASPESTATIONSAN PEDROSAN PEDRO ANDSAN PABLOVOLCANOES30'45'15'69º00'68º00'45'21º00'30'22º00'15'OLLAGÜEOLLAGÜEVOLCANOHILLARARALHILL ORINACALIRIDEL CAJÓNLEGENDRADIOMETRIC DATINGS (Ma ± 2 σ)K-Ar total rockK-Ar biotiteK-Ar amphiboleK-Arno information40Ar/39Ar biotite40Ar/39Argroundmass 40Ar/39Ar amphiboleU-Pbzircon14Cpeat (years ky) 14Cwood (years ky) 14Cwater(years ky)* Recalculated age, approximate locationclt Age obtained in clast included in depositSOURCE OF RADIOMETRIC DATING1 This paper2 3Baker report)Renzulli et & written communication; 2017)HYDROTHERMAL ALTERATIONIntermediate argillic (clay)SolfataricSRQ#ÉÉÉÉÉÉÉ3672""3672RiverDry ravine or intermittent watercourseLagoonHot spring WetlandPaved roadUnpaved roadTrack or trailSettlementIndex contour line Secondary contour lineSnow index contour lineSecondary snow index contour lineLevel m.a.s.l.)Snow level ma.s.l.)International boundary"Observed / inferred geological contact Intra-formational contactIndeterminate observed faultInferred / covered faultNormal fault observed / inferred Reverse fault observed / inferredCovered reverse faultPhoto-lineamentObserved / inferred topographic caldera margin Observed / inferred volcanic crater Craterless volcanic eruption centreLava flow and direction of flowLevee Glacial cirque Morainic crestAlignment of volcanic eruption centresMass wasting escarpmentTopographic escarpmentPaleo-coastline Direction and dip angle of strataMetallic deposits and prospectsRock and industrial ore depositsCorrelative deposit number in Tables 9 and 10, annexedto text Geological profile lineAA'!!!EE¡¹11í®®®(a)(d)(((((((((((((((((((((d)(a)(av1)(e)(b)((((((((((((((((d)(a)(av)!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!(d)(fp)(1)(a)(d)(av)(sv)-!5/Hie(ab)(l)(av4)6.5005.5004.5003.5002.500A’A5.5004.5003.5002.5001.5006.5001.500B5.5004.5003.5002.5001.5006.5006.5005.5004.5003.5002.500B’’ 1.500Plmv MORPHOSTRUCTURAL SCHEMATIC MAPSPREVIOUS WORKSRamírez& Huete (1980)D. Sellés, M. Gardeweg(1:100.000)SOURCE OF INFORMATION68º30'15'68º21º30'22º45'68º30'15'68º21º30'22º45'Main faultsInternational boundaryAlignments (ridges)¡Eruptive centresO’Callaghan& Francis (1986)Herrera et al. (2007)Ahumada& Mercado (2010)Bertín& Amigo (2015)Bertín & Amigo (2015), modifiedMiddle SierraLoa River depressionInter-mountain depressions of the Main Mountain Range (salt flats) Polapi Perdíz ravine10.-San Pedro and Silala RiversMain Mountain Range, Neogene volcanic ridge1.- Palpana-Cebollar-Polapi-Carasilla volcanic ridge 2.- Araral-Ascotán-Barrancane volcanic ridge3.- Azufre and Aguilucho volcanoes ridge4.- San Pedro-San Pablo volcanoes5.- Colorado-Lay Lay-Colana hills ridge6.- Graben of Inacaliri ridgellllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll!.04020Km8REPÚBLICADEBOLIVIA68°68°30'22°21°30'4109571236EstaciónSanPedroRío LoaSalar de AscotánSierra del MedioRío San PedroRío SilalaEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEElllllllllllllllllllllllllll0105Km68°10’21°50’Complejo Volcánico Cordón Inacaliri10569PLIOCENE MIOCENE UPPERUPPERNEOGENE PivPivLacustrine Deposits (Upper Pleistocene – Holocene)Fine sedimentary sequences, laminated, deposited in depressions between alluvial fans or basins dammed by lavas and mass wasting.Lacustrine deposits with abundant diatomite deposits.Glacial Deposits (Upper Pleistocene)Poorly to weakly consolidated deposits, massif, poorly sorted, formed by gravels having abundant matrix of sand and clay, commonly matrix-supported, with blocks of up to 2 m in diameter, wormy texture, and with morainic arcs or crests that reveal different stages of glacial retreat. They form frontal and lateral moraines on the slopes of the volcanoes. Young moraines.PlHsPlHsPlHlPlHlPlHrmPlHrmPlHePlHePlacPlacPlHacPlHacPlgPlgPlacvPlacvPsvPsvLOWERMsvMsvMmvMmvMIDDLE(a)(d)(((((((((((((((((((((((((((((((d)(1)(a)(d)(av)(sv)(d)(b)(HieHieiv(scale 1:100.000Horizontal scale 1:100.000Vertical complexm.a.s.l.m.a.s.l.POLECHILEANANTARCTIC TERRITORY.bb"ACUERDO ENTRE LA REPÚBLICA DE CHILE Y LA REPÚBLICA ARGENTINA PARA PRECISAR EL RECORRIDO DEL LÍMITE DESDE EL MONTE FITZ ROY HASTA EL CERRO DAUDET". (Buenos Aires, 16 de diciembre de 1998).GEOLOGÍA DEL ÁREAASCOTÁN - CERRO INACALIRIREGIÓN DE ANTOFAGASTADaniel Sellés M.Moyra Gardeweg P.SERIE GEOLOGÍA BÁSICACARTA GEOLÓGICA DE CHILENo. xxxEscala 1:100.000201XSUBDIRECCIÓN NACIONAL DE GEOLOGÍAISSN xxxx-xxxx90°TERRITORIO CHILENO ANTÁRTICO53°POLO SUREquidistancia curvas de nivel: 50 mESCALA 1:100.0001000 m 0 1 2 3 4 5 km DECLINACIÓN MAGNÉTICAENERO-2015NORTE MAGNÉTICONORTE ASTRONÓMICOGEOLOGÍA MINERÍAESCALA 1:100.000MAPA DE UBICACIÓNReferencia bibliográficaSellés M., D.1;Gardeweg P., M.1, N.1, 2016. Geología delárea PampaLirima-Cancosa,Regiónde Tarapacá.Servicio Nacional de Geología y Minería, Carta Geológica de Chile, Serie Geología BásicaXXX:XX p.,1 mapaescala 1:100.000. Santiago.1 Aurum Consultores-Servicios Geológicos y Mineros Ltda. [email protected]; [email protected] 0717-7283Inscripción No.271.810© Servicio Nacional de Geología y Minería. Av. Santa María 0104 Casilla 10465, Santiago, Chile.Director Nacional (S): Mario Pereira A.Subdirector Nacional de Geología (S): Omar Cortés C.Derechos reservados, prohibida su reproducción.EdiciónJefa Comité Editor: Renate Wall Z.Comité Editor: Rodrigo Carrasco O., Aníbal Gajardo C., Jorge Muñoz B., Andrew Tomlinson.Editores:Constantino Mpodozis M., Jorge Muñoz B.Jefa de Publicaciones: Soraya Amar N.Jefa de Unidad de Sistemas de Información Geológica (USIG): Constanza Casanova de L.Normas utilizadasEscala Geológica del Tiempo: Gradstein, F.M.; Ogg, J.G.; Schmitz, M.D.; Ogg, G.M. (editores) 2012.Base topográficaCartas escala 1:50.000,Ascotan, Cerro Araral, Volcanes San Pedro y San Pabloy Cerro Inicaliri o del Cajóndel Instituto Geográfico Militar (Chile), modificadas.Referencia geodésicaProyección Universal Transversal de Mercator (UTM), Zona 19S, SIRGAS.Apoyo científico y técnicoEstudios petrográficos: Los autores.Factibilidades de dataciones radiométricas: Eugenia Fonseca P. y los autores.Determinaciones radiométricas K-Ar:Marcelo Yáñez B. y César Vásquez B. 40Ar/39Ar: Carlos Pérez de Arce R.,Laboratorio de Geocronología del Servicio Nacional de Geología y Minería.Determinaciones radiométricas U-Pb LA-ICP-MS: Dirk Frei, Stellenbosch University, Sudáfrica, y Víctor Valencia, Washington State University, Estados Unidos.Determinaciones radiométricas U-Pb SHIRMP: Mark Fanning, Australian National University, Australia.Producción digital: Luis Delcorto A., Aurum Consultores Ltda.Apoyo financieroFondos Sectoriales del Servicio Nacional de Geología y Minería, Plan Nacional de Geología."Autorizada su circulación, en cuanto a los mapas y citas que contiene esta obra, referenteso relacionadas con los límites internacionales y fronteras del territorio nacional, por Resolución No. 126del13 de octubre 2016de la Dirección Nacional de Fronteras y Límites del Estado. La edición y circulación de mapas, cartas geográficas u otros impresos y documentos que se refieran o relacionen con los límites y fronteras de Chile, no comprometen, en modo alguno, al Estado de Chile de acuerdo con el Art. No. 2, letra g) del DFL No. 83 de 1979 del Ministerio de Relaciones Exteriores".LEYENDACENOZOICOCUATERNARIOPLIOCENOMIOCENOHOLOCENOSUPERIORSUPERIORINFERIORNEÓGENOPLEISTOCENOESTRATOVOLCANES Y SECUENCIAS VOLCANICASProductos volcánicos del HolocenoProductos volcánicos holocenos del volcán San Pedro.Domos y coladas domo de dacitas de hornblenda y, biotita y piroxenos subordinados, de reducido volumen, con un alcance no superior a los 2,5 km desde la cumbre del edificio actual del volcán San Pedro.Depósitos de bloques y ceniza, macizos, mal seleccionados, no soldados, matriz soportados de color gris claro, conabundantesbloquescondiaclasamiento prismáticos originados por elcolapsodecoladas domos ydomosVolcanes y lavas del Pleistoceno Superior.Secuencias Cerro Pabellón. y dacíticos de hornblenda y piroxeno (Vn. San Pedro)olivino(ab) Lavas "aa" andesítico basálticas de olivino y andesíticas de piroxeno y olivino asociadas al cono de escorias La Poruña(l) Depósitos de lahares(b) (d) hornblenda(av1) Avalancha Polapi(av2) Avalancha Pedro(av3) Avalancha Paniri(av4) Avalancha Apacheta, coladas distales del cerro Cañapa, los domos Ascotán, Pabelloncito bajo y Cachimba y lavas tempranas de los volcanes Azufre y Aguilucho.Superior(ca. 11,3-5,4 Ma)Domos, son de composición andesítica con fenocristales de clinopiroxeno, ortopiroxeno y anfíbola y contenidos menores y no siempre presentes de biotita y olivino.(d) Domos dacíticos y riolíticos de anfíbola y biotita, localmente con bandeamiento de flujo(av) Depósitos de avalanchas volcánicas y de block and ash que se encuentran intercalados entre las ignimbritas San Pedro y Sifón así como sobre la Ignimbrita Polapi.(sv) Cuerpo subvolcánico del cerro del Diablo. Dique dacítico blanco de biotita y anfíbola, con ojos de cuarzo.Secuencias volcánicas y volcanoclásticas del Mioceno Medio (ca. 15,6-12,1 Ma)Secuencia discontínua de lavas andesíticas de anfíbola, domos dacíticos de anfíbola y biotita y brechas muy gruesas, con bloques de hasta 5-6 m, de orígen diverso (colapso, lahares y bloques y ceniza). Afloramientos discretos, pequeños y de superficies redondeadas en las quebradas que drenan la cuenca de la Estación Polapi, en Cerritos Bayos, al este del Salar de Ascotán y al suroeste del volcán San Pedro.(d) Relictos de domos dacíticos a riolíticos de biotita y anfíbola, alineados NO-SO(b) Depósitos de brechas volcánicas polimícticas, de brechas piroclásticas de bloques y cenizas, tobas no soldadas de pequeño volumensubordinadas e intercalaciones de arenas y conglomerados epiclásticos.Estratos del Cerro del Diablo (Mioceno Inferior)Secuencia volcano-sedimentaria continental plegada, compuesta por brechas epiclásticas finas,conglomerados finos a medios y areniscas rojas gruesas a finas, localmente con cemento calcáreo. Clastos exclusivamente volcánicos, de composición intermedia a silícea. PlsvPlmvPlivPivDEPÓSITOS Y ROCAS ESTRATIFICADASDepósitos antrópicos(Holoceno)Depósitos y acumulaciones de materiales áridos y minerales industriales producto de la actividad humanaDepósitos lacustre efímeros(Holoceno)Depósitos de limos terrígenos y evaporíticos depositados en cuencas que albergan efímeros lagos durante períodos húmedos.Depósitos fluviales activos(Holoceno)Gravas, gravas arenosas y arenas no consolidadas, bien a moderadamente seleccionadas, asociadas a cursosfluviales permanentes o intermitentes. Localmente incluyen depósitos palustres y evaporíticos.Depósitos aluviales activos(Holoceno)Depósitos pobremente consolidados de gravas, arenas y limos polimícticos, con variables grados de selección,que desarrollan planicies aluviales de diversa extensión y abanicos de baja a moderada pendiente.Glaciares caída (Pleistoceno Superior-Holoceno)Depósitos de costras salinas con variable retrabajo eólico que rellenan la cuenca del Salar de Ascotán. Consisten principalmente en yeso, halita y ulexita.Depósitos de costras salinas con variable retrabajo eólico que rellenan la cuenca del Salar de Ascotán. Consisten principalmente en yeso, halita y ulexita.Depósitos lacustres(Pleistoceno Superior-Holoceno)(d) Depósitos lacustres con abundantes depósitos de diatomeas.Depósitos de flujos de detritos de extensión reducida, no consolidados, caóticos, mal seleccionados, constituidospor gravas de fragmentos angulosos a subangulosos y matriz de grava fina, arena y limo.Depósitos eólicos(Pleistoceno Superior-Holoceno)Depósitos de arenas terrígenas y evaporíticas que cubren parcialmente la topografía baja al oriente del Salar cuyos depósitos constituyen generalmente vastas planicies aluviales y conos de deyección con escaso transporte en la laderas altas. Presentan recurrentes intercalaciones de depósitos piroclásticos de caída.Depósitos glaciales(Pleistoceno Superior)Depósitos pobre a débilmente consolidados, macizos, mal seleccionados, formados por gravas con abundantematriz de arena y arcilla, comúnmente matriz soportados, con bloques de hasta 2 m de diámetro, de superficie (Pleistoceno Medio;ca. 0,6 Ma)Depósito clástico caótico, muy grueso, con bloques de hasta varios de metros de lavas oscuras de clinopiroxeno, algunos con textura “jigsaw” o diaclasamiento prismático que incorpora además grandes bloques de varias decenas de metros, internamente deformados y fracturados, de la ignimbrita Polapi. Aflora a lo largo del curso superior del estero Polapi, con un característico aspecto de grandes manchas o dominios de colores claros y oscuros.Avalanchas de detritos del PleistocenoDepósitos caóticos macizos con bloques de tamaños hasta métricos en matriz areno-limosa. Corresponden acolapsos gravitacionales parciales de aparatos volcánicos aunque ocurridos con posterioridad a la actividadvolcánica misma.Ignimbrita Aguilucho (Pleistoceno Inferior;ca.1 Ma)Tobas de ceniza y de lapilli rojizas y brechas piroclásticas que aflora en la quebrada Aguilucho, al norte del graben de Inacaliri y en lo alto del volcán Apacheta, con 20 m de espesor máximo. Localmente ricas en fragmentos juveniles con mezcla incompleta de magmas (escorias bandeadas). Incluye líticos andesíticos densos, bombas escoriáceas, pómez de anfíbola y biotita y escorias bandeadasDepósitos piroclásticos de caída(Plioceno Superior-Pleistoceno)Secuencias piroclásticas de caída que alternan niveles oscuros de escorias con niveles claros pumíceos, adosados a laderas de cerros en el sector noroccidental del mapa. Probablemente producto de la actividad de los volcanes Palpana y/o Chela.Ignimbrita Polapi(Mioceno Superior;ca. 8,2 Ma)Toba riolítica de biotita no soldada, muy rica en fragmentos de pómez y líticos. Compuesta por más de una unidad de flujo, con variaciones verticales locales, pero de aspecto en general macizo, con hasta 45 m de espesor. Aflora en quebradas y lomajes de la cuenca de la Estación Polapi. Rellena gargantas labradas en la Ignimbrita Sifón.Depósitos de remoción en masa y aluviales del Mioceno SuperiorSecuencia clástica dominada por un depósito de avalancha de detritos compuesta exclusivamente por fragmentosde lasecuencia clástica de Cerro El Diablo, probablmente originada por undeslizamiento gravitacional, novolcánico. Hacia el techo, secuencias aluviales inmaduras producto del retrabajo del depósito de avalancha.Ignimbrita Sifón (Mioceno Superior;ca. 8,6-8,2 Ma)Toba dacítica de ceniza, moderadamente soldada, rica en cristales de biotita, plagioclasa y cuarzo, con escasos y localizados fragmentosdepómez y líticos. Aspecto macizoy homogéneo, condiaclasamientocolumnarincipiente. Aflora a lo largo del río San Pedro y en la cuenca de la Estación Polapi con hasta 20 m de espesor, rellenando una topografía muy irregular.Ignimbrita San Pedro (Mioceno Superior;ca.10-12 Ma)Toba dacítica no soldada pero medianamente consolidada, muy rica en fragmentos de pómez y líticos. Cristales de lagioclasa y menos biotita y anfíbola. Maciza, sin estructuras internas, caracterizada por desgaste en tafonis. Aflora exclusivamente a lo largo del río San Pedro, con 40 m de espesor máximo, sin base expuesta.(c) Con cobertura aluvialHanHlHfHaPlHpcPlHsPlHlPlHrmPlHePlacPlHacMsipMrmMsispPlgHgrHvSUPERIORMEDIOPlacvINFERIORPladPsvINFERIORMsvMmvMidMEDIOPliiaPPl(RíoLoaTocopillaNCalamaOCÉANOPACÍFICOBOLIVIAREGIÓNDE AtacamaHoja Ollagüe, escala 1:250.000Carta Ascotan - Cerro Inacaliri, escala 1:ACUERDO DE 1998"BOLIVIACHILEPERÚ500 POLO SURTERRITORIOCHILENO ANTÁRTICOSantiagoCUADRO DE SITUACIÓNEN HOJAOLLAGÜEHOJA ASCOTAN-CERRO INACALIRI, escala 100.000Cartas Catálogo IGM, escala 1:50.000QUEHUITACHELACHITIGUAVOLCÁNMIÑOCERROLLOCASCERROPALPANASALAR DE SAN MARTÍNO CARCOTEASCOTANCERROJASPEESTACIÓNSAN PEDROVOLCANESSAN PEDRO Y SAN PABLO30'OLLAGÜEVOLCÁNOLLAGÜECERROARARALCERRO INACALIRI O DEL CAJÓNSIMBOLOGÍADATACIONES RADIOMÉTRICAS (Ma ± 2σ) roca total K-biotitaK-anfíbolaK-Ar sin información40Ar/biotita 40Ar/39Ar masa fundamental40Ar/39Ar anfíbolaU-Pb circón14C turba (años ka)14C madera (años ka)14C agua (años ka) Edad recalculada, ubicación aproximadaclt Edad obtenida en clasto incluido en depósitoFUENTE DE LAS DATACIONES RADIOMÉTRICAS1Este trabajo2 Roobol et al. (1976), Francis y Rundle (1976) 3 Baker (1977a) (=memoria) 4 Payne (1998) 5 Wörner et al. (2000)6 Tomlinson et al. (2001)7 Urzúa et al. (2002)8 Renzulliet al. (2006)9Tierney, 201110 Bertín y Amigo (2015)11 Rivera et al. (2015)12 Enel (com. escrita; 2017)ALTERACIÓN HIDROTERMALArgílica intermediaSolfatáricaSRQ#3672RíoQuebrada seca o curso intermitenteLagunaFuente termalBofedalCamino pavimentadoCamino sin pavimentarHuella o senderoPobladoCurva de nivel índiceCurva de nivel secundariaCurva de nivel nieve índiceCurva de nivel nieve secundariaCota (msnm)Cota nieve (msnm)Límite internacional"Contacto geológico observado/inferidoContacto intraformacionalFalla observada indeterminadaFalla inferida/cubiertaFalla normal observada/inferidaFalla inversa observada/inferidaFalla inversa cubiertaFotolineamientoMargen topográfico de caldera observado/inferidaCráter volcánico observado/inferidoCentro de emisión volcánico sin cráterColada de lava y dirección de flujoLevéeCirco glacialCresta de morrenaAlineamiento de centros de emisión volcánicosEscarpe remoción en masaEscarpe topográficoPaleolínea de costaRumbo y manteo de estratosYacimientos y prospectos metálicosYacimientos de rocas y minerales industrialesNúmero correlativo de yacimiento en tablas 9 y 10 anexas al textoTraza de perfil geológicoAA'((®(av2)(av3)(d)(fd)(d)(c)(d)(b)IncaCERROSDELINCAOBARRANCANEQuebradaLaPerdízHITO S/N LXXIICo. 7570km22°55'758050'759045'760040'761035'7620km21°30'580590600km5's.n.m.6.5005.5004.5003.5002.500A’Am s.n.m.6.5005.5004.5003.5002.500B’’ vertical 100.000Escala horizontal 1:100.000B’ ESQUEMA MORFOESTRUCTURALTRABAJOSANTERIORESRamírez y FUENTE DE LA INFORMACIÓN68Fallas principalesLímite internacionalAlineamientos¡Centros de emisiónO’Callaghan y Francis (1986)Herreraet Ahumada y Bertín y y Amigo (2015), modificadoSierra del MedioDepresión del río LoaDepresiones intermontanas de la Cordillera Principal 7.- Salar de Ascotán 8.- Quebrada Polapi 9.- Quebrada Perdíz10.- Ríos San Pedro y SilalaCordillera Principal, cordón volcánico neógeno 1.- Cordón volcánico Palpana-Cebollar-Polapi-Carasilla 2.- Cordón volcánico Araral-Ascotán-Barrancane 3.- Cordón volcanes Azufre y Aguilucho 4.- Volcanes San Pedro-San Pablo 5.- Cordón cerros Colorado-Lay Lay-Colana 6.- Graben Cordón de Inacalirillllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll!.Inacaliri10569EN EDICIÓNEN EDICIÓNPLIOCENOEN EDICIÓNPLIOCENOMIOCENOEN EDICIÓNMIOCENOEN EDICIÓNSUPERIOREN EDICIÓNSUPERIORSUPERIOREN EDICIÓNNEÓGENOEN EDICIÓNSecuencias volcánicas de aspecto reciente, formadas por lavas y domos andesíticos a dacíticos, con brechas EN y tobas intercaladas. Corresponden principalmente a lavas dacíticas de hornblenda y piroxeno del volcán San EDICIÓNy Pedro, el cono de escoria y colada de lava La Poruña y, los domos dacíticos Pabellón, Chac Inca, Chanka y EDICIÓNPedro, Domos y coladas domo dacíticos de biotita y hornblenda (Pabellón, Chac Inca, Chanca, Azufrera del Bajo) EDICIÓNDomos Secuencias de flujos de escoria aglutinados, tefra y coladas de lava delgadas, rojizas a negras, andesíticas. Depósitos de flujos de detritos asociados a los domos Chanka.EDICIÓNDepósitos Cono de escorias La Poruña, andesítico basáltico de olivino y andesítico de piroxeno y olivinoEN EDICIÓNCono olivinoLavas EDICIÓNLavas Brechas de escombro en la base de talud de domosEN EDICIÓNBrechas domosVolcanes y lavas del Pleistoceno MedioEN EDICIÓNVolcanes MedioColadas de lava, lavas domo y domos con rasgos superficiales medianamente conservados, que forman losEN EDICIÓNColadas losvolcanes San Pablo, Azufre, Aguilucho y Paniri y, la etapa temprana de los domos Chanca.EDICIÓNvolcanes Domos andesítico de piroxeno y dacíticos de biotita y hornblendaEN hornblendaAvalancha de la estación PolapiEN EDICIÓNAvalancha PolapiAvalancha del volcán San PedroEN PedroAvalancha del volcán PaniriEN PaniriAvalancha del volcán AguiluchoEN AguiluchoVolcanes del Pleistoceno InferiorEN InferiorAparatos volcánicos predominantemente andesíticos a dacíticos, formadas por lavas, domos y lavas domo, EDICIÓNAparatos además de brechas y tobas intercaladas. Dentro de esta unidad se encuentran los volcanes Las Cueva y EDICIÓNademás EDICIÓNApacheta, EDICIÓNde (a) EDICIÓN(Secuencias rojizas y negras de composición máfica, formada por lavas, aglomerados y tobas andesíticas.(d) Domos dacíticos de anfíbolaEN anfíbola(av) Avalancha volcánica del volcán AzufreEN AzufreVolcanes del Plioceno InferiorAparatos volcánicos predominantemente andesíticos a dacíticos, formadas por lavas, domos y lavas domo, además de brechas y tobas intercaladas. Dentro de esta unidad se encuentran los volcanes Palpana y el alineamiento de volcanes N-NO Araral-Ascotán-Barrancane que definen el límite internacional.EDICIÓNalineamiento (a) Secuencias rojizas y negras de composición máfica, formada por lavas, aglomerados y tobas andesíticas.(d) Domos dacíticos de anfíbolaVolcanes del Plioceno InferiorEN InferiorEdificios volcánicos alineados NO-SE sobre lavas y domos del Mioceno Superior. Compuestos EDICIÓNEdificios predominantemente por coladas de lava andesíticas de orto y clinopiroxeno ± olivino o anfíbola y, dacíticas de EDICIÓNpredominantemente hornblenda y biotita y ±ortopiroxeno. Incluye los volcanes Cebollar, Polapi y Carasilla.EDICIÓNhornblenda (a) Secuencias rojizas y negras de composición máfica, formada por lavas, aglomerados y tobas andesíticas.Volcanes y secuencias volcánicas y volcanoclásticas del Mioceno SuperiorEN SuperiorDomos, lavas domo, lavas y brechas autoclásticas que representan los remanentes de aparatos volcánicos EDICIÓNDomos, variablemente desmantelados. Destacan los centros alineados en dirección NO-SE de los cerros Colana, Lay EDICIÓNvariablemente Lay y Colorado, al pie sur del cordón Inacaliri. Incluye el flanco norte del voluminoso cerro Carcanal. Las lava EDICIÓNLay EDICIÓNson flujoEN EDICIÓNSan EDICIÓNCuerpo PivEN EDICIÓNPivGlaciares de roca EDICIÓNGlaciares (Holoceno)Gravas compuestas de fragmentos de roca y material fino no consolidado, con cemento o núcleo de hielo EDICIÓNGravas perenne, depositadas en la base de taludes, la mayor parte con estrías superficiales que señalan movimiento.EDICIÓNperenne, Depósitos piroclásticos de caídaEN caídaDepósitos EDICIÓNprincipalmente Depósitos salinosEN salinos (Pleistoceno Superior-Holoceno)EDICIÓN lacustresEN Secuencias sedimentarias finas,laminadas, depositadas en depresiones entre abanicos aluviales o cuencasEN cuencasrepresadas por lavas y remociones en masa.EDICIÓNrepresadas Depósitos de remoción en masaEN masa(Pleistoceno Superior-Holoceno)constituidosEN EDICIÓNpor eólicosEN de Ascotán.Depósitos aluviales y coluvialesEN coluviales(Pleistoceno Superior-Holoceno)Secuencias sedimentarias no consolidadas, moderadamente seleccionadas, polimícticas y matriz-soportadas,EDICIÓNcuyos EDICIÓNen glacialesEN abundanteEN EDICIÓNmatriz amorrillada y con crestas o arcos morrénicos que evidencian distintos estados de retroceso glacial. Conforman EDICIÓNamorrillada morrenas frontales y laterales en los flancos de los volcanes.EDICIÓNmorrenas (1) Morrenas jóvenesEN jóvenesIgnimbrita El Encanto EDICIÓNIgnimbrita (Holoceno;ca. EDICIÓNca. 11 ka)Depósito de flujo piroclástico pumíceo, dacítico, de pequeño volumen, no soldado, asociado a la etapa más EDICIÓNDepósito reciente del volcán San Pedro. EDICIÓNreciente Depósitos aluviales y coluviales del PleistocenoEN PleistocenoSecuencias de gravas y gravas arenosas mediana a pobremente seleccionadas que constituyen planicies aluviales fósiles y abanicos aluviales y coluviales antiguos, con abundantes incisiones tanto de quebradasEN EDICIÓNaluviales quebradasinactivas como actuales.EDICIÓNinactivas Avalancha Cuero de Vaca EDICIÓNalgunos PlHsEN EDICIÓNPlHsPlHlEN EDICIÓNPlHlPlHrmEN EDICIÓNPlHrmPlHeEN EDICIÓNPlHePlacEN EDICIÓNPlacPlHacEN EDICIÓNPlHacPlgEN EDICIÓNPlgPlacvEN EDICIÓNPlacvEN EDICIÓNPsvEN EDICIÓNPsvINFERIOREN EDICIÓNINFERIORMsvEN EDICIÓNMsvMmvEN EDICIÓNMmvMEDIOEN EDICIÓNMEDIOEN EDICIÓN((((((((((EDICIÓNHieEN EDICIÓN"EDICIÓN""EDICIÓN"EDICIÓNEEN EDICIÓNEEEN EDICIÓN)))EDICIÓN))))))EDICIÓN))))EDICIÓN)EDICIÓN¡EDICIÓN¡¡EDICIÓN¡"¡"EDICIÓN"¡"¡EDICIÓN"¡""¡"EDICIÓN¡)))¡)))EDICIÓN)))¡))))¡)EDICIÓN)¡)EDICIÓN"¡"EEN EDICIÓNE¡EDICIÓN¡¡¡¡EDICIÓN¡¡¡¡¡¡EDICIÓN¡¡¡"¡"¡"¡"EDICIÓN"¡"¡"¡"EDICIÓNMsv(EDICIÓNPliv(PlacEN EDICIÓNPlacMsvEN EDICIÓNMsvMsvEN EDICIÓNMsvPlacEN EDICIÓNPlacPlgEN EDICIÓNPlgPlgEN EDICIÓNPlgPlHacEN EDICIÓNPlgMsv(HaEN EDICIÓNHaHaEN EDICIÓNHaEN EDICIÓNPlg1EN EDICIÓNPlg1PlgEN EDICIÓNPlg"Plg"PlHacEN EDICIÓNPlHacPlHacEN EDICIÓNPlHacHaEN EDICIÓNHaHgrEN EDICIÓNHgrHgrEN EDICIÓNHgrPlg1EN EDICIÓNPlg1PlHl(EDICIÓNPlHl(HgrEN EDICIÓNPlg1Plg1EN EDICIÓNPlg1PlHacEN EDICIÓNPlHacEN EDICIÓN1,612±EDICIÓN1,094±1EN 1HEN EDICIÓNHHEN EDICIÓNHEN EDICIÓN16EN EDICIÓN1616EN EDICIÓN16CEN EDICIÓNCEEN EDICIÓNEREN EDICIÓNRREN EDICIÓNROEN EDICIÓNOSEN EDICIÓNSDEN EDICIÓNDEEN EDICIÓNECEN EDICIÓNCOEN EDICIÓNOLEN EDICIÓNL¡AEN EDICIÓNA¡NEN EDICIÓNNNEN EDICIÓNN¡EDICIÓNAAEN EDICIÓNACEN EDICIÓNOOEN EDICIÓNOREN EDICIÓNRDEN EDICIÓNDDEN EDICIÓNDÓEN EDICIÓNÓÓEN EDICIÓNÓNEN EDICIÓNNDEN EDICIÓNEIEN EDICIÓNINEN EDICIÓNNAEN EDICIÓNCCEN EDICIÓNCAEN EDICIÓNALEN EDICIÓNLIEN EDICIÓNIREN EDICIÓNRIEN EDICIÓNICEN EDICIÓNIIEN EDICIÓNLLEN EDICIÓNIB''EDICIÓNB''4539EN EDICIÓN45394539EN EDICIÓN45394770EN EDICIÓN47704770EN EDICIÓN4770E4770EEN EDICIÓNE4770EE4770EEN EDICIÓNE4770E4777EN EDICIÓN47774756EN EDICIÓN47564756EN EDICIÓN4756E4756EEN EDICIÓNE4756EE4756EEN EDICIÓNE4756E5373EN EDICIÓN53735373EN EDICIÓN5373¡4808EN EDICIÓN48084808EN EDICIÓN48085265EN EDICIÓN52655265EN EDICIÓN5265E5265EEN EDICIÓNE5265EE5265EEN EDICIÓNE5265E¡5212EN EDICIÓN52125212EN EDICIÓN52125373EN EDICIÓN53735330EN EDICIÓN53305361EN EDICIÓN53615381EN EDICIÓN53815381EN EDICIÓN53815163EN EDICIÓN51635476EN EDICIÓN54765476EN EDICIÓN5476¡EDICIÓN)))¡)))5426EN EDICIÓN54265426EN EDICIÓN54265397EN EDICIÓN53975397EN EDICIÓN53975264EN EDICIÓN52645264EN EDICIÓN5264¡5446EN EDICIÓN54465117EN EDICIÓN51175117EN EDICIÓN51174996EN EDICIÓN49965373EN EDICIÓN53735408EN EDICIÓN54085408EN EDICIÓN54085342EN EDICIÓN53425342EN EDICIÓN53425147EN EDICIÓN51475147EN EDICIÓN5147¡EDICIÓN¡¡¡5385EN EDICIÓN53855385EN EDICIÓN53855422EN EDICIÓN54225422EN EDICIÓN54224861EN EDICIÓN48615360EN EDICIÓN53605330EN EDICIÓN5330¡5584EN EDICIÓN5584"EDICIÓN55845584EN EDICIÓN55845628EN EDICIÓN56285628EN EDICIÓN5628"5628EN EDICIÓN56284182EN EDICIÓN41824182EN EDICIÓN41823861EN EDICIÓN38614672EN EDICIÓN46724662EN EDICIÓN46623876EN EDICIÓN3876EN EDICIÓNVn. CerroEN CerroInacaliriEN EDICIÓNInacaliriInacaliriEN EDICIÓNInacaliri5330Inacaliri5330EN EDICIÓN5330Inacaliri5330InacaliriEN EDICIÓNInacaliriEN EDICIÓN55'7580EN EDICIÓN758050'EDICIÓN50'EDICIÓN10'590EN EDICIÓN5905'EDICIÓN5'ESQUEMÁTICOSEN EDICIÓNPERFILES EDICIÓNPivEN EDICIÓNPlHsEN EDICIÓNSalar AscotánEN CarasillaEN EDICIÓNVn. EDICIÓNDomo EDICIÓNPliv EDICIÓNMmv EDICIÓNPliv(EDICIÓNPliiaEN EDICIÓNPlacEN EDICIÓNMmvEN EDICIÓNMsisEN EDICIÓNMsipEN EDICIÓNNational ServiceScale 1:100.000GEOLOGY Annex VIII Appendix G
433
434
Annex IX
Suárez, F., Muñoz, J.F., Maass, T., Mendoza, M., 2017. Evapotranspiration Estimation in the Silala River Basin - Methods Review and Estimation of Wetland Evaporation
435
436
Annex IX
EVAPOTRANSPIRATION ESTIMATION IN THE SILALA RIVER BASIN
METHODS REVIEW AND ESTIMATION OF WETLAND EVAPORATION
Francisco Suárez (PhD)
Associate Professor, Pontificia Universidad Católica de Chile
José Francisco Muñoz (PhD)
Professor, Pontificia Universidad Católica de Chile
Tamara Maass, Civil Engineer
Magdalena Mendoza, Civil Engineer
May, 2017
Annex IX
437
GLOSSARY This glossary of hydrological terms is based on the following: http://www.wmo.int/pages/prog/hwrp/publications/international_glossary/…http://www.nws.noaa.gov/om/hod/SHManual/SHMan014_glossary.htmhttp://www.geo.utexas.edu/faculty/jmsharp/sharp-glossary.pdfEvapotransp… Combination of evaporation from free water surfaces and transpiration of water from plant surfaces to the atmosphere. Landsat: Group of satellites built and placed in orbit by the USA for high-resolution observation of the Earth’s surface. Mean Saturation Vapor Pressure: Mean maximum possible partial pressure of water vapour in the air at a given temperature. Net Radiation: Difference between incident and reflected radiation. Normalized Difference Vegetation Index: Indicator that can be used to analyze remote sensing measurements and assess whether the target being observed contains live green vegetation or not. Penman-Monteith Approach: Method for estimating evapotranspiration. Psychometric Constant: Constant that relates the partial pressure of water in air to the air temperature. Slope Of The Saturation Vapor Pressure Curve: Rate of change of saturation specific humidity with air temperature. Soil Heat Flux Density: Heat flux entering the ground per unit area. 438
Annex IX
TABLE OF CONTENTS
1 INTRODUCTION .................................................................................................... 1
1.1 Study zone .......................................................................................................... 1
1.2 Objectives ........................................................................................................... 1
1.3 Summary of the methodology ............................................................................ 5
1.4 Structure of the report......................................................................................... 6
2 SUMMARY AND CONCLUSIONS ....................................................................... 6
3 EVAPOTRANSPIRATION METHODS ................................................................. 8
3.1 Penman-Monteith (ETo) ................................................................................... 10
3.2 Turc (ETo)......................................................................................................... 10
3.3 Priestley-Taylor (ETo) ...................................................................................... 10
3.4 Taylor-de Bruin (ETo) ...................................................................................... 11
3.5 Jensen-Haise (ETo) ........................................................................................... 11
3.6 Actual evapotranspiration (ETr) ....................................................................... 11
4 AVAILABILITY OF METEOROLOGICAL DATA ............................................ 12
4.1 Meteorological data collected in this study ...................................................... 12
4.2 Meteorological data collected by other institutions or studies ......................... 14
4.2.1 Temperature and relative humidity ........................................................... 14
4.2.2 Wind speed data ........................................................................................ 15
4.2.3 Monthly Precipitation ............................................................................... 16
4.2.4 Monthly Streamflow ................................................................................. 17
5 NDVI DATA........................................................................................................... 17
6 RESULTS ............................................................................................................... 19
6.1 Potential ET using the Penman-Monteith formula and the meteorological data
collected in this study .................................................................................................. 19
6.2 Potential ET using the meteorological data collected by the DGA .................. 21
6.3 Actual ET using the meteorological data collected by the DGA and the NDVI..
.......................................................................................................................... 23
6.4 ETo and ETr, expressed as equivalent flow in the wetlands and percentage of
the monthly flow measured in DGA fluviometic station. ........................................... 25
Annex IX
439
6.5 Potential impacts of channelization on wetland ET ......................................... 27 7 CONCLUSIONS ..................................................................................................... 28 8 REFERENCES........................................................................................................ 30 APPENDIX A – CALCULATION OF ATMOSPHERIC PARAMETERS ................. 30 440
Annex IX
1
1 INTRODUCTION
The National Director of the Dirección Nacional de Fronteras y Límites del Estado
(DIFROL) of the Ministry of Foreign Affairs, Mrs. Ximena Fuentes, requested that
Professors Francisco Suárez and José Muñoz of the Pontificia Universidad Católica de
Chile undertake an interdisciplinary study aimed at deepening the hydrological
knowledge of the transboundary basin of the Silala River, located in the north of Chile.
This report presents the methodology and results related to the estimation of
evapotranspiration (ET) losses over the Orientales and Cajones wetlands, which are the
main wetlands in the basin that are located in Bolivia. Quantification of ET losses from
wetlands is important to understand their relative importance compared to the river flow
of the basin. In addition, knowing the ET losses will help to determine the impact of
channelization on the wetlands that are located within the Silala River basin. Using data
obtained from a meteorological station located in the basin and regional information, ET
was calculated using different methods, which were selected according to the data
available. This study was led by Drs. Francisco Suárez and José Muñoz and the report
was written under the supervision and instruction of Drs. Howard Wheater and Denis
Peach.
1.1 Study zone
The area of the study is the Silala River basin, a transboundary watershed shared by
Bolivia (upstream) and Chile (downstream). The Silala River basin is located in the
Andean Plateau of the Atacama Desert, approximately 300 km northeast of Antofagasta
(Figure 1-1). Figure 1-2 presents the meteorological and fluviometric stations of the
General Directorate of Water (Dirección General de Aguas, DGA) with more than 15
years of data that were used in this study. Figure 1-3 shows the meteorological and
fluviometric stations that are located in the Silala River basin.
1.2 Objectives
The general objective of this study is to estimate evapotranspiration rates, i.e., potential
evapotranspiration (ETo) and actual evapotranspiration (ETr), in the wetlands of the
Silala River basin. These ET estimates can be used to understand the magnitude of the
water lost to the atmosphere from the wetlands.
The specific objectives are to:
1. Evaluate ETo using recent data from a meteorological station installed in the Silala
River basin, using the Penman-Monteith method.
Annex IX
441
2 2.Combine estimates of the spatial extent of the wetlands with ETo and ETr estimates,using the Penman-Monteith method and other formulations available in theliterature, to estimate the historical ET losses in the wetlands of the Silala Riverbasin.3.Compare wetland evaporation losses in the Silala River basin with river flows at theDGA Fluviometric Station, located at the Chile-Bolivia international boundary.It is important to note that data are limited for remote regions such as the Silala River basin. Consequently, results from ET estimation methods are uncertain. Therefore, this report does not aim to obtain accurate ET estimates but rather to determine the relative importance of wetland ET for the basin hydrology. 442
Annex IX
3
Figure 1-1. Location of the Silala River at the Chile-Bolivia international boundary.
Annex IX
443
4 Figure 1-2. Location of the DGA’s stations that were used in this study. 444
Annex IX
5
Figure 1-3. The Silala River basin with the location of different hydrometeorological stations
near the study site.
1.3 Summary of the methodology
In this study, the following well-established ET estimation methods were used to
determine ETo: Penman-Monteith, Turc, Priestley-Taylor, Taylor-de Bruin, and Jensen-
Haise. In addition, ETr was evaluated using the method described by Groeneveld et al.
(2007), which combines estimates of ETo with the normalized difference vegetation
index (NDVI) obtained from Landsat images. Because meteorological stations that are
Annex IX
445
6 near or in the Silala River basin do not have a long record of measurements, and do not monitor all the variables required to estimate ETo (e.g., they do not record net radiation), a new meteorological station was installed at the Silala River basin (the “UC meteorological station”). This station has been fully operational since November 2016 and allows us to obtain ETo values from that date to the present. Similarly, because of the lack of meteorological data in the Silala River basin, historical ETr estimates were calculated using data from a meteorological station of the General Directorate of Water (DGA), which is located at Chiu-Chiu (70 km southwest from the Silala meteorological station (Figure 1-2). The Chiu-Chiu meteorological station has daily temperature records since September 1990 and daily relative humidity records since October 1986. To quantify the relative importance of ETo and ETr in the basin, the ET estimates were compared to the flow of the Silala River measured at the DGA Fluviometric Station, which is located at the Chile-Bolivia international boundary (Figure 1-3). This fluviometric station has nine years of complete data (from 2001 to 2007 and from 2010 to 2011). Because of the difficulty of collecting meteorological data in remote locations such as in the Silala River basin, the estimates provided in this study are only an approximation. The results should therefore be seen in terms of the relative importance of wetland evaporation compared to the other hydrological processes. 1.4 Structure of the report The structure of the remainder of this report is as follows: Chapter 2 states a summary and the conclusions of the study; Chapter 3 presents the methods used to evaluate ETo and ETr; Chapter 4 describes the meteorological data that are available to perform the evapotranspiration estimations; Chapter 5 describes the normalized difference vegetation index (NDVI) data from the wetlands; Chapter 6 presents the main results of this study; Chapter 7 presents the main conclusions of this study. The references cited in the report are shown in Chapter 8. An Appendix containing the calculation of atmospheric parameters is hereto enclosed. 2 SUMMARY AND CONCLUSIONS In this study, we estimated ETo and ETr in the main wetlands of the Silala River basin, i.e., the Orientales and Cajones wetlands. ETo is an idealized estimate of theevapotranspiration from a specified vegetated surface that is actively growing and not short of water, and therefore is determined by atmospheric conditions. ETr is the evapotranspiration that is expected to actually occur, and is normally (for short 446
Annex IX
7
vegetation) less than the potential rate. It depends on vegetation type and condition, and
any effects of water limitation. These estimates are subject to large uncertainties due to
the difficulty of measuring meteorological variables in remote areas. Therefore, they
can only be seen in terms of the relative importance of evaporation processes compared
to the water fluxes due to other hydrological processes, e.g., river flow.
Our results show that ETo during the period November, 2016 to January, 2017,
determined using the data collected at the UC meteorological station and the Penman-
Monteith approach, varied between 2.1 and 7.6 mm day-1, with a daily standard
deviation of 1.1 mm day-1.
The monthly ETo values varied between 3.4  0.4 and 5.9  0.5 mm day-1 (mean 
standard deviation) for winter and summer, respectively. The previous ETo estimates
were determined using all the methods presented in this study, the temperature and
relative humidity data from the Chiu-Chiu station, and the wind speed collected from
the Wind Energy Explorer web platform. The annual ETo determined using all the
methods is 1,685  142 mm year-1 (mean  standard deviation). These values were
obtained using a specific time period that corresponds to the years where simultaneous
records of daily temperature and relative humidity were available, as well as satellite
images that allow estimation of the NDVI, i.e., 1992, 1994, 1995, 1999, 2002, 2004,
2005 and 2006. The monthly and annual results obtained are similar to those presented
by AGRIMED (2015) for towns located in areas near or with similar characteristics to
the study zone. For example, the annual ETo calculated by AGRIMED (2015) ranges
from 1,317 to 1,942 mm year-1. Also, García et al. (2004) measured the grass crop
evapotranspiration under standard conditions without water stress in four zones of the
Bolivian Altiplano and obtained monthly average ETo values of 4.3 mm day-1 for the
time period between October and April.
The monthly ETr values, determined using the methods presented in this study, vary
between 0.6 (winter) and 2.7 mm day-1 (summer) using the Groeneveld et al. (2007)
method. These ETr values represent 19 and 45.4% of the ETo of each season,
respectively. On an annual basis, the ETr is ~496 mm year-1 (29.4% of the annual ETo).
In the Orientales and Cajones wetlands, the highest monthly average of ETr for all the
methods presented, converted to an equivalent flow of water by multiplying the
evaporation rate by the wetland area, is 5.9 l/s and occurs in February. So the highest
monthly water loss to the atmosphere is approximately 3.3% of the river flow at the
Chile-Bolivia international boundary. On an annual basis the average loss is 1.3 l/s
(0.7% of river discharge at the Chile-Bolivia international boundary). Under the
conservative assumption that the wetlands evaporate at the potential ET rate, the highest
loss to the atmosphere occurs during January (11.5  3.9 l/s (mean value  standard
Annex IX
447
8 deviation)) and is approximately 6.5  2.2% (mean value  standard deviation) of the river flow at the Chile-Bolivia international boundary. During the month of January, the monthly average river flow is ~176.7 l/s (calculated using the nine years of complete monthly river flow data). On an annual basis, the ETo loss is equivalent to a flow of 3.4 l/s, or 2% of the average annual discharge. Channelization in the wetlands most likely will lower groundwater levels by less than 0.5 m in the vicinity of the river channel, and this effect will decrease with distance from the channel. It is expected that soil capillarity will retain enough water to support vegetation. The main effect of channelization is therefore expected to be a reduction in the area of ponded water, in the vicinity of the channels. Given that the total wetland losses are rather small in comparison with river flows, it is likely that the impacts of channelization on wetland ET are negligible in comparison with river flows at the border. 3 EVAPOTRANSPIRATION METHODS Evapotranspiration (ET) is the combination of two separate processes that occur simultaneously, whereby water is lost on the one hand from the soil surface by evaporation and on the other hand from the crop by transpiration (i.e. due to evaporation within the plant leaf). Soil evaporation is mainly determined by the fraction of the solar radiation reaching the soil surface and the water availability. Transpiration is mainly determined by the plant cover and the water available to the plant root system (Allen et al., 1998). The ET from a reference surface is called the reference crop evapotranspiration, reference evapotranspiration or potential evaporation (ETo). ETo expresses the evaporating power of the atmosphere for a reference surface at a specific location and time of the year. The reference surface is normally a hypothetical grass reference crop with specific characteristics (Allen et al., 1998). ETo assumes that water is unlimited and is an idealized value that does not depend on crop type, crop development and management practices (for which corrections may be made). ETo values measured or calculated at different locations or in different seasons are comparable as they refer to the ET from the same reference surface. Thus, the only factors affecting ETo are climatic parameters. On the other hand, actual evapotranspiration (ETr) is the amount of water that is actually evaporated from the soil and transpired by the plants, and it is usually smaller (and often much smaller) than ETo. Allen et al. (1998) provide a simple and convenient way to characterize the potential evapotranspiration from natural vegetation, as the product of ETo and a crop coefficient (Kc). The same equation is used to estimate the crop evapotranspiration under standard conditions (well-managed, large, 448
Annex IX
9
well-watered fields that achieve full production under the given climatic conditions), but
the methods for estimating Kc values for natural vegetation are different and are
explained in Allen et al. (1998). The differences between ETo and potential evaporation
for the vegetation type of interest, under the same meteorological conditions, are
typically due to differences in leaf anatomy, stomatal characteristics and aerodynamic
properties of the vegetation (Allen et al., 1998). However, if the vegetation is short of
water, the actual evaporation ETr will be reduced to a fraction of the potential rate.
Evapotranspiration can be estimated using standardized pans designed to measure
evaporation (Epan). Pans provide a measurement of the integrated effect of radiation,
wind, temperature and humidity on the evaporation from an open water surface. The
evaporation rate from pans filled with water is easily obtained but requires qualified
personnel to perform the measurement procedures. In the absence of rain, the amount of
water evaporated during a period (mm day-1) corresponds with the decrease in water
depth in that period. The US Class A pan is probably the most widely used
internationally, and has proved its practical value to estimate reference or potential
evapotranspiration by observing the evaporation loss from a water surface. However,
the Class A pan yields evaporation estimates that are consistently greater than open
water evaporation from shallow lakes (Eagleman, 1967), and is also different from
evaporation from a vegetated surface. Thus to obtain ETo estimates, the pan evaporation
has to be corrected using a pan coefficient (Kp), i.e., ETo = Kp · Epan (Doorenbos and
Priutt, 1977; Allen et al., 1998).
There are many methods available in the literature to determine ET (Allen et al., 1998;
Summer and Jacobs, 2005; Yoder et al., 2005). The Penman-Monteith method is
generally the preferred method to determine ETo as it is based on all of the major
controlling atmospheric variables, such as net radiation, wind speed, relative humidity,
and air temperature, among others (Campbell and Norman, 2012). Indeed, the Penman-
Monteith method is recommended by the FAO for the calculation of ETo (Allen et al.,
1998), and the FAO has developed procedures and recommendations for the use of the
method with limited climatic data. Also, Garcia et al. (2004) demonstrated that the FAO
Penman–Monteith approach is suitable in the Bolivian Altiplano. However, in remote
areas such as in the Silala River basin, the available data are scarce and no direct
measurement of evaporation from open water springs and associated vegetation is
currently available near the study site (Houston, 2006), so other empirical methods that
require a smaller number of atmospheric variables are often the only suitable option
(e.g., see DGA, 2013). For this reason, in this study several methods were evaluated, to
assess the uncertainty associated with the evaporation estimates, even when some of
them apply to other situations such as open water bodies.
Annex IX
449
10 In this section we present the methods that were used to determine ETo and ETr with the aim of estimating how much water is lost to the atmosphere in the wetlands of the Silala River basin. These methods were selected because they are the most commonly used. It should be noted that data are limited in remote locations such as in the Silala River basin and thus, any estimation of ET will be uncertain. For this reason, the estimates presented in this report should be viewed in relative terms in the context of an uncertainty analysis. 3.1 Penman-Monteith (ETo) The Penman-Monteith approach calculates ETo (mm day-1) using the following equation (Allen et al., 1998): ETo=0.408Δ(Rn−G)+γ900T+273u2(es−ea)Δ+γ(1+0.34u2)(2.1) where Rn is the net radiation at the crop surface [MJ m-2 day-1]; G is the soil heat flux density [MJ m-2 day-1]; T is the mean daily air temperature measured at 2 m height [℃]; u2 is the wind speed at 2 m height [m s-1 ]; es and ea are the saturation and actual vapor pressure, respectively, and the term (es – ea) is called the vapor pressure deficit [kPa]; Δ is the slope of the saturated vapor pressure-temperature curve [kPa ℃-1] and γ is the psychometric constant [kPa ℃-1]. 3.2 Turc (ETo) The ETo in method proposed by Turc is given by (Yoder et al., 2005): ETo=aT0.013TT+15(23.8856Rs+50)λ(2.2) where Rs is the solar radiation [MJ m-2 day-1]; λ is the latent heat of vaporization (~2.45 MJ kg-1); aT = 1 when the daily relative humidity (RH) is larger or equal to 50% and aT=1+(50+𝑅𝑅𝑅𝑅)70⁄ when RH  50%. 3.3 Priestley-Taylor (ETo) The Priestley-Taylor approach is given by the following formula (Summer and Jacobs, 2005): ETo=αΔΔ+γ(Rn−G)1λ(2.3) 450
Annex IX
11
where α is the Priestley-Taylor empirically derived dimensionless constant and all the
other variables are defined above.
3.4 Taylor-de Bruin (ETo)
The Taylor-de Bruin formula is typically used to determine evaporation from open
water bodies and is given by (de Bruin and Keijman, 1979):
ETo = Δ
0.85Δ+0.63γ
(Rn−Qx)
λ ρ 86.4 (2.4)
where Qx is the change in heat stored in the water body [W m-2]; and ρ is the water
density (1000 kg m-3). The Taylor-de Bruin method can be used to determine ETo of a
wetland when Qx is replaced by G.
3.5 Jensen-Haise (ETo)
The Jensen-Haise method determines ETo using the following equation (McGuinness
and Bordne, 1972):
ETo = (0.014T − 0.37)(Rs 3.523 x 10−2) (2.5)
where all the parameters have been previously defined.
All the calculation procedures related to the determination of atmospheric parameters
are presented in Appendix A.
3.6 Actual evapotranspiration (ETr)
Determining actual evapotranspiration (ETr) is challenging as it depends on many
biophysical processes, including the effects on vegetation of water limitation (Campbell
and Norman, 2012). One approach for the estimation of ETr is to combine the estimates
of ETo with the normalized difference vegetation index (NDVI [-]) (Senay et al., 2011).
The NDVI is a vegetation index that determines the surface extent of the vegetation.
NDVI is based on the ratio of the reflectance in the red band and the reflectance in the
near infrared (NIR) band; vegetation strongly absorbs red light and reflects NIR,
providing a quantitative measure of green plant cover over a landscape. The raw NDVI
is normalized by setting bare soil values at 0 (NDVI = 0) and values for full vegetation
cover at 1.0 (NDVI = 1.0). To estimate annual ETr, Groeneveld et al. (2007) used
equation (3.1):
ETr = (ETo − Annual Precipitation) ∗ NDVI + Annual Precipitation (3.1)
Groeneveld et al. (2007) estimated ETr of a wide variety of phreatophyte communities
in arid and semi-arid shallow groundwater environments in the western USA.
Annex IX
451
12 Phreatophytes communities obtain water from groundwater. They are typically located in arid regions along rivers and areas where the groundwater table is very shallow such as in wetlands, which are environments similar to that found at the source of Silala River, i.e., at both the Orientales and Cajones wetlands. Groeneveld et al. (2007) found that annual phreatophyte ET predicted the actual ET at moisture flux tower sites with a coefficient of determination (r2) of 0.95; they also stated that their method can be applied to regions where sufficient meteorological data are available for calculation of annual values of precipitation and ETo. Although these formulas use an annual temporal scale, in this study we estimated ETr using a monthly temporal scale because of the availability of the satellite images that were used to estimate the NDVI of the wetlands at the headwaters of the Silala River (Alcayaga, 2017). 4 AVAILABILITY OF METEOROLOGICAL DATA 4.1 Meteorological data collected in this study As shown in Figure 4-1, a meteorological station, named “UC meteorological station”, was installed in the Silala River basin (22.03°S 68.04°W). The meteorological station was installed alongside the Silala River, in the ravine near the junction with Quebrada Negra (Figure 1-3). The UC meteorological station has a precipitation gauge, an anemometer, a relative humidity and air temperature sensor, a pressure transducer to measure atmospheric pressure, and a net radiometer that is installed over a riparian wetland area alongside the river. The meteorological station was fully operational on November 4th, 2016. 452
Annex IX
13
Figure 4-1. Meteorological station installed for this study: (a) photograph of the site where the
meteorological station was installed; (b) diagram showing the position of each instrument. A:
anemometer; TH: temperature and relative humidity sensor; P: precipitation gauge; (c)
photograph of the precipitation gauge (installed at an elevation of 0.75 m); (d) photograph of
the anemometer (elevation of 2.0 m) and of the temperature and relative humidity sensor
(elevation of 1.48 m); and (e) photograph of the net radiometer (NR) that was installed at an
elevation of 2 m.
The precipitation gauge and the anemometer correspond to the Vantage Pro2 system
(Davis Instruments, IL, USA). The precipitation gauge was installed at an elevation of
0.75 m above the ground. This sensor (tipping bucket) measures precipitation and
cumulative precipitation at time intervals of 1 h. The anemometer was installed at an
elevation of 2.0 m above the ground. The anemometer measures mean wind speed
during time intervals of 1 h. It also records the maximum instantaneous velocity over a
time interval of 1 h. The air temperature and relative humidity sensor is the HOBO U23
Pro v2 Temperature/Relative Humidity Data Logger (Onset Computer Corporation,
Bourne, MA). This sensor was installed at an elevation of 1.48 m above the ground and
collects data at 15-min intervals. The pressure transducer is the HOBO Water Level (0-4
m) Data Logger (Onset Computer Corporation, Bourne, MA), which allows measuring
pressure both in the water and in the air. The pressure transducer was installed at an
elevation of approximately 1.7 m and collects data at 15-min intervals. The net
radiometer (NR Lite 2 Net Radiometer, Kipp & Zonen, The Netherlands) was installed
at an elevation of 2.0 m above the vegetation of the wetland. Additionally, the
meteorological station was enclosed using a fence (1 m tall and 30 m2) to protect the
instruments. With the exception of the anemometer, all the instruments were installed
following the recommendations of the World Meteorological Organization (WMO,
Annex IX
453
14 2008). On the other hand, the anemometer was installed following the recommendations of the American Association of State Climatologists (AASC) (The State Climatologist, 1985). 4.2 Meteorological data collected by other institutions or studies 4.2.1 Temperature and relative humidity The General Directorate of Water (DGA is its acronym in Spanish) meteorological stations that are near the Silala River basin (i.e., Silala, Inacaliri, Quebrada Negra, Mirador Silala, Línzor and Toconce stations) do not have a long record of measurements. In addition, the quality of the data is not as good as other stations that are located in the region. DGA advised us to use the daily temperature and relative humidity data from the Chiu Chiu meteorological station because it is the closest station to the Silala basin with reliable data. Chiu Chiu station is located 70 km southwest from the Silala meteorological station (Figure 1-2). This station has a daily temperature record since September 1990 and daily relative humidity record since October 1986, but also presents gaps in its records that were infilled. Missing average temperature data from Chiu Chiu were infilled using three meteorological stations (Inacaliri, Toconce and Línzor) that are located near the study site (Table 4-1) and using the regional temperature gradient estimated by the DGA (2008). The time period analyzed corresponds to the years where simultaneous records of daily temperature and relative humidity were available, as well as satellite images that allow estimation of the NDVI, i.e., 1992, 1994, 1995, 1999, 2002, 2004, 2005 and 2006. Meteorological station Latitude Longitude Altitude Meteorological station Latitude Longitude Altitude Chiu Chiu 22°20’S 68°38’W 2,524 Toconce 22°16’S 68°10’W 3,310 Línzor 22°13’S 68°01’W 4,100 Inacaliri 22°01’S 68°’04’W 4,040 Table 4-1. Meteorological stations used to infill missing temperature data. Altitude is in meters above sea level (m.a.s.l.). The missing data were infilled using the approach proposed by Fernández and Salas (1995): Y1̂=α+Y2β+ε Sy√1−R2 (4.1) 454
Annex IX
15
where Y1 ̂ is the variable to be filled, Y2 is the predictor variable,  and  are the
regression coefficients,  is a random error ~N(0,1), Sy is the standard deviation of the
original Y1 and R2 is the coefficient of determination.
In terms of relative humidity (RH), according to the recommendations made by the
DGA, we used the RH synoptic data measured at 8:00 and 14:00 h as the daily
maximum and minimum, respectively. Radiation estimates were obtained through the
methods described in Appendix A.
4.2.2 Wind speed data
Since the DGA did not measure wind speed at the Silala River basin, to estimate ETo for
the historical calculations using the Penman-Monteith method we used the data
provided by the Wind Energy Explorer web platform
(http://walker.dgf.uchile.cl/Explorador/Eolico2/), which publicly delivers wind speed
data throughout the Chilean territory at different timescales (hourly and monthly) and
for different heights above the ground (it has 12 vertical levels from 0 to 200 m). The
Wind Energy Explorer was developed by the Department of Geophysics of the
Universidad de Chile by instruction of the Chilean Ministry of Energy and with the help
of the German Agency for International Cooperation (GIZ). The wind speed
information was constructed with numerical simulations of the Weather Research and
Forecasting (WRF) model version 3.2, with a horizontal resolution of 1 km and vertical
resolution of 10 m. They simulated the wind speed for a full year (2010), the results of
which were validated at more than 350 sites with local observations. Also, using the
same model, they provided estimates of wind speed for the time period between 1980
and 2013 for the entire Chilean territory. Since the wind speed data used to determine
the historical calculations of ETo are simulated data, the Penman-Monteith estimates
should be treated with caution. Nonetheless, as previously noted the estimates presented
in this report should be viewed in relative terms in the context of an uncertainty
analysis.
Figure 4-2 shows the daily wind speed at 2 m above the ground for selected years,
obtained from the simulation of the Wind Energy Explorer web platform. The values
range from 2 to 11 m s-1.
Annex IX
455
16 Figure 4-2. Daily wind speed at 2 m above the ground for selected years in the wetlands of Silala River basin. 4.2.3 Monthly Precipitation The monthly precipitation was used to estimate ETr using equation (3.1). To be consistent in the calculations, the monthly precipitation was obtained from the DGA’s Silala meteorological station that is at the Chile-Bolivia international boundary (22.01°S 68.03°W) (Figure 1-1). Table 4-2 shows the monthly and annual precipitation registered at this station for the analyzed period. Precipitation Silala Station (mm) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual 1992 24 0 0 6 2 3 0 0 0 0 0 18 54 1994 23 16 15 14 0 2 0 10 6 0 0 16 102 1995 106 0 10 0 3 0 0 0 8 0 0 0 127 1999 0 139 83 0 0 0 0 0 0 0 0 19 241 2002 2 30 43 0 0 1 6 0 0 1 1 0 84 2004 21 59 5 0 0 0 0 9 0 0 0 0 93 2005 67 33 12 0 0 0 0 0 2 0 0 2 114 2006 69 38 5 2 0 0 0 0 0 0 0 0 114 Table 4-2. Monthly and annual precipitation at the Silala meteorological station. 456
Annex IX
17
4.2.4 Monthly Streamflow
To estimate the relative importance of wetland ETo and ETr for the basin, the ET was
compared to the average monthly river flow measured at the DGA’s fluviometric station
located at the Chile-Bolivia international boundary: DGA Fluviometric Station (22.01°S
68.03°W). Table 4-3 shows the average monthly river flow for nine years of complete
data: from 2001 to 2007 and from 2010 to 2011 (Muñoz et al., 2017).
River flow Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Mean (l/s) 176.7 178.8 176.8 176.0 176.5 172.5 166.5 165.8 170.5 173.1 174.5 170.8
Table 4-3. Monthly streamflow in DGA fluviometric station.
5 NDVI DATA
To estimate Silala River headwater wetland ETr, Landsat images were used to estimate
the NDVI and the wetlands’ surface extent (Alcayaga, 2017). These images enabled the
determination of the spatial distribution of NDVI, which was calculated for the
following ranges: 0.10-0.20, 0.20-0.29, 0.30-0.39, 0.40-0.49, 0.50-0.59 and 0.6-1.00.
NDVI values greater than 0.1 represent zones with vegetation. Alcayaga (2017) presents
the wetland surface extent, as a function of the NDVI, for all the Landsat images found
between 1987 and 2016. Here, we selected periods where both Landsat images and
meteorological data were simultaneously available (between 1992 and 2006) to estimate
ETo (Table 5-1). In Table 5-1 (and in Alcayaga (2017), Figures 16 and 17), it can be
seen that there is much more vegetation during the months of January, February and
March, while less vegetation occurs between June and September. Figure 5-1 presents
an example of the NDVI images reported by Alcayaga (2017).
Annex IX
457
18 Surface (km2) per NDVI Values Range Month ≤ 0.1 0.1 - 0.2 0.2 - 0.3 0.3 - 0.4 0.4 - 0.5 0.5 - 0.6 > 0.6 Total January 0.781 0.102 0.042 0.014 0.006 0.001 0.000 0.165 February 0.799 0.068 0.051 0.022 0.006 0.003 0.000 0.147 March 0.793 0.090 0.041 0.016 0.005 0.001 0.000 0.153 April 0.891 0.039 0.013 0.004 0.000 0.000 0.000 0.055 May 0.920 0.025 0.002 0.000 0.000 0.000 0.000 0.027 June 0.932 0.013 0.001 0.000 0.000 0.000 0.000 0.014 July 0.936 0.010 0.000 0.000 0.000 0.000 0.000 0.010 August 0.938 0.008 0.000 0.000 0.000 0.000 0.000 0.008 September 0.932 0.014 0.000 0.000 0.000 0.000 0.000 0.014 October 0.923 0.023 0.001 0.000 0.000 0.000 0.000 0.023 November 0.932 0.011 0.004 0.000 0.000 0.000 0.000 0.015 December 0.909 0.027 0.011 0.001 0.000 0.000 0.000 0.038 Average 0.890 0.036 0.014 0.005 0.001 0.000 0.000 0.056 Maximum 0.938 0.102 0.051 0.022 0.006 0.003 0.000 0.165 Minimum 0.781 0.008 0.000 0.000 0.000 0.000 0.000 0.008 Standard Dev. 0.062 0.033 0.019 0.008 0.003 0.001 0.000 0.062 Table 5-1. Average of the monthly wetland area (km2) used to estimate evapotranspiration. 458
Annex IX
19
Figure 5-1. Example of the NDVI images for the main wetlands in the Silala River basin
(Alcayaga, 2017). The wetlands shown are the Orientales and Cajones that are located in
Bolivian territory.
6 RESULTS
6.1 Potential ET using the Penman-Monteith formula and the meteorological
data collected in this study
Figure 6-1 presents the ETo results obtained with the meteorological data collected for
this study (November, 2016 to January, 2017). In general, the daily ETo that are less
than 3 mm day-1 are due to days with low net radiation. The estimated ETo ranges
Annex IX
459
20 between 2.1 and 7.6 mm day-1, which represent 3.8 and 13.7%, respectively, of the average daily flow for this period (0.18 m3 s-1). This percentage was obtained considering that ETo occurs in the wetlands area where water is readily available and using an area of the wetlands of 0.28 km2, which corresponds to the maximum area measured by Alcayaga (2017) with Landsat images. Johnson et al. (2010) measured evaporation rates in the Chilean Altiplano using the chamber method. Their focus was to determine evaporation from bare soils in zones where shallow groundwater exists. For the sites located in the II Region of Antofagasta, where the depth to groundwater was zero, they measured daily values that ranged between 6.3 and 7.1 mm day-1 (during December 2007), which are within the range of the estimated ETo values at the Silala River basin. The monthly average ETo in both November and December 2016 was 5.6 mm day-1, while in January 2017 it was 4.2 mm day-1. The standard deviation of the whole series is 1.1 mm day-1 (2.0 % of average daily flow for this time period). The DGA (1987) reported a mean annual pan evaporation of 6.7 mm day-1 at Inacaliri, with mean pan evaporation rates (Epan) of 9.7, 9.6 and 7.0 mm day-1 during November, December and January, respectively. These values can be compared with ETo using a pan coefficient. Table 5 of Allen et al. (1998) presents pan coefficients (Kp) for Class A Pan for different ground covers and levels of daily relative humidity and wind speed. Using the data collected in the UC meteorological station (wind speed of ~2.8 m s-1 and relative humidity of ~40%), we estimated that Kp varies between 0.70 and 0.85. Therefore, the ETo values estimated using Epan are ~7.5, ~7.4 and ~5.4 mm day-1 for November, December and January, respectively. These values are slightly larger than those obtained using the Penman-Monteith approach populated with the data measured at the UC meteorological station. Figure 6-1. Potential evapotranspiration (ETo) using the Penman-Monteith formula using the data collected for this study (November 2016-January 2017). 460
Annex IX
21
6.2 Potential ET using the meteorological data collected by the DGA
Figure 6-2 shows the monthly mean ETo for the analyzed period (1992, 1994, 1995,
1999, 2002, 2004, 2005 and 2006) using the five different approaches. Table 6-1 shows
the monthly mean, the inter-annual standard deviation and the annual mean of each
method. ETo ranges between 2 mm day-1 (winter) and 8 mm day-1 (summer). The ETo
calculated with the Penman-Monteith method has larger annual average values than the
other methods and shows less variability throughout the year. The Penman-Monteith
method results in a maximum ETo of 7.3 mm day-1, a minimum ETo of 5.0 mm day-1
and a mean standard deviation of 0.6 mm day-1 (11% of the annual mean ETo calculated
with this method). The Turc method has the highest mean standard deviation (0.8 mm
day-1, which corresponds to 22% of the annual mean ETo calculated with this method).
The other methods (Priestley –Taylor, Taylor-de Bruin and Jensen–Haise) show that
ETo is lower in winter and higher in summer, and have a similar mean standard
deviation of around 0.3-0.4 mm day-1. The annual totals in Table 6-1 ranged from 1304
± 210 (Turc) to 2164 ± 175 (Penman-Monteith) mm/year (mean ± standard deviation).
The approaches of Penman-Monteith and Taylor-de Bruin present an annual ETo greater
than 2,000 mm/year.
Figure 6-2. Monthly Mean Potential Evapotranspiration (ETo) from five estimation methods.
Annex IX
461
22 Method Average Month Penman-Monteith Turc Priestley-Taylor Taylor-de Bruin Jensen-Haise Jan 6.7 ± 0.8 4.7 ± 0.5 5.6 ± 0.4 7.4 ± 0.5 5.1 ± 0.5 5.9 ± 0.5 Feb 6.2 ± 0.9 4.1 ± 0.6 5.2 ± 0.4 6.8 ± 0.5 4.7 ± 0.5 5.4 ± 0.6 Mar 5.9 ± 0.6 3.3 ± 1.4 4.4 ± 0.4 5.8 ± 0.5 4.2 ± 0.5 4.7 ± 0.7 Apr 5.6 ± 0.5 3.4 ± 0.7 3.5 ± 0.4 4.8 ± 0.3 3.5 ± 0.6 4.2 ± 0.5 May 5.2 ± 0.6 2.9 ± 0.7 2.7 ± 0.3 3.7 ± 0.3 2.8 ± 0.4 3.5 ± 0.5 Jun 5 ± 0.5 2.6 ± 0.6 2.2 ± 0.1 3.1 ± 0.1 2.5 ± 0.2 3.1 ± 0.3 Jul 5.2 ± 0.6 2.5 ± 0.7 2.3 ± 0.2 3.3 ± 0.2 2.4 ± 0.3 3.1 ± 0.4 Aug 5.7 ± 0.5 3.2 ± 0.7 3.1 ± 0.2 4.2 ± 0.2 3.1 ± 0.4 3.9 ± 0.4 Sep 6.4 ± 0.9 3.9 ± 1 4.1 ± 0.2 5.7 ± 0.2 3.9 ± 0.5 4.8 ± 0.6 Oct 6.7 ± 0.9 4 ± 1.9 5.1 ± 0.3 6.9 ± 0.3 4.7 ± 0.4 5.5 ± 0.8 Nov 7.1 ± 0.6 5.1 ± 0.7 5.6 ± 0.4 7.6 ± 0.4 5.2 ± 0.5 6.1 ± 0.5 Dec 7.3 ± 0.5 5.2 ± 0.5 5.9 ± 0.4 7.7 ± 0.4 5.5 ± 0.4 6.3 ± 0.4 Average 6.1 ± 0.6 3.7 ± 0.8 4.1 ± 0.3 5.6 ± 0.3 4 ± 0.4 4.7 ± 0.5 Annual (mm/year) 2164 ± 175 1304 ± 210 1506 ± 93 2033 ± 95 1419 ± 138 1685 ± 142 Table 6-1. Monthly mean  inter-annual standard deviation of ETo (mm day-1). The monthly and annual values obtained in this study are similar to those presented by AGRIMED (2015) for locations in areas near or with similar characteristics to the study site (located in the Altiplano and above 2,450 m.a.s.l.). AGRIMED (2015) used the Penman-Monteith method to calculate ETo on an monthly and annual basis. ETo varied from 3 mm day-1 in winter to 6.3 mm day-1 in summer in Putre, San Pedro de Atacama and Socompa towns. Also the annual ETo in these locations varied between 1,294 and 1,942 mm year-1. These results are similar to those obtained by García et al. (2004), through lysimetric measurements carried out in four zones of the Bolivian Altiplano. They measured the grass crop evapotranspiration under standard conditions without water stress and obtained a monthly average of 4.3 mm d-1 for the months of October to April. Note that the precipitation in the Bolivian Altiplano sites ranged between 363 and 450 mm year-1. These precipitation rates are greater than that observed at the Silala River basin, which is of ~165 mm year-1 (Muñoz et al., 2017). 462
Annex IX
23
6.3 Actual ET using the meteorological data collected by the DGA and the NDVI
Using the ETo methods described above, the monthly ETr was estimated using the
Groeneveld et al. (2007) method (equation (3.1)). Because the NDVI was reported as a
range (e.g., between 0.30 and 0.39), we used the upper boundary as a conservative
estimate of monthly NDVI. We calculated the monthly NDVI using a weighting
between the surface area (km2) per NDVI value range, multiplied by the upper boundary
of the NDVI range, and divided by the total area of wetlands (equation (4.1)).
Therefore, our estimates are considered to be an upper limit of ETr.
NDVI =
Σ NDVIUpperBoundaryi ∗𝐴𝐴𝐴𝐴𝐴𝐴𝑎𝑎𝑖𝑖
𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 𝑟𝑟𝑟 𝑟𝑟𝑟𝑟𝑟𝑟
𝑖𝑖=𝑟𝑟𝑟 𝑟𝑟𝑟𝑟𝑟𝑟 1
𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑇𝑇 𝐴𝐴𝐴𝐴𝐴𝐴𝐴 (4.1)
Figure 6-3 and Table 6-2 show the monthly mean ETr calculated with the available
NDVI index for the analyzed periods (1992, 1994, 1995, 1999, 2002, 2004, 2005 and
2006) using equation (3.1). Table 6-3 shows the average percentage of ETr / ETo for
each month and each equation.
Method
Average
Month Penman-
Monteith Turc Priestley-
Taylor
Taylor-de
Bruin
Jensen-
Haise
Jan 2.5 1.9 2.2 2.6 2.1 2.2
Feb 3.9 3.6 4.0 4.4 3.8 3.9
Mar 1.8 1.2 1.3 1.7 1.3 1.5
Apr 1.3 0.9 0.9 1.2 0.9 1.0
May 1.1 0.6 0.6 0.8 0.6 0.7
Jun 1.0 0.5 0.4 0.6 0.5 0.6
Jul 1.0 0.4 0.4 0.6 0.4 0.6
Aug 1.1 0.6 0.6 0.8 0.6 0.7
Sep 1.4 0.8 0.9 1.2 0.8 1.0
Oct 1.3 0.9 1.0 1.4 0.9 1.1
Nov 1.6 1.1 1.1 1.5 1.0 1.3
Dec 2.2 1.6 1.7 2.1 1.6 1.8
Average 1.7 1.2 1.3 1.6 1.2 1.4
Annual
(mm/year) 602.9 426.2 450.6 566.5 434.5 496.1
Table 6-2. Monthly mean ETr (mm day-1) using the Groeneveld et al. (2007) method.
Annex IX
463
24 In Figure 6-3, ETr estimates range between 0.4 mm day-1 (winter) and 4.4 mm day-1 (summer). The monthly standard deviation depends on the monthly availability of the NDVI images. January is the month with the highest number of images for the analyzed period (5 images), and has one of the highest NDVI values with a mean standard deviation among all methods of 1.1 mm day-1 (corresponding to 49 % of the ETr monthly average). It is observed that the value of ETr is higher in February with all the evaluated methods, which occurs because the modified equation of Groeneveld et al. (2007) for a monthly time scale uses the monthly precipitation, and one of the two satellite images evaluated for February corresponds to the year 1999 where the monthly precipitation in the Silala meteorological station was very high (139 mm). Figure 6-3. Monthly mean of the actual evapotranspiration (ETr) using the Groeneveld et al. (2007) method. Table 6-3 shows that ETr estimates represent at least 19% of ETo (winter) and at most 81.4% of ETo (February). The ETr estimates of all the months except February are similar to those obtained by Groeneveld et al. (2007) for an annual time scale. They obtained an annual average ratio of ETr/ETo of 17.4% in Owens Valley, California and 32.6% in San Luis Valley, Colorado; each of these areas has an arid climate where vegetation is sustained by shallow groundwater. On an annual basis, the ETr estimates range between 426 mm year-1 (Turc) and 603 mm year-1 (Penman-Monteith). On average, using all the approaches presented in this study, the annual ETr is of 496 mm year-1 (29.4 % of annual ETo). 464
Annex IX
25
ETr / ETo (%)
Jan 36.9
Feb 81.4
Mar 29.1
Apr 22.8
May 20.8
Jun 19.3
Jul 19.0
Aug 19.0
Sep 21.1
Oct 19.2
Nov 21.4
Dec 30.4
Average 28.4
Table 6-3. Monthly mean ratio of ETr / ETo for all methods expressed as a percentage using the
Groeneveld et al. (2007) method.
6.4 ETo and ETr, expressed as equivalent flow in the wetlands and percentage of
the monthly flow measured in DGA fluviometic station.
The evaporation losses from the wetlands can be considered as an equivalent
streamflow, in the sense that if the evaporation losses had not occurred, additional water
would in principle be available to the river. The simple calculation takes the product of
the monthly evaporation rates and the NDVI-derived wetland area to derive monthly
equivalent flows of ETo and ETr (in L s-1 – litres per second) for the wetlands. Table 6-4
shows the monthly and yearly averages of ETr flows (in l/s) using the Groeneveld et al.
(2007) method and the wetland surface obtained from the Landsat images. These water
flows are presented for each ETo method, and are also compared to the monthly and
yearly river flows. The ETr estimates are higher in the summer months and lower in the
winter months, reaching values of 5.9 l/s and 0.1 l/s, respectively. These values
represent 3.3% and 0% of the monthly river flow. The yearly ETr is of ~1.2 l/s, which
represents 0.7% of the yearly river flow. These values are the best estimation of
evaporation losses under the uncertainty context of the data used to determine these
estimates. The average standard deviation of ETr for all methods in January is 0.74 l/s.
Annex IX
465
26As a consequence, one would expect that wetland ET should be less than 3.3% of the monthly river flow. MonthMETHODAveragePercentage of streamdischarge Penman-MonteithTurcPriestley-TaylorTaylor-deBruinJensen-HaiseETr(l/s) using equation (3.1) Jan4.13.13.64.43.33.72.1Feb5.85.36.06.85.75.93.3Mar3.12.22.32.92.32.61.5Apr0.80.60.60.80.60.70.4May0.30.20.20.20.20.20.1Jun0.20.10.10.10.10.10.1Jul0.10.10.00.10.10.10.0Aug0.10.10.10.10.10.10.0Sep0.20.10.10.20.10.20.1Oct0.30.20.30.40.30.30.2Nov0.50.40.40.50.30.40.2Dec1.00.70.70.90.70.80.5Annual1.41.11.21.41.11.20.7Table 6-4. Average monthly ETr, expressed as an equivalent streamflow (l/s) using the Groeneveld et al. (2007) method combined with the five different ETomethods, and the percentage of the monthly and yearly streamflow of the mean of all methods. On the other hand, to indicate an upper bound to the wetland evaporation, the potential evaporation rates can be assumed to apply. Table 6-5 shows the monthly averages of ETo obtained for each ETo method, and the relative importance of ET compared to the monthly river flow. ETo values also are higher in the summer months and lower in the winter months, reaching values of 11.5l/s and 0.4 l/s respectively. These values are the most conservative estimates of evaporation losses and represent 6.5% and 0.2% of the monthly river flow. The average standard deviation of ETo for all methods in January is 3.9l/s. If, further, allowance is made for the uncertainty in ETo estimates, the evapotranspiration loss from the wetlands should be less than 8.7% (mean + standard deviation) of the average river flows. In fact, on an annual basis, theETo represents only 2% of the annual streamflow. Hence, it is likely that any changes in the magnitude of 466
Annex IX
27
this hydrological process, e.g., due to channelization, would be negligible compared to
surface water discharge or groundwater flow at the basin scale.
Month
METHOD
Average Percentage of
Penman- Monthly Stream
Monteith Turc Priestley-
Taylor
Taylor-de
Bruin
Jensen-
Haise
ETo (l/s)
Jan 13.2 9.0 10.9 14.4 9.9 11.5 6.5
Feb 8.1 6.3 8.8 11.6 7.6 8.5 4.7
Mar 11.0 7.3 7.9 10.3 7.8 8.9 5.0
Apr 3.7 2.5 2.4 3.3 2.5 2.9 1.7
May 1.6 0.9 0.8 1.1 0.9 1.1 0.6
Jun 0.8 0.4 0.4 0.5 0.4 0.5 0.3
Jul 0.6 0.3 0.3 0.4 0.3 0.4 0.2
Aug 0.6 0.3 0.3 0.4 0.3 0.4 0.2
Sep 1.1 0.6 0.7 0.9 0.6 0.8 0.5
Oct 1.8 1.2 1.4 2.0 1.3 1.6 0.9
Nov 2.5 1.8 1.7 2.4 1.5 2.0 1.1
Dec 3.4 2.2 2.4 3.2 2.2 2.7 1.6
Annual 4.0 2.7 3.1 4.2 2.9 3.4 2.0
Table 6-5. Average monthly ETo, expressed as an equivalent streamflow (l/s), obtained with the
five different ETo methods and the percentage of the monthly and yearly streamflow of the mean
of all methods.
6.5 Potential impacts of channelization on wetland ET
One important aspect to analyze in the Silala River basin is the impact of channelization
on wetland ET. When the river is channelized, it is likely to have lower water levels in
the channel compared to those observed in the river. Therefore, one would expect lower
groundwater levels in the vicinity of the channel. Nonetheless, as one moves away from
the river, groundwater levels are not expected to be affected by the channelization. In
general, the Silala River flows have depths below ground surface that are shallower than
0.5 m and thus, it is likely that the groundwater levels will not be lowered by more than
this depth. As a result, it is likely that soil capillarity will maintain soil moisture
conditions that will still be able to provide water to the root system of the riparian
vegetation. However, some reduction in areas of ponded water would be expected.
Annex IX
467
28 Overall it is likely that the impacts of channelization on wetland ET are negligible in comparison with river flows at the international border. 7 CONCLUSIONS In this study, we estimated potential and actual evapotranspiration (ETo and ETr) rates for the Cajones and Orientales wetlands in the headwaters of the Silala River, located in Bolivia. These estimates are subject to large uncertainties due to the difficulty of measuring meteorological variables in remote areas. Therefore, they can only be seen in terms of relative importance compared to the water flow of other hydrological processes, e.g., surface runoff. The following conclusions can be drawn from this study: The potential evapotranspiration, ETo, during November 2016 to January 2017,determined using the local data collected for this study and the Penman-Monteithapproach, varied between 2.1 and 7.6 mm day-1, with a daily standard deviation of1.1 mm day-1.The monthly ETo values, determined using the methods presented in this study, themeteorological data from the study site and the Chiu-Chiu station, and the WindEnergy Explorer web platform, varied between 2 and 8 mm day-1. ETo is lower inwinter and higher in summer, with a mean standard deviation that ranges between0.3 and 0.8 mm day-1. The annual ETo ranged from 1304 ± 210 to 2164 ± 175 mmyear-1 (mean ± standard deviation). These values are similar to those obtained byAGRIMED (2015) in towns near or with similar characteristics to the Silala Basin,and by Garcia et al. (2004) in four zones of the Bolivian Altiplano. The monthly ETrvalues, determined using the methods presented in this study, vary between 0.6(winter) and 2.7 mm day-1 (summer), which represent 19% and 45.4% of the ETo ofeach season, respectively.On an annual basis, ETr values, determined using the methods presented in thisstudy, vary between 426 and 603 mm year-1 using the Groeneveld et al. (2007)method. The annual average represents 29.4% of the annual ETo.ETr is the best estimate of evaporation losses in the wetlands. On an annual basis theaverage loss is 1.3 l/s (0.7% of river discharge at the Chile-Bolivia internationalboundary). The highest monthly average of ETr, expressed as an equivalent flow, forall the methods presented is 5.9 l/s and occurs in February, i.e., approximately 3.3%of the river flow at the Chile-Bolivia international border.468
Annex IX
29
 On the other hand, ETo is a conservative estimation of evaporation losses for
wetlands. On an annual basis, the ETo loss is equivalent to a flow of 3.4 l/s, or 2% of
the average annual river flow at the Chile-Bolivia international border. The highest
monthly average ETo corresponds to a flow of 11.5  3.9 l/s (mean value  standard
deviation), which occurs in January. This is approximately 6.5  2.2% (mean value
 standard deviation) of the corresponding monthly flow at the international border.
The other months have smaller ETo values, so they represent a lesser percentage of
the monthly river flow, because the monthly streamflow measured at the DGA
fluviometric station is nearly constant (Muñoz et al., 2017).
 Since channelization most likely will lower groundwater levels by less than 0.5 m in
the vicinity of the river channel, with a decreasing effect with distance from the
channel, it is expected that soil capillarity will retain enough water to support
vegetation. However, some reduction in area of ponded water in the riparian
wetlands could be expected. Given that total wetland evaporation is a small
percentage of river flow, it is likely that the overall impacts of channelization on
wetland ET are negligible in comparison with river flow at the international border.
Annex IX
469
30 8 REFERENCES AGRIMED, 2015. Evapotranspiración de referencia para la determinación de las demandas de riego en Chile. Facultad de Ciencias Agronómicas, Universidad de Chile. Alcayaga, H., 2017. Characterization of the Drainage Patterns and River Network of the Silala River and Preliminary Assessment of Vegetation Dynamics Using Remote Sensing. (Vol. 4, Annex I). Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evapotranspiration-Guidelines for Computing Crop Water Requirements-FAO Irrigation and Drainage Paper 56. FAO, Rome, 300(9), D05109. Campbell, G.S. and Norman, J. M., 2012. An Introduction to Environmental Biophysics. Springer Science & Business Media. De Bruin, H.A.R. and Keijman, J.Q., 1979. The Priestley-Taylor evaporation model applied to a large, shallow lake in the Netherlands. Journal of Applied Meteorology, 18(7), 898-903. DGA (General Directorate of Water), 1987. Balance Hídrico de Chile. Available at: http://documentos.dga.cl/SUP1540.pdf. Last accessed: April, 2017. DGA (General Directorate of Water), 2013. Actualización del Balance Hídrico Nacional Considerando Cambio Climático: Diagnóstico. Available at: http://documentos.dga.cl/Informe_BH.pdf. Last accessed: April, 2017. Doorenbos, J. and Priutt, W.O., 1977. Guidelines for predicting crop water requirements. Irrigation and Drainage Paper 24, Food and Agriculture Organization of the United Nations. Eagleman, J.R., 1967. Pan evaporation, potential and actual evapotranspiration. Journal of Applied Meteorology, 6, 482-488. Fernández, B. and Salas, J., 1995. Transferencia de información hidrológica por correlación. Revista de la Sociedad Chilena de Ingeniería Hidráulica, Vol. 10 Nº1, Julio 1995. Johnson, E., Yáñez, J., Ortiz, C., Muñoz, J., 2010. Evaporation from shallow groundwater in closed basins in the Chilean Altiplano. Hydrological Sciences Journal–Journal des Sciences Hydrologiques, 55(4), 624-635. 470
Annex IX
31
Garcia, M., Raes, D., Allen, R., & Herbas, C., 2004. Dynamics of reference
evapotranspiration in the Bolivian Highlands (Altiplano). Agricultural and Forest
Meteorology, 125(1), 67-82.
Groeneveld, D.P., Baugh, W.M., Sanderson, J.S., Cooper, D.J., 2007. Annual
groundwater evapotranspiration mapped from single satellite scenes. Journal of
Hydrology, 344, 146-156.
Houston, J., 2006. Evaporation in the Atacama Desert: an empirical study of spatiotemporal
variations and their causes. Journal of Hydrology, 330, 402-412.
McGuinness, J.L., Bordne, E.F., 1972. A Comparison of Lysimeter-derived Potential
Evapotranspiration with Computed Values (No. 1452). US Dept. of Agriculture.
Muñoz, J.F., Suárez, F., Fernández, B., Maass, T., 2017. Hydrology of the Silala River
Basin. (Vol. 5, Annex VII).
Senay, G.B., Leake, S., Nagler, P.L., Artan, G., Dickinson, J., Cordova, J.T., Glenn,
E.P. (2011). Estimating Basin scale evapotranspiration (ET) by water balance and
remote sensing methods. Hydrological Processes, 25(26), 4037-4049.
Summer, D.M. and Jacobs, J.M., 2005. Utility of Penman–Monteith, Priestley–Taylor,
reference evapotranspiration, and pan evaporation methods to estimate pasture
evapotranspiration. Journal of Hydrology, 308(1), 81-104.
The State Climatologist, 1985. Publication of the American Association of State
Climatologists: Heights and Exposure Standards for Sensors on Automated Weather
Stations, 9.
WMO, 2008. Guide to Meteorological Instruments and Methods of Observation. WMO
No. 8, 7th ed. WMO, Geneva.
Yoder, R.E., Odhiambo, L.O., Wright, W.C., 2005. Evaluation of methods for
estimating daily references crop evapotranspiration at a site in the humid southeast
United States. Applied Engineering in Agriculture, 21, 197-202.
Annex IX
471
32 APPENDIX A CALCULATION OF ATMOSPHERIC PARAMETERS Mean saturation vapor pressure The mean saturation vapor pressure (es, [kPa]) is calculated as follows (Allen et al., 1998): 𝑒𝑒𝑠𝑠 =𝑒𝑒 °(𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚)+𝑒𝑒 °(𝑇𝑇𝑚𝑚𝑚𝑚𝑚𝑚)2(A.1) where T is the air temperature [°C] and e°(T) [kPa] is the saturation vapor pressure at a temperature T [°C] (Allen et al., 1998): e°(T)=0.6108𝑒𝑒𝑒𝑒𝑒𝑒[17.27 TT+237.3 ] (A.2) Slope of the saturation vapor pressure curve The slope of the saturation vapor pressure-temperature curve (, [kPa °C-1]), at a temperature T [°C], can be calculated using the following formula (Allen et al., 1998): Δ=4098𝑒𝑒𝑒𝑒𝑒𝑒(17.27 TT+237.3) (T+237.3)2 (A.3) Actual vapor pressure derived from relative humidity data The actual vapor pressure (ea, [kPa]) derived from relative humidity data was calculated using the following equation (Allen et al., 1998): ea = e°(Tmin)RHmax 100+e °(Tmax)RHmin1002(A.4) where the subscripts min and max correspond to the minimum and maximum values observed during the day. 472
Annex IX Appendix A
33
Atmospheric pressure
The atmospheric pressure (P, [kPa]) is a function of the elevation of the site (z, [m])
(Allen et al., 1998):
P = 101.3 ( 293−0.0065 z
293 )
5.26
(A.5)
Psychrometric constant
The psychrometric constant (γ, [kPa °C-1]) was estimated as follows (Allen et al., 1998):
γ = cp P
ϵ λ (A.6)
where cp is the specific heat at constant pressure (1.103×10-3 MJ kg-1 ℃-1) and ϵ is the
ratio of the molecular weight of water vapor and dry air (0.622).
Net radiation
To estimate the net radiation, the extraterrestrial radiation for daily periods (Ra, [MJ m-2
day-1]) is required (Allen et al., 1998):
Ra = 24 (60)
π Gscdr[ωs 𝑠𝑠𝑠𝑠𝑠𝑠() 𝑠𝑠𝑠𝑠𝑠𝑠() + 𝑐𝑐𝑐𝑐𝑐 () 𝑐𝑐𝑐𝑐𝑐 () 𝑠𝑠𝑠𝑠𝑠𝑠(ωs)] (A.7)
where Gsc is the solar constant (0.082 MJ m-2 min-1); dr [-] is the inverse relative
distance earth-sun (equation 4.8);  [rad] is the solar declination (equation 4.9); s [rad]
is the sunset hour angle (equation 4.10); and  [rad] is the latitude.
dr = 1 + 0.033 𝑐𝑐𝑐𝑐𝑐 ( 2π
365 J ) (A.8)
 = 0.409 𝑠𝑠𝑠𝑠𝑠𝑠 ( 2π
365 J − 1.39) (A.9)
ωs = π
2 − 𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎𝑎𝑎 [− 𝑡𝑡𝑡𝑡𝑡𝑡() 𝑡𝑡𝑡𝑡𝑡𝑡()
X0.5 ] (A.10)
𝑋𝑋 = {1 − [tan()]2[tan()]2 𝑖 𝑖𝑖 𝑋𝑋 > 0
0.00001 𝑖 𝑖𝑖 𝑋𝑋 ≤ 0 (A.11)
where J is the number of the day in the year (January 1st: J = 1; December 31st: J = 365).
To derive solar radiation (Rs, [MJ m-2 day-1]) from air temperature data, the Hargreaves’
radiation formula can be used (Allen et al., 1998):
Rs = kRs √Tmax − 𝑇𝑇min Ra (A.12)
Annex IX Appendix A
473
34 where kRs is an adjustment coefficient (0.16 ℃-0.5 ). For clear-sky solar radiation (Rso, [MJ m-2 day-1]), the following formula has been used (Allen et al., 1998): Rso=(0.75+2×10−5 z ) Ra (A.13) The net shortwave radiation (Rns, [MJ m-2 day-1]) is estimated using the albedo ( = 0.23) (Allen et al., 1998): Rns=(1−α)Rs (A.14) The net longwave radiation (Rnl, [MJ m-2 day-1]) is estimated using the following equation (Allen et al., 1998): Rnl=σ[Tmax,K4+Tmin,K42](0.34−0.14√ea)(1.35RsRso−0.35) (A.15) where σ is the Stefan-Boltzmann constant (4.903×10-9 MJ K-4 m-2 day-1), and Tmax,K and Tmin,K are the maximum and minimum absolute temperatures during the 24-hour period [K]. Finally, the net radiation is given by the sum of the net shortwave and longwave radiation. 474
Annex IX Appendix A
Annex X
Suárez, F., Sandoval, V., Sarabia, A., 2017. River-Aquifer
Interactions Using Heat as a Tracer in the Transboundary Basin of the Silala River
475
476
Annex X
RIVER-AQUIFER INTERACTIONS USING HEAT AS A TRACER IN THE
TRANSBOUNDARY BASIN OF THE SILALA RIVER
Francisco Suárez (PhD)
Associate Professor, Pontificia Universidad Católica de Chile
Research Assistants: Victoria Sandoval (MSc), Andrés Sarabia (BSc)
May, 2017
Annex X
477
GLOSSARY This glossary of hydrological terms is based on the following: http://www.wmo.int/pages/prog/hwrp/publications/international_glossary/…http://www.nws.noaa.gov/om/hod/SHManual/SHMan014_glossary.htmhttp://www.geo.utexas.edu/faculty/jmsharp/sharp-glossary.pdfAquifer: Geological formation capable of storing, transmitting and yielding exploitable quantities of water. Artesian Well: Well tapping a confined aquifer whose piezometric surface lies above the ground surface. Basin (Or River Basin): Area having a common outlet for its surface runoff. Conduction-Advection Equation: (in the context of heat transfer) Principle used to quantify the transport of heat. Discharge: Volume of water flowing per unit time, for example through a river cross-section or from a spring or a well. Environmental Tracers: Tracers that are found in the natural environment. Evapotranspiration: Combination of evaporation from free water surfaces and transpiration of water from plant surfaces to the atmosphere. Flowrate: Volume flow per unit time. Groundwater: Subsurface water occupying the saturated zone (i.e. where the pore spaces (or open fractures) of a porous medium are full of water). Heat Capacity: Heat absorbed or released by a system per unit of temperature rise or fall. Hydraulic Gradient: The decrease in hydraulic head per unit length in the direction of flow, for example in a closed conduit, open channel or porous medium. Hydraulic Head: Sum of the elevation head, pressure head and velocity head, where head refers to a measure of pressure as the equivalent height of a column of water. Infiltration: The movement of water from the surface of the land into the subsurface. Losing Reach: Segment of a river where water infiltrates through the riverbed sediments. 478
Annex X
Perennial River: River that flows continuously all through the year.
Piezometer: Device that measures the pressure (piezometric head) of groundwater at a
specific point, typically by monitoring water level in a well that is entirely cased except
at its lowest end.
Porosity: Ratio of the volume of the pore space in a given sample of a porous medium
to the gross volume of sample.
Pressure Head: Head of a column of static water that can be supported by the static
pressure at a point.
Pumping Test: A field testing procedure to quantify aquifer properties at a site
involving pumping water out of (or less commonly injecting water into) an aquifer and
measuring the effect on water levels in that aquifer and sometimes in adjacent strata.
Recharge: Contribution of water to an aquifer either by direct infiltration or by runoff
and subsequent infiltration.
Reservoirs: Body of water, either natural or man-made, used for storage, regulation and
control of water resources.
River Reach: Length of a channel between two defined cross sections.
River Stage: Water level of the river at a specific location.
River-Aquifer Interactions: Term used to describe the water exchange between
surface waters and groundwater.
Saturation: (in the context of soils or rocks) The state in which the pore spaces or open
fractures of a porous medium are full of water – the definition of ‘groundwater’.
Sediments: Material transported by water either in suspension or as bed load from the
place of origin to the place of deposition.
Spring: Place where groundwater emerges naturally from the rock or soil.
Streamflow: General term for water flowing in a watercourse.
Surface Water: Water which flows over or is stored on the ground surface.
Temperature Rod: Instrument used to measure the thermal profile in the sediments.
Thermal Conductivity: Property of a solid or fluid which relates conductive heat flux
to the temperature gradient according to Fourier’s law.
Thermal Dispersivity: Property that quantifies the spreading of heat as a result of the
heterogeneity in the velocity field.
Annex X
479
Vadose Zone: (also known as unsaturated zone) Subsurface zone above the water table in which the interstices are filled with air and water, and the water pressure is less than atmospheric pressure. Water Flux: Volume of water flowing per unit time per unit area, e.g. through a river cross-section or an aquifer. Weir: Overflow structure that may be used for controlling upstream water level, and/or for measuring discharge. 480
Annex X
TABLE OF CONTENTS
1 INTRODUCTION ..................................................................................................... 1
1.1 Study zone ................................................................................................... 1
1.2 Objectives .................................................................................................... 1
1.3 Summary of the methodology ..................................................................... 4
1.4 Structure of the report .................................................................................. 5
2 SUMMARY AND CONCLUSIONS ........................................................................ 5
3 RIVER-AQUIFER INTERACTIONS USING TEMPERATURE TIME
SERIES IN THE SEDIMENTS BENEATH THE RIVER ....................................... 8
3.1 Introduction .................................................................................................. 8
3.2 Materials and methods ................................................................................. 9
3.2.1 Theory and analytical approach to determine the water exchange
between groundwater and surface waters .................................................... 9
3.2.2 Experimental methods ............................................................................... 11
3.2.3 Temperature time series analysis ............................................................... 18
3.3 Results and discussion ............................................................................... 19
3.3.1 Characterization of the thermal and hydraulic properties of the
streambed sediments .................................................................................. 19
3.3.2 Temperature time series analysis ............................................................... 21
3.4 Conclusions ................................................................................................ 45
4 RIVER-AQUIFER INTERACTIONS USING DISTRIBUTED
TEMPERATURE SENSING ALONG THE RIVER .............................................. 46
4.1 Introduction ................................................................................................ 46
4.2 Materials and methods ............................................................................... 47
4.2.1 Distributed temperature sensing (DTS) theory .......................................... 47
4.2.2 Field deployment ....................................................................................... 48
4.2.3 Calibration and quality analysis of measurements .................................... 50
4.2.4 Location of the cable along the Silala River .............................................. 55
Annex X
481
4.2.5 Data filtering to remove thermal anomalies not related to water temperature ................................................................................................ 56 4.3 Results and discussion ............................................................................... 61 4.4 Conclusions ................................................................................................ 70 5 ANALYSIS OF WATER FLOWS AT THE DIFFERENT WEIRS ALONG THE SILALA RIVER.............................................................................................. 71 5.1 Introduction ................................................................................................ 71 5.2 Materials and methods ............................................................................... 72 5.2.1 Weirs and stage-discharge curves .............................................................. 72 5.2.2 Water level monitoring at the weirs ........................................................... 74 5.2.3 Anthropogenic water use in the study site ................................................. 75 5.3 Results ........................................................................................................ 75 5.4 Conclusions ................................................................................................ 77 6 CONCLUSIONS...................................................................................................... 78 7 REFERENCES ........................................................................................................ 80 APPENDIX A 482
Annex X
1
1 INTRODUCTION
The National Director of the Dirección Nacional de Fronteras y Límites del Estado
(DIFROL) of the Ministry of Foreign Affairs of Chile, Mrs. Ximena Fuentes, requested
that Professor Francisco Suárez undertake an interdisciplinary study aimed at deepening
the hydrogeological knowledge of the transboundary basin of the Silala River, located
in the north of Chile.
This report presents the methodology and results obtained from a study that used heat as
an environmental tracer to investigate surface water-groundwater interactions in the
Silala River, near the international border between Chile and Bolivia. This study was
led by Dr. Francisco Suárez and the report was written under the supervision and
instruction of Drs. Howard Wheater and Denis Peach.
1.1 Study zone
The area of the study is the Silala River basin, a transboundary watershed shared by
Bolivia (upstream) and Chile (downstream). The Silala River basin is located in the
Andean Plateau of the Atacama Desert, approximately 300 km northeast of Antofagasta
(Figure 1-1 and Figure 1-2). The river reach studied in this report is approximately 2 km
long, beginning near the international border and finishing after the confluence of the
Silala River and Quebrada Negra, as shown in Figure 1-3.
1.2 Objectives
The general objective of this study is to investigate the interactions between
groundwater and surface waters in the Silala River using thermal methods.
The specific objectives are to:
1. Use temperature time series obtained at different depths in the sediments beneath the
Silala River to estimate the water exchange between the river and the aquifer.
2. Use distributed temperature sensing (DTS) methods to identify potential locations of
groundwater contribution along the Silala River.
3. Construct a conceptual model of the river-aquifer interactions based on the
temperature time series, DTS data, and streamflow and meteorological data.
Annex X
483
2 Figure 1-1. Location of the Silala River basin at approximately 300 km northeast of Antofagasta. 484
Annex X
3
Figure 1-2. Location of the Silala River basin at the Chile-Bolivia international boundary.
Annex X
485
4 Figure 1-3. River reach where the interactions between groundwater and surface waters were investigated. 1.3 Summary of the methodology To estimate the water exchange between the river and the aquifer, the temperature profile in the sediments below the river and the sediments’ thermal properties were measured at five locations. Then, the water exchange between the river and the aquifer was quantified by solving the one-dimensional conduction-advection equation using the Hatch et al. (2006) amplitude method. To identify potential locations of groundwater entering into the river, a fiber-optic cable was installed along a ~1.3 km reach of the Silala River. The cable was connected to a DTS instrument and average temperatures were measured, with a spatial resolution of 486
Annex X
5
0.5 m (sampling resolution of 0.25 m) and a temporal resolution of 5 min. The
measurements were carefully calibrated to obtain a thermal resolution of ~0.1°C. Using
the thermal profile obtained along the river reach, and using average data collected
during the night, different locations of groundwater discharge into the river were
identified. The identification was based on locations that exhibit a distinct increase in
the river temperature of 0.1°C or more.
Finally, a conceptual model of the river functioning was made based on the information
collected from the two heat tracer experiments described above. This conceptual model
was also supported using river flow data collected at the weirs installed along the Silala
River.
1.4 Structure of the report
The structure of the remainder of this report is as follows: Chapter 2 states a summary
and the conclusions of the study; Chapter 3 reports the investigation that used
temperature time series beneath the river to estimate the water exchange between the
river and the aquifer; Chapter 4 presents the study of river-aquifer interactions using
DTS methods and a conceptual model of the river-aquifer interactions that was
constructed based on the data collected in the field; Chapter 5 presents an estimation of
the river-aquifer interactions based on the information collected at the weirs that are
located along the Silala River; Chapter 6 presents the main conclusions of this study.
Finally, the references cited in the report are shown in Chapter 7. An Appendix
containing detailed information and photographs is included hereto.
2 SUMMARY AND CONCLUSIONS
In this study, the interactions between the river and the aquifer of the Silala River basin
were evaluated using heat as a tracer. This study included two investigations. The first
investigation consisted in analyzing the thermal profile of the sediments beneath the
river to determine the water exchange between the river and the aquifer. The second
investigation identified locations of groundwater entering the river through the use of
DTS methods. Using this information, a conceptual model of the river-aquifer
interactions in the Silala River basin was made.
In the first investigation, temperature time series combined with the sediments’ thermal
properties and hydraulic head data were used to quantify the fluxes that occur between
the river and the aquifer. Temperature rods (TR’s) were constructed to monitor the
temperatures of the sediments at different depths, and five TR’s were installed at
Annex X
487
6 different locations along the Silala River (named TR1, TR2, TR3, TR4 and TR5). Our results, based on the thermal analysis combined with Monte Carlo simulations, show that at four of the locations investigated (TR1, TR2, TR4, and TR5), it is very likely that water flows from the river into the aquifer through the streambed. At TR3 the conclusions are less certain; the thermal analysis yielded fluxes going from the aquifer towards the river, but the direction of the hydraulic gradient at that location implies that water flows in the other direction (from the river to the aquifer). The overall conclusion from these point measurements is that the river reach investigated in this study appears to be a losing reach. Nonetheless, at two locations the hydraulic head is small, implying that changes in the river stage or in the aquifer groundwater level could reverse the direction of the water fluxes. At TR1, we estimated water fluxes of 22  6 cm day-1 (with a probability of 98% that water is flowing towards the aquifer). At TR2, we estimated water fluxes of 5  7 cm day-1 (with a probability of 78% that water is flowing towards the aquifer). At TR3, we estimated water fluxes of -7  21 cm day-1 (with a probability of 56% that water is flowing towards the river). At TR4, we estimated water fluxes of 44  7 cm day-1 (with a probability of 100% that water is flowing towards the aquifer). And at TR5, we estimated water fluxes of 3  16 cm day-1 (with a probability of 62% that water is flowing towards the aquifer). It is important to recall that the water exchanges between the river and the aquifer quantified in this study are point-in-space measurements. Therefore, they do not mean that the same behavior occurs along the entire river reach. In the second investigation, DTS measurements along a ~1.3 km reach of the Silala River allowed high-resolution monitoring of water temperatures, both in space and time. The river reach began at ~750 m from the Chile-Bolivia international border and finished ~2 km downstream of the border. Along this reach, temperature was measured with a spatial resolution of 0.5 m (sampling resolution of 0.25 m) and using an integration time of 5 min. When the data were integrated to 5 min, the thermal resolution of the DTS system was 0.12 °C. The collected temperature data were separated into mean profiles taken during the day and the night. The mean day traces corresponded to thermal data averaged during the 4 h of maximum irradiance of each day, while the mean night traces were data averaged during the 4 h before reaching the minimum temperature along the river. The mean day and night traces were averaged over a time period of ~104 days and resulted in a thermal resolution of 0.14 and 0.11 °C, respectively. Using the night traces collected along the river reach, different locations of groundwater discharge into the river were identified. The identification was based on locations that exhibit a distinct increase in the river temperature of 0.1 °C or more that can only be explained by a warm groundwater input. 488
Annex X
7
Three types of river-aquifer interactions were defined for this study: the Type-1 riveraquifer
interactions correspond to localized groundwater inputs, which increase the
river temperature by more than 0.1 °C; the Type-2 river-aquifer interactions correspond
to locations on the fiber-optic cable that exhibit a temperature peak during the night and
a temperature valley during the day; Finally, the Type-3 river-aquifer interactions
correspond to groundwater input coming from the SPW-DQN artesian well. The waters
from the SPW-DQN well originate in a confined aquifer that is deeper than the fluvial
aquifer that is located below the Silala River (Arcadis, 2017). Here, the term confined
aquifer is based on its hydrogeological definition (one containing groundwater that is
under pressure exceeding atmospheric pressure, and where the recharge area to a
confined aquifer is at some distance and is unconfined but at higher elevation that the
confined aquifer). The Type-3 groundwater contribution is similar to those described as
Type-1 river-aquifer interactions but is described as another type of interaction since its
source is different and its effect on the thermal dynamics of the river is important.
Using DTS methods, several locations of groundwater discharge were identified but the
most distinct groundwater input was a warm groundwater discharge that corresponds to
SPW-DQN (Type-3 river-aquifer interaction), which heavily influences the thermal
behavior of the river downstream, producing an increase of ~5 °C in the mean water
temperature (during the entire day) and up to ~9 °C at specific times (during the night).
The Type-1 river-aquifer interaction was the second groundwater discharge in
importance at the study site. The thermal anomalies caused by the Type-1 inputs are
clearly present during the whole measuring period and were consistent in their location,
increasing the temperature of the river downstream. These locations are highly
correlated with the positions of springs that discharge from the walls of the ravine at
elevations higher than the river stage. Finally, Type-2 river-aquifer interactions most
likely correspond to locations where the DTS cable is buried. Our results suggest that, at
the locations where the fiber-optic cable was buried, the water discharges from the river
to the aquifer, consistent with the findings using the TR’s.
From the results obtained using the TR and DTS methods, a conceptual model of the
river-aquifer interactions was built. This conceptual model was supplemented by data
from weirs that are located along the study site. Our results show that the Silala River
reach investigated in this study is a gaining reach (i.e., river-groundwater interactions
are dominated by groundwater inputs), where ~35.9 l/s of groundwater enters the river
from different springs that discharge water that is warmer than the river from the walls
of the ravine at elevations higher than the river stage. In addition, approximately 3.3 l/s
of water flows from the river into the fluvial aquifer below the river through the
riverbed sediments. Near the end of the reach, groundwater is discharged into the river
Annex X
489
8 from the SPW-DQN artesian well, which discharges ~91.6 l/s of warm water. Therefore, the water balance shows a net gain of the order of ~124 l/s in the investigated reach. The data collected in this study reveal that river-aquifer interactions in the Silala River are undeniable and support the hypothesis that the Silala River is a perennial river supported by groundwater. 3 RIVER-AQUIFER INTERACTIONS USING TEMPERATURE TIME SERIES IN THE SEDIMENTS BENEATH THE RIVER 3.1 Introduction Heat has been successfully used as an environmental tracer to describe flow through saturated porous media, such as in the sediments that surround rivers (Anderson, 2005). Heat transport in the streambed sediments is influenced by the thermal fluctuations in the river, by the physical properties of the sediments, and by the water exchange between the groundwater and the river (Constantz, 2008). In the last ten years, new methods have been developed to determine the water exchange between rivers and aquifers using heat. Temperature time series have also been investigated during recent years, and have resulted in improved determination of the spatiotemporal distribution of the groundwater-surface water exchange (Hatch et al., 2006; Gordon et al., 2012; Briggs et al., 2012, 2014, 2016; Vandersteen et al., 2015; Irvine et al., 2016). These thermal data have been combined with both analytical (Hatch et al., 2006; Keery et al., 2007) and numerical (Healey and Ronan, 1996; Anibas et al., 2009; Gordon et al., 2012; Vandersteen et al., 2015) models with different levels of complexity that allow determination of the water exchange between groundwater and surface waters. The main objective of this study is to investigate the water exchange between the Silala River and the fluvial aquifer beneath it. The specific objectives are: (1) to determine the thermal properties (thermal conductivity and heat capacity) and hydraulic conductivity of the streambed sediments of the Silala River; (2) to obtain temperature time series at different locations along the river and at different depths in the streambed sediments; (3) to estimate water exchange fluxes between the river and the aquifer; and (4) to perform an uncertainty analysis on the estimated fluxes. The structure of this chapter is as follows: first, the theory to determine the water exchange between the river and aquifer is shown. Then, the study zone is presented, with emphasis on the locations where the temperature time series were collected. Next, the experimental methods used in this study are described, and the information needed to carry out the analysis is explained. Then, results and discussions of the investigation are presented. Finally, the main conclusions of this work are stated. 490
Annex X
9
3.2 Materials and methods
3.2.1 Theory and analytical approach to determine the water exchange between
groundwater and surface waters
The temperatures in the upper part of a streambed can be characterized using a semiinfinite
half-space domain. When the thermal conditions in this domain are assumed to
be governed by a one-dimensional conduction-advection equation, the following heat
transport equation can be used to describe the thermal dynamics (Stallman, 1965; Hatch
et al., 2006):
z
C q T
z
T
t z
C T p w 

 










 (3.1)
where T [°C] is temperature; z [m] is the vertical direction; Cp and Cw [J m-3 K-1, Joules
per cubic meter-Kelvin] are the volumetric heat capacity of the saturated soil and water,
respectively; q [m s-1] is the water flux in the vertical direction; and λ [W m-1 K-1, Watts
per meter-Kelvin] is the effective thermal conductivity of the saturated soil, defined as
(Jury and Horton, 2004):
t w f     C v 0 (3.2)
where βt [m] is the thermal dispersivity; λ0 [W m-1 °C-1] is the thermal conductivity in
absence of fluid flow; and vf [m s-1] is the linear velocity of a fluid particle. Note that q
is a volumetric flux (volume per unit area per unit time) while vf is a velocity (distance
per time); and that both are related to the velocity of the thermal front, v [m s-1], in the
following way (Hatch et al., 2006):
e
f n
v  q (3.3)
v
C
C
q
w
 p (3.4)
where ne [m3 m-3] is the porosity. Equation (1) implies that if the thermal spatiotemporal
evolution in the sediments near the river is known, the water flux (q) can be
inferred.
Different approaches have been used to determine the water fluxes from equations
(3.1)-(3.4) when temperature time series at different depths in the sediments below a
river are available (Lautz, 2010; Shanafield et al., 2011; Vandersteen et al., 2015). In
Annex X
491
10 particular, analytical methods have been successfully used to investigate groundwater-surface water interactions (Hatch et al., 2006; Keery et al., 2007; McCallum et al., 2012; Luce et al., 2013). Therefore, in this study we performed a time series analysis that is implemented in an open-source computational software called VFLUX (Gordon et al., 2012). This software is populated with field data collected from temperature rods (TR’s – see Figure 3-1) and combined with hydraulic gradient information collected from piezometers (the experimental methods are described below). Figure 3-1. Diagram illustrating the upper (shallowest) part of a streambed and the thermal observations that are used to determine the water exchange between groundwater and surface water using thermal methods: (a) a representation of a temperature rod (TR) used to collect temperature time series at different depths below the water-sediment interface; and (b) a representation of the thermal signals collected at different depths in the TR. The temperature time series shows a reduction in amplitude (A) and a shift in phase () with greater depth. A and  are used to determine the water exchange between groundwater and surface waters (modified from Hatch et al., 2006). The VFLUX (Vertical Fluid [Heat] Transport Solver software, available at the following link: http://hydrology.syr.edu/vflux/), processes the temperature time series, and estimates the water fluxes between each pair of sensors within a TR using signal analysis (Gordon et al., 2012). As shown in Figure 3-1, as the depth increases, the amplitude is reduced (A) and a phase change () occurs. The variations in A and  492
Annex X
11
are associated with the water travel time through the sediments. In VFLUX, a dynamic
harmonic regression (DHR) method is used to extract A and  from the time series,
and the water exchange between the river and the groundwater is determined using a
total of six different analytical approaches: the amplitude and the phase change methods
proposed by Hatch et al. (2006); the amplitude and phase change methods of Keery et
al. (2007) and the combined methods of McCallum et al. (2012) and Luce et al. (2013).
In the current study, we selected the Hatch et al. (2006) amplitude method to determine
the water exchange between groundwater and surface water as this method has shown
good agreement for both downwelling (towardss the aquifer) and upwelling (towards
the river) conditions (e.g., see Briggs et al., 2014). Additionally, the amplitude method
of Hatch et al. (2006) was the approach that yielded results that agreed better with our
measurements of hydraulic gradient.
The Hatch et al. (2006) amplitude method estimates the water fluxes according to the
following relationships:
 


 

 



2
2 ln v2 A
C zC
C
q r
w p
p  
(3.5)
2
4 8
 


 


 
p PC
v 
 (3.6)
where Ar [-] is the ratio of amplitude variations between pairs of temperature
measurement points at different depths, and separated at a distance of z [m]; and P [s-1]
is the period of temperature fluctuations. The Hatch et al. (2006) amplitude method also
requires knowledge of the following media properties: thermal conductivity, heat
capacity, thermal dispersivity, and porosity.
3.2.2 Experimental methods
3.2.2.1 Study site, construction and installation of temperature rods and
piezometers
To quantify the water exchange between surface waters and groundwater, five
temperature rods (TR’s), each with a corresponding pair of piezometers, were installed
along the Silala River (Figure 3-2). The TR’s were installed along a ~2-km reach of the
river, beginning near the Chile-Bolivia international border and finishing downstream of
the confluence of the Silala River and Quebrada Negra.
Annex X
493
12 The TR’s were built using steel rods (hollow tubes 1.5 m length and 0.0254 m (1 inch) diameter), with temperature loggers (DS1922L iButtons, Maxim Integrated, San Jose, CA; with an accuracy of 0.5 °C and a resolution of 0.0625 °C) that can be installed at different depths below the river-sediment interface. Figure 3-2. Study site showing the location of each temperature rod (TR1-TR5) along the study site in the Silala River. The figure also shows photographs of each TR. The construction of the TR’s was based on the design proposed by Naranjo and Turcotte (2015). As shown in the schematics and photograph depicted in Figure 3-3, each TR 494
Annex X
13
consists of a steel pipe that contains 6-7 temperature sensors that are integrated into an
electrical circuit board that allow their connection. The temperature sensors are
connected to the circuit board using clips that allows manual installation of the sensors
at different depths. The top of the TR has an access to facilitate the data collection
process. To collect the data, a phone cable with an RJ-11 connection is coupled to the
electronics at one end, while at the other end a DS9490R USB adapter is used to
connect the TR to the computer. The WeeButtonRF software (Alphamach, Ste-Julie,
QC, Canada) is used to extract the data from the temperature sensors. The advantages of
this system are similar to those described by Naranjo and Turcotte (2015): (1) the
thermal sensors can be replaced without the need to weld them into the board; and (2)
the electrical circuit board has the flexibility of allowing the addition of more sensors at
different depths. Initially, circuit boards of 1.23 m length with 10 clips spaced between
0.05 and 0.2 m apart were constructed, with the small spacings located at the shallowest
depths to allow monitoring of upwelling conditions (Briggs et al., 2014). However, after
a couple of days of measurements we observed that in some river reaches the thermal
signal was attenuated in the first 0.5 m (or less) below the water-sediment interface. As
a consequence, new circuit boards were constructed with a spacing that varied between
0.03 and 0.15 m.
As described previously, five transects were identified along the Silala River. In each
transect, one TR and two piezometers were installed (Figure 3-2). The TR’s were driven
manually into the sediments and the resulting depths of these systems varied between
0.6 and 1.0 m (Table 3-1). Prior to the insertion of the TR’s into the sediments, a
0.0191-m diameter metal guide was manually introduced into the ground. Inserting the
metal guide allowed locations to be found where the TR’s could be inserted into the
sediments without damaging the rods or the piezometers. Then, the metal guide was
carefully removed from the sediments and the TR was inserted into the hole made by
the metal guide. Note that the metal guide had a smaller diameter than the TR’s and the
piezometers to avoid preferential flow along the TR’s.
TR5 was installed on September 29th, 2016, while TR1 through TR4 were installed on
October 18-19th, 2016. The data collected prior to November 4th, 2016 allowed
optimization of the depth of the temperature sensors in each TR, i.e., in each TR the
sensor depths were selected so that the thermal signal below the water-sediment
interface was captured, trying to avoid having more than one sensor registering a
constant temperature at a specific depth. Initially, the thermal data were collected using
time intervals of 30 min, but subsequently the data collection frequency was reduced to
1 h (November 4th, 2016). This reduction in the data collection frequency was carried
out because further analysis showed that time intervals of 1 h did not affect the
Annex X
495
14 determination of the water exchange fluxes. This change in the settings also increased the time before the data-logger capacity was filled up (approximately 3 months). Figure 3-3. (a) Example of the configuration of a temperature rod (TR); (b) the corresponding pair of piezometers near the TR; and (c) a photograph that shows when the electronics were inserted into the steel rod. The depths of each temperature sensor (P1, P2, P3, P4, P5, P6 and P7) and the angle between the TR and the vertical direction () are depicted in subfigure (a). The distances (D1, D2, D3) that define the location of each pressure transducer (used to determine the hydraulic gradient, H/z) and the angle between the vertical direction and the piezometer () are shown in subfigure (b). In this example, subfigure (b) indicates that the flow goes from the river to the aquifer. 496
Annex X
15
The water level in the aquifer beneath the river-sediment interface was measured in
piezometers made of steel at time intervals of 15 min. The piezometers had a diameter
of 0.0381 m (1.5 inches) and a total length of 1.5 m. The openings in the piezometer
were located at 0.1 m above the tip of the rod. A pressure transducer (HOBO Water
Level (4 m) Data Logger, Onset Computer Corporation, Bourne, MA) was installed
inside each piezometer. This instrument had a typical error of 4 mm, a maximum error
of 8 mm, and a resolution of 1.4 mm. The water level in the river was measured using
similar pressure transducers inserted inside a PVC pipe (0.040 m diameter and 0.5 m
length) that had holes all along its length. The PVC pipe was tied to the steel
piezometer. This instrumentation allowed determination of the hydraulic gradient
(H/z) between the river and the aquifer at each TR, which indicates the flow
direction. The distances between pressure transducers and holding clamps in the
piezometers were measured manually using a measuring tape. Therefore, the difference
in hydraulic head (H) may be subject to an error that is estimated to be ~0.01 m or
less. To reduce this error, a wooden stick was inserted into the piezometers where the
aquifer water level is being recorded and the distance between the top of the piezometer
and the water level was measured using the measuring tape. Then, the distance between
the top of the piezometer and the river water level was measured along the side of the
piezometer, allowing estimation of the head difference. This head difference was used
to correct the estimates of the pressure transducers, i.e., both estimates of the head
difference must agree at the time of the manual measurement. Note also that since not
all the TR’s or piezometers were installed vertically, the angles between the vertical
direction and the rod and piezometer direction were measured to determine the depth of
each temperature sensor (Figure 3-3). Table 3-1 presents the most relevant geometrical
features of each TR and their pair of piezometers, which shows that the depths reached
by the TR’s varied between 0.6 and 1.0 m.
Annex X
497
16 Temperature rod (TR) TR1 TR2 TR3 TR4 TR5 Latitude 22°0’54.87’’S 22°1’13.61’’S 22°1’20.20’’S 22°1’31.18’’S 22°1’38.05’’S Longitude 68°1’49.53’’O 68°1’59.29’’O 68°2’4.97’’O 68°2’14.26’’O 68°2’23.99’’O Northing (m) * -22.015242 -22.020448 -22.022278 -22.025329 -22.027237 Easting (m) * -68.030424 -68.033136 -68.034715 -68.037294 -68.039998 Elevation (m.a.s.l.) 4278 4249 4241 4181 4194 Total depth of TR (m) 1.00 1.00 0.98 0.99 1.00 Angle of the TR,  (°) 6 0 6 27 0 Number of temperature sensors 7 6 5 6 4 P1 (m) 0.005 0.030 0.040 -0.040*** -0.010*** P2 (m) 0.035 0.080 0.070 0.086 0.090 P3 (m) 0.065 0.130 0.100 0.175 0.190 P4 (m) 0.095 0.180 0.250 0.354 0.490 P5 (m) 0.125 0.280 0.380 0.533 -- P6 (m) 0.274 0.380 -- 0.712 -- P7 (m) 0.423 -- -- -- -- Total depth of piezometer (m)** 1.00 0.75 0.60 0.99 1.03 Angle of the piezometer,  (°) 10 0 0 0 0 D1 (m) 0.325 0.265 0.285 0.310 0.160 D2 (m) 0.115 0.185 0.290 0.100 0.045 D3 (m) 0.830 0.870 0.880 0.875 0.205 Table 3-1. Geometrical information related to the temperature rods (TR’s) and their corresponding pair of piezometers. The distances D1, D2 and D3 were measured at the time of installation. * WGS84 projection. ** Total depth of piezometer corresponds to the distance from the water-sediment interface to the openings of the piezometer. *** Negative distance means that the sensor is above the water-sediment interface. 498
Annex X
17
3.2.2.2 Characterization of the thermal and hydraulic properties of the streambed
sediments
To characterize the streambed sediments, the thermal properties (thermal conductivity
and heat capacity) and the saturated hydraulic conductivity of the sediments were
measured. The thermal properties were measured in the field and in the laboratory,
while the saturated hydraulic conductivity was measured in the laboratory.
To determine the sediments’ thermal properties, the dual probe method with transient
heating was used (Bristow et al., 1994). In this method, a probe with two needles is
inserted into the sediments. Then, a heat pulse is applied to one of the needles while the
other one measures the thermal response. This information allows the thermal
conductivity () and the heat capacity (Cp) to be determined by solving the radial heat
flow equation (Bristow et al., 1994). This method was implemented both in the
laboratory and in the field using a thermal properties analyzer (KD2 Pro/SH-1, Decagon
Devices, Pullman, WA). For the laboratory measurements, sediment samples were
collected from the field and carried to the laboratory. From each one of the TR transects
one sediment sample was collected, for a total of five samples (see Figure 3-4). The
samples were collected in 250-cm3 rings (8-cm diametre, 5-cm length). The samples
were slowly saturated from below and then the thermal properties analyzer was used to
determine the  and the Cp. Three measurements were carried out in different locations
of each sample (Figure 3-4a). For the field measurements, the thermal properties
analyzer was brought to the field and it was used near each of the TR’s, where at least
three measurements were made (Figure 3-4b).
Figure 3-4. Thermal property measurements of the streambed sediments: (a) in the laboratory;
(b) in the field.
Annex X
499
18 The saturated hydraulic conductivity was measured with a variable-head permeameter (KSAT, UMS, Munich, Germany). This permeameter uses Darcy’s Law (Darcy, 1856) to determine the saturated hydraulic conductivity as a function of the water flow that passes through the sediment sample. The variable-head permeameter uses the same 250-cm3 rings that were used to collect the sediment samples. Thus, the samples described before were used to determine the saturated hydraulic conductivity. Note that this test was carried out before the analysis of the thermal properties. Also, since the sediment samples that were brought into the laboratory were slightly disturbed, the saturated hydraulic conductivity values were not used in any calculations and are only used as a reference. 3.2.3 Temperature time series analysis The determination of the surface water/groundwater exchange in each TR installed along the Silala River was performed using the temperature time series, and the difference in the hydraulic head (H) between the river and the aquifer. The time series analysis was carried out using the amplitude method proposed by Hatch et al. (2006), which is implemented in the VFLUX computational software (Gordon et al., 2012). The Hatch et al. (2006) amplitude method also requires knowledge of the following media properties: thermal conductivity, heat capacity, thermal dispersivity, and porosity. The thermal conductivity and heat capacity were measured both in the laboratory and the field (as described above). The thermal dispersivity was fixed at a value of 
t = 0.001 m (Hatch et al., 2006), which is the suggested value for the spatial scales used in this investigation (i.e., the spacing between the sensor that ranges between 0.03 and 0.20 m) and when it is assumed that the thermal dispersivity has the same magnitude as the chemical dispersivity. The porosity was not measured because it was not possible to collect undisturbed sediment samples. Therefore, the porosity was fixed to a value of 38%, as recommended by Briggs et al. (2016). Additionally, a sensitivity analysis was performed to investigate the impact of this parameter on the determination of the water fluxes. Monte Carlo simulations were carried out to investigate the uncertainty of the estimated water fluxes. The Monte Carlo method consisted of 1000 realizations to determine statistical confidence intervals of the water flux estimates. The input parameters of each realization were thermal conductivity, volumetric heat capacity, and porosity. The thermal conductivity and the volumetric heat capacity were defined assuming a normal distribution with mean and standard deviation equal to those values measured in the field, whereas the porosity was assumed to be inversely correlated to the thermal 500
Annex X
19
conductivity (Lapham, 1989). In addition, the standard deviation of the porosity was
fixed at 4% (Gordon et al., 2012). After carrying out the 1000 realizations, the outliers
(data points that lie an abnormal distance from other values in a random sample from a
population) were removed from the results, and the mean, median and standard
deviation of the water fluxes were determined.
Finally, even when we had many sensors within each TR, the shallowest pair of sensors
was used to quantify the water fluxes as they allow estimation of the water exchange
near the water-sediment interface.
3.3 Results and discussion
3.3.1 Characterization of the thermal and hydraulic properties of the streambed
sediments
Table 3-2 presents the laboratory measurements of the thermal conductivity, heat
capacity and saturated hydraulic conductivity of the sediment samples collected at the
study site. Table 3-2 shows the field measurements of thermal conductivity and heat
capacity. The thermal properties of the streambed sediments are also shown in Figure
3-5 for both the laboratory and field measurements. The average thermal conductivity
values varied between 0.733 and 1.186 W m-1 K-1 (laboratory), and between 0.995 and
1.990 W m-1 K-1 (field). These values are smaller than thermal conductivities found in
similar studies. For instance, Gordon et al. (2012) utilized thermal conductivity values
of 1.3 and 2.26 W m-1 K-1. Also, the field thermal conductivities were larger than those
obtained in the laboratory (both the average values and the variability). For instance, at
the location of TR3, the average field thermal conductivity is 80% larger than that
obtained in the laboratory, and its variability is one order of magnitude higher. The only
location that did not exhibit large differences between the laboratory and field
measurements was at TR4, where the average thermal conductivity only increased by
4%, although the variability observed in the field was larger than that obtained in the
laboratory. The measured volumetric heat capacity ranged between 2.67 and 3.77 MJ m-
3 K-1 [mega Joules per cubic meter-Kelvin] (laboratory), and between 2.83 and 3.61 MJ
m-3 K-1 (field). These values are in agreement with the volumetric heat capacity
measured in different sediments (Briggs et al., 2016). On the other hand, the saturated
hydraulic conductivity values vary between 1.7 and 41.5 m day-1, which is consistent
with the values obtained for sand (Rawls et al., 1982).
Annex X
501
20 Temperature rod (TR) TR1 TR2 TR3 TR4 TR5 Thermal conductivity,  (W m-1 K-1) 0.974  0.004 0.733  0.124 1.108  0.091 1.186  0.019 1.087  0.061 Heat capacity, Cp (MJ m-3 K-1) 2.670 ± 0.520 3.774 ± 0.024 2.740 ± 0.547 2.905 ± 0.163 2.851 ± 0.150 Saturated hydraulic conductivity, Ks (m day-1) 1.7 ± 1.2 41.5 ± 31.8 13.0 ± 1.3 78.5 ± 7.5 14.3 ± 1.4 Table 3-2. Thermal conductivity (), heat capacity (Cp), and saturated hydraulic conductivity (Ks) of the sediment samples. These values correspond to the measurements performed in the laboratory (average value  standard deviation based on three repetitions). Temperature rod (TR) TR1* TR2* TR3** TR4*** TR5** Thermal conductivity,  (W m-1 K-1) 1.632 ± 0.963 0.995 ± 0.370 1.990 ± 1.600 1.241 ± 0.436 1.940 ± 1.45 Heat capacity, Cp (MJ m-3 K-1) 3.031 ± 0.408 2.836 ± 0.587 2.901 ± 1.329 2.903 ± 0.950 3.613 ± 1.268 * Three measurements; ** Four measurements; *** Two measurements. Table 3-3. Thermal conductivity () and heat capacity (Cp) for each TR. These values correspond to the measurements performed in the field (average value  standard deviation). Figure 3-5. Thermal properties obtained for each temperature rod (TR): (a) thermal conductivity measured in the laboratory (red) and in the field (green); and (b) volumetric heat 502
Annex X
21
capacity measured in the laboratory (red) and in the field (green). Vertical lines represent the
standard deviation of each measurement.
3.3.2 Temperature time series analysis
This section presents the results of the water exchange between the river and the aquifer
using the information collected by the TR’s. The sign convention of these results is that
positive fluxes are directed towards the aquifer and negative fluxes are directed towards
the river. The same convention is used for the hydraulic gradient. Also, the results of the
uncertainty analysis are reported using boxplots (also known as box-and-whisker
diagrams). In these plots, the lower edge of the box represents the first quartile of the
data (q1 or the 25th percentile), the horizontal red line inside the box corresponds to the
median of the data (q2 or the 50th percentile, i.e., the median), and the upper edge of the
box is the third quartile of the data (q3 or the 75th percentile). The data points that are
larger than q3 + 1.5(q3 –q1) or lower than q1 - 1.5(q3 –q1) are considered outliers and
consequently were eliminated from the data (Williamson et al., 1989). In our boxplots,
the plotted whiskers extend to the most extreme data value that is not an outlier.
3.3.2.1 TR1: water exchange fluxes and uncertainty analysis
The temperature time series at different depths at TR1 are shown in Figure 3-6(a).
These time series show variable daily fluctuations between 10 and 16 °C at 0.5 cm
depth. It can be observed that all the temperature time series increase their daily
minimum temperature as the austral summer (which corresponds to the wet season)
approaches. Figure3-6(a) also shows that the thermal amplitude is attenuated below 42.3
cm depth. Figure 3-6(b) shows the river stage and groundwater level measured by the
pressure transducers at TR1, and the precipitation measured at the Inacaliri
meteorological station (Muñoz et al., 2017). The water levels are relative to the
groundwater level measured at 0:00 a.m. on November 5th, 2016. The river stage
displays diurnal cycles, which may be used to understand some of the dominant
processes affecting the water balance of a river basin. According to Lundquist and
Cayan (2002), the mechanisms that explain diurnal cycles can be evapotranspiration,
infiltration in losing reaches, diurnal cycles in precipitation, and snowmelt. However,
because in our study we only have approximately three months of water level data, it is
not yet possible to determine the mechanism that explains these diurnal cycles. 3-6 (b)
shows that the effect of precipitation is more noticeable for groundwater levels than for
the river stage. The abrupt change in water levels observed on January 10th, 2017 was
Annex X
503
22 due to a sediment transport experiment where The Antofagasta (Chili) and Bolivia Railway Company Ltd. (FCAB) stopped the diversion of water into the FCAB intake (Mao, 2017). This experiment caused a spike in the river flow and a notable groundwater response that is much bigger than the effect of precipitation events on groundwater levels (this can be observed when comparing the increase in groundwater levels between the day of the sediment transport experiment and the day when the largest precipitation occurred). These data suggest that groundwater recharge occurs rapidly and that water leaks through the riverbed, flowing towards the aquifer. 3-6 (b) also shows that there are some perturbations in the water levels when there is no precipitation. The perturbations observed between November 28th, 2016 and December 2nd, 2016 are due to pump tests carried out in the well PW-BO, where the water pumped was discharged into the river (at a location downstream the well but upstream of TR1). The other perturbations that are observed in the water levels cannot be explained by the field data that have been collected for this study, although they might be related to the operation of the FCAB water intake (e.g., see fluctuation on November 19th, 2016 or December 11th, 2016) or due to natural processes (e.g., see small fluctuations occurring on November 14th, 2016). 3-6 (c) presents the hydraulic gradient at TR1. The average hydraulic gradient is approximately 0.33 m m-1 with a corresponding pressure head difference of 0.33 m between the river and the aquifer. This pressure head difference is larger than the estimated error that is associated with the reference position of each pressure transducer (~0.01 m), implying that the water exchange must be from the river to the aquifer. 504
Annex X
23
Figure 3–6. Thermal and hydraulic data collected at TR1: (a) temperature time series at different depths; (b) river
stage, groundwater level and daily precipitation; and (c) hydraulic gradient between the river and the aquifer.
Annex X
505
24 Figure 3-7(a) presents the daily mean water fluxes (cm3 cm-2 day-1 or cm day-1) between the river and the aquifer estimated with the average thermal properties at TR1 (continuous black line) and with Monte Carlo simulations (boxplot). These water fluxes were determined at 0.02 m depth using VFLUX. Figure 3-7 (a) shows that the water fluxes estimated with the average thermal properties vary between 9 and 35 cm day-1 during the monitoring period, with an average value of 21 cm day-1 (dashed line in Figure 3-7 (b)). The uncertainty analysis performed with the Monte Carlo simulations reveals that when using heat as a tracer without taking into account the hydraulic gradient, the probability of having a water exchange from the river to the aquifer is 98%. The uncertainty analysis yields water fluxes of 22  6 cm day-1 (mean  standard deviation of 921 simulations), with a median of 24 cm day-1. Figure 3-7 (b) shows the frequency distribution of the average fluxes obtained with the Monte Carlo simulations. It can be seen that the distribution is asymmetrical and that the estimated water fluxes are significantly different from zero, which implies that river-aquifer interactions are occurring. Figure 3-7 (c) shows the mean daily values of hydraulic head difference in the period of study. Overall, the direction of the water fluxes estimated using temperature time series agrees with the hydraulic gradient data. There is a good correlation between the timing of the variations in the water fluxes and of the fluctuations in the hydraulic head difference, although this timing shows a small shift at some days. The differences related to temporal fluctuations of water fluxes and hydraulic head difference are likely due to limitations of the analytical method used in this study to quantify the water exchange between the river and the aquifer (Hatch et al., 2006; Vandersteen et al., 2015), and are described in a subsequent section of this Chapter. 506
Annex X
25
Figure 3–7. (a) Daily mean water fluxes (cm3 cm-2 day-1) between the river and the aquifer at 0.02 m
depth, estimated with the average thermal properties at TR1 (continuous black line) and with Monte
Carlo simulations (boxplot). Positive values correspond to downwelling conditions and negative
values correspond to upwelling conditions. (b) Histogram of the mean water flux of each simulation
throughout all the studied period obtained with the Monte Carlo simulations (red columns). The
dashed line corresponds to the mean water flux estimated using the average thermal properties. The
numbers above the red columns are the frequency percentages below 10%. (c) Daily hydraulic head
difference between the river and the aquifer.
Annex X
507
26 3.3.2.2 TR2: water exchange fluxes and uncertainty analysis The temperature time series at different depths at TR2 are shown in Figure 3-8(a). These data show daily fluctuations between 10.5 and 14.0 °C at 3.0 cm depth. All the time series increase their minimum temperature as the austral summer approaches. Figure 3-8 (a) also shows that the thermal amplitude is almost completely attenuated below 38.0 cm depth. Figure 3-8 (b) shows the river stage and groundwater level measured by the pressure transducers at TR2, and the precipitation measured at the Inacaliri meteorological station (Muñoz et al., 2017). The water levels are relative to the groundwater level measured at 0:00 a.m. on November 5th, 2016. At this location, the river stage also displays diurnal cycles, although the variations in amplitude of diurnal water levels fluctuations are different than those observed at TR1, most likely due to the different cross sections of TR1 and TR2. Figure 3-8 (b) shows that, as above, the effect of precipitation on groundwater levels is more noticeable than on the river stage. As described before, the abrupt change in water levels observed at January 10th, 2017 was due to a sediment transport experiment (Mao, 2017). Similarly to TR1, the data from the sediment transport experiment suggest that groundwater recharge occurs rapidly and that water leaks through the riverbed, flowing towards the aquifer Figure 3-8 (b) also shows that there are some perturbations on the water levels when there is no precipitation. The perturbations observed between November 28th, 2016 and December 2nd, 2016 are due to pump tests carried out in the well PW-BO, where the water pumped was discharged into the river (at a location downstream of the well but upstream of TR1). The other perturbations that are observed in the water levels cannot be explained by the field data that have been collected for this study, although they might be related to the operation of the FCAB water intake (e.g., see fluctuation on November 19th, 2016 or December 11th, 2016) or due to natural processes (e.g., see small fluctuations occurring on November 16th, 2016). Note that at this location, the increase in river stage on November 19th, 2016 and December 11th, 2016 is greater than at TR1. This behavior occurs because of the different geometry of the cross section at each location. Figure 3-8 (c) presents the hydraulic gradient at TR2. The average hydraulic gradient is approximately 0.039 m m-1 with a corresponding pressure head difference of 0.029 m between the river and the aquifer. This pressure head difference is larger than the estimated error that is associated with the reference position of each pressure transducer (~0.01 m), implying that the water exchange must be from the river to the aquifer. However, the small pressure head difference implies small water fluxes that can be easily reversed, which can be observed in Figure 3-8 (c) during some short periods when the hydraulic gradient becomes negative. 508
Annex X
27
Figure 3-9(a) presents the daily mean water fluxes between the river and the aquifer
estimated with the average thermal properties at TR2 (continuous black line) and with
Monte Carlo simulations (boxplot). These water fluxes were determined at 0.055 m
depth using VFLUX. Figure 3-9 (a) shows that the water fluxes estimated with the
average thermal properties vary between 2 and 14 cm day-1 during the monitoring
period, with an average value of 6 cm day-1 (dashed line in Figure 3-9(b)). The Monte
Carlo simulations reveal that when using the Hatch et al. (2006) amplitude method, the
probability of having water exchange from the river to the aquifer is ~78%. The
uncertainty analysis yields water fluxes of 5  7 cm day-1 (mean  standard deviation of
986 simulations), with a median of 6 cm day-1. Therefore, even when there is an
uncertainty in the streambed thermal properties, in general the estimated fluxes have the
same direction as that expected from the vertical hydraulic gradient. Nonetheless, the
Hatch et al. (2006) amplitude method was unable to capture the hourly-flux reversal
episodes that are seen from the 15-min hydraulic gradient data. This limitation occurred
because the Hatch et al. (2006) method solves the transient regime using an analytical
solution that is valid for steady and uniform fluid velocity (Stallman, 1965). As
explained by Hatch et al. (2006), this analytical solution has represented well slow
fluctuations in seepage rates (e.g., at temporal scales of days, weeks or months) but
deviates from the exact solution when rapid changes occur (see a subsequent section of
this Chapter for more details). Figure 3-9 (b) shows the frequency distribution of the
fluxes obtained with the Monte Carlo simulations. It can be seen that the distribution is
asymmetrical and that the estimated water fluxes are significantly different from zero,
which implies that river-aquifer interactions are occurring. Figure 3-9(c) shows the
mean daily values of hydraulic head difference in the period of study. Overall, the
direction of the water fluxes estimated using temperature time series agrees with the
hydraulic gradient data. There is a good correlation between the timing of the variations
in the water fluxes and of the fluctuations in the hydraulic head difference, although this
timing shows a small shift at some days. The differences related to temporal fluctuations
of water fluxes and hydraulic head difference are likely due to limitations of the
analytical method used in this study to quantify the water exchange between the river
and the aquifer (Hatch et al., 2006; Vandersteen et al., 2015), and are described in a
subsequent section of this Chapter.
Annex X
509
28 Figure 3–8. Thermal and hydraulic data collected at TR2: (a) temperature time series at different depths; (b) river stage, groundwater level and daily precipitation; and (c) hydraulic gradient between the river and the aquifer 510
Annex X
29
Figure 3–9. (a) Daily mean water fluxes (cm3 cm-2 day-1) between the river and the aquifer at 0.055 m depth,
estimated with the average thermal properties at TR2 (continuous black line) and with Monte Carlo
simulations (boxplot). Positive values correspond to downwelling conditions and negative values correspond to
upwelling conditions. (b) Histogram of the mean water flux of each simulation throughout all the studied
period obtained with the Monte Carlo simulations (red columns). The dashed line corresponds to the mean
water flux estimated using the average thermal properties. The numbers above the red columns are the
frequency percentages below 10%. (c) Daily hydraulic head difference between the river and the aquifer.
Annex X
511
30 3.3.2.3 TR3: water exchange fluxes and uncertainty analysis The temperature time series at different depths at TR3 are shown in Figure 3-10(a). These time series show daily fluctuations between 10 and 16 °C at 4.0 cm depth. It can be observed that all the temperature time series increase their minimum temperature as the austral summer (which corresponds to the wet season) approaches. Figure 3-10 (a) also shows that the daily thermal amplitude is almost completely attenuated below 38 cm depth. Figure 3-10 (b) shows the river stage and groundwater level measured by the pressure transducers at TR3, and the precipitation measured at the Inacaliri meteorological station (Muñoz et al., 2017). The water levels are relative to the groundwater level measured at 0:00 a.m. on November 5th, 2016. At this location, the river stage also displays diurnal cycles, although the variations in amplitude of diurnal water levels fluctuations are different from those observed at TR1 and TR2, most likely due to the different cross section properties of each TR. Figure 3-10 (b) shows, as above, that the effect of precipitation on groundwater levels is more noticeable than on the river stage. The abrupt change in water observed at January 10th, 2017 was due to a sediment transport experiment (Mao, 2017). These data suggest that groundwater recharge occurs rapidly when the river stage increases. Figure 3-10 (b) also shows that there are some perturbations in the water levels when there is no precipitation. These perturbations are due to human activities or natural processes (as discussed previously). At TR3, the perturbations of the river stage are less noticeable than at TR1 and TR2 because the cross section at TR3 is wider than at TR1 and TR2. Consequently, for the same water flow fluctuation the water levels will vary less at TR3. Figure 3-10 (c) presents the hydraulic gradient at TR3. The average hydraulic gradient is approximately 0.089 m m-1 with a corresponding pressure head difference of 0.05 m between the river and the aquifer. This pressure head difference is slightly larger than the estimated error that is associated with the reference position of each pressure transducer (~0.01 m). Hence, the water exchange should go from the river to the aquifer. Nonetheless, since the pressure head difference is very small the associated fluxes are also going to be very small and can be easily reversed if the water stage in the river decreases or if the water level in the aquifer increases. Figure 3-11 (a) presents the daily mean water fluxes (cm3 cm-2 day-1) between the river and the aquifer estimated with the average thermal properties at TR3 (continuous black line) and with Monte Carlo simulations (boxplot). These water fluxes were determined at 0.055 m depth using VFLUX. Figure 3-11 (a) shows that the water fluxes estimated with the average thermal properties vary between -19 and 3 cm day-1 during the monitoring period, with an average value of -9 cm day-1 (dashed line in Figure 3-11 (b)). Therefore, at this location, the direction of the water flux determined using thermal 512
Annex X
31
methods does not agree with that determined using the hydraulic gradient data. This
error is probably due to uncertainties related to the variability of the streambed
volumetric heat capacity. The uncertainty analysis performed with the Monte Carlo
simulations analysis shows that the water flux estimations at TR3 have a large
uncertainty (on the same order of magnitude as the mean water flux), and reveals that
the probability of having water flowing from the aquifer to the river is ~56%. The
uncertainty analysis yields water fluxes of -7  21 cm day-1 (mean  standard deviation
of 888 simulations), with a median of -4 cm day-1. Given that the estimated water flux
shows a large variability, which hinders the determination of the water direction, the
actual water flux direction at TR3 was determined by manually measuring the hydraulic
gradient. This measurement was performed on the field campaign carried out on
January 31st, 2017. In this field campaign, a positive hydraulic gradient was found (see
Figure 3-11 (c)), which demonstrated that the water was flowing from the river towards
the aquifer. Figure 3-11 (b) shows the frequency distribution of the fluxes obtained with
the Monte Carlo simulations. It can be seen that the distribution is asymmetrical and
that the estimated water fluxes are significantly different from zero, which implies that
river-aquifer interactions are occurring. Figure 3-11 (c) shows the mean daily values of
hydraulic head difference in the period of study. Overall, the changes in the estimated
water fluxes are very small, which could occur because the hydraulic head difference is
also small. As described before, the large uncertainty in the volumetric heat capacity
measured at TR3 and the small hydraulic gradient hinders the precise determination of
the direction and temporal fluctuations of the water fluxes.
Annex X
513
32 Figure 3–10. Thermal and hydraulic data collected at TR3: (a) temperature time series at different depths; (b) river stage, groundwater level and daily precipitation; and (c) hydraulic gradient between the river and the aquifer. 514
Annex X
33
Figure3–11. (a) Daily mean water fluxes (cm3 cm-2 day-1) between the river and the aquifer at 0.055 m depth,
estimated with the average thermal properties at TR3 (continuous black line) and with Monte Carlo
simulations (boxplot). Positive values correspond to downwelling conditions and negative values correspond
to upwelling conditions. (b) Histogram of the mean water flux of each simulation throughout all the studied
period obtained with the Monte Carlo simulations (red columns). The dashed line corresponds to the mean
water flux estimated using the average thermal properties. The numbers above the red columns are the
frequency percentages below 10%. (c) Daily hydraulic head difference between the river and the aquifer.
Annex X
515
34 3.3.2.4 TR4: water exchange fluxes and uncertainty analysis The temperature time series at different depths within the sediments at TR4 are shown in Figure 3-12(a). These time series show daily fluctuations between 10 and 16 °C at 8.6 cm depth. Figure 3-12 also shows that the daily thermal amplitude is almost completely attenuated below 71.2 cm depth. Note that the attenuation of the thermal signal at TR4 occurs deeper than at TR1, TR2 and TR3, suggesting that heat is being transported deeper into the sediments by water advection. At this location, the piezometer installed to measure the hydraulic head of the aquifer was dry, implying that the water level was deeper than the openings of the piezometer (> 1.0 m). Therefore, the TR4 was actually measuring the temperature of the sediments in the unsaturated zone (also called vadose zone) and water is flowing from the river to the aquifer through the vadose zone. During a field campaign performed on November 4th, 2016, water was poured into the piezometer and we observed that the water level inside it decreased rapidly, suggesting that gravity flow occurred at this location (Jury and Horton, 2004), where water flows through the vadose zone driven by a unit hydraulic gradient, i.e., the total head gradient is equal to 1. According to Buckingham-Darcy’s Law (Jury and Horton, 2004), which states that water flux in porous media is proportional to the hydraulic gradient, one would expect a relatively constant water flux, at a rate very similar to the sediments’ hydraulic conductivity, and any variation on the water flux is expected to occur as a result of river stage fluctuations. Figure 3-12 (b) shows the river stage (relative to the river level measured at 0:00 a.m. on November 5th, 2016) and the daily precipitation measured at the Inacaliri meteorological station (Muñoz et al., 2017). It can be seen that the river stage has daily fluctuations that occur along the entire monitoring period. These fluctuations have different amplitude to those observed at TR1-TR3, most likely due to the different cross section properties of each TR. The fluctuations in the river stage observed between November 28th and December 2nd correspond to water flow contributions to the river associated with pumping tests performed at the PW-BO well, which is located near the Bolivia-Chile international border. Also, the rise in river stage that is shown in January 10th, 2017 was due to a sediment transport experiment (Mao, 2017). Figure 3-12 (b) clearly shows the effect of daily precipitation on the river stage at TR4, where the water level increased as much as 0.15 m. Note that the formulation of the time series analysis can be applied to saturated or unsaturated systems (Hatch et al., 2006). At TR4, in the zone near the water-sediment interface where the temperature loggers used to determine the water fluxes are installed, the sediments should have moisture conditions near saturation (Brunner et al., 2009). 516
Annex X
35
Therefore, the Hatch et al. (2006) amplitude method applied using the assumption that
the sediments are fully saturated is a reasonable approximation to determine the water
flux infiltrating into the sediments, even when unsaturated conditions are expected to
occur in between the zone that is below the riverbed sediments and the top of the
capillary fringe (Brunner et al., 2009).
Figure 3-13(a) presents the daily mean water fluxes (cm3 cm-2 day-1) between the river
and the sediments estimated with the average thermal properties at TR4 (continuous
black line) and with Monte Carlo simulations (boxplot). These water fluxes were
determined at 0.13 m depth using VFLUX. Figure 3-13(a) shows that the water fluxes
estimated with the average thermal properties vary between 34 and 56 cm day-1 during
the monitoring period, with an average value of 46 cm day-1 (dashed line in Figure 3-
13(b)). The uncertainty analysis performed with the Monte Carlo simulations reveals
that the probability of having a water exchange from the river to the aquifer is 100%,
which agrees with the fact that the piezometer installed to measure the groundwater
level was dry (installed in the vadose zone). The Monte Carlo simulations yield water
fluxes of 44  7 cm day-1 (mean  standard deviation of 987 simulations), with a median
of 44 cm day-1. Figure 3-13(b) shows the frequency distribution of the fluxes obtained
with the Monte Carlo simulations. It can be seen that the distribution is symmetrical and
that the estimated water fluxes are always greater than zero, which implies that riveraquifer
interactions are occurring. Figure 3-13(c) shows the mean daily values of the
river stage. The river stage is relative to the river stage measured at 0:00 a.m. on
November 5th, 2016. At TR4, the water fluxes estimated using temperature time series
are always directed downwards, which agrees with the condition of gravity flow. Also,
there is a correlation between the timing of the variations in the water fluxes and of the
fluctuations in the river stage, although this timing shows a small shift at some days.
The differences related to temporal fluctuations of water fluxes and hydraulic head
difference are likely due to limitations of the analytical method used in this study to
quantify the water exchange between the river and the aquifer (Hatch et al., 2006;
Vandersteen et al., 2015), and are described in a subsequent section of this Chapter.
Annex X
517
36 Figure 3–12. Thermal and hydraulic data collected at TR4: (a) temperature time series at different depths; (b) river stage and daily precipitation. 518
Annex X
37
Figure 3–13. (a) Daily mean water fluxes (cm3 cm-2 day-1) between the river and the sediments at 0.13 m
depth, estimated with the average thermal properties at TR4 (continuous black line) and with Monte Carlo
simulations (boxplot). Positive values correspond to downwelling conditions and negative values
correspond to upwelling conditions. (b) Histogram of the mean water flux of each simulation throughout
all the studied period obtained in the Monte Carlo simulations (red columns). The dashed line corresponds
to the mean water flux estimated using the average thermal properties. The numbers above the red columns
are the frequency percentages below 10%. (c) Daily relative river stage in the study period.
Annex X
519
38 3.3.2.5 TR5: water exchange fluxes and uncertainty analysis The temperature time series at different depths at TR5 are shown in Figure 3-14(a). These time series show variable daily fluctuations between 16.5 and 19.0 °C at the water-sediment interface. At TR5, the daily thermal amplitude and the increment of minimum temperatures as the austral summer approaches are smaller than those monitored at the other TR’s. This difference in the thermal behavior is attributed to SPW-DQN, an artesian well that flows into the river ~10-20 m upstream of TR5 (see next chapters for more details about Well SPW-DQN). Figure 3-14(a) also shows that the thermal amplitude is attenuated below 49 cm depth. Figure 3-14(b) shows the river stage and groundwater level measured by the pressure transducers at TR5, and the precipitation measured at the Inacaliri meteorological station (Muñoz et al., 2017). The water levels are relative to the groundwater level measured at 0:00 a.m. on November 5th, 2016. As observed in the previous TR’s, at TR5 the river stage and the groundwater levels also exhibit daily cycles. These daily cycles have different amplitude to those observed at TR1-TR4, most likely due to the different cross section properties of each TR. The impact of daily precipitation on the river stage at TR5 is not as noticeable as in the other TR’s, which can also be explained by the water contribution of SPW-DQN. Figure 3-14 (b) also shows that there are some perturbations on the water levels when there is no precipitation. These perturbations are due to human activities or natural processes (as discussed previously). The abrupt change in water levels and observed at January 10th, 2017 was due to a sediment transport experiment (Mao, 2017). These data suggest that groundwater recharge occurs rapidly when the river stage increases. The average hydraulic gradient is approximately 0.21 m m-1 with a corresponding pressure head difference of 0.21 m between the river and the aquifer (Figure 3-14(c)). This pressure head difference is larger than the estimated error that is associated with the reference position of each pressure transducer (~0.01 m), implying that the water exchange must be from the river to the aquifer. Figure 3-15(a) presents the daily mean water fluxes (cm3 cm-2 day-1) between the river and the aquifer estimated with the average thermal properties at TR5 (continuous black line) and with Monte Carlo simulations (boxplot). These water fluxes were determined at 0.14 m depth using VFLUX Figure 3-15(a) shows that the water fluxes obtained with the average thermal properties vary between -4 and 25 cm day-1 during the monitoring period, with an average value of 9 cm day-1(dashed line in Figure 3-15(b)). Although the estimated water fluxes at TR5 display more variability than those estimated at the other TR’s, the water fluxes at TR5 always remain of the same order of magnitude. The uncertainty analysis performed with the Monte Carlo reveals that the probability of having a water exchange from the river to the aquifer is 62%, which is in agreement 520
Annex X
39
with the measured vertical hydraulic gradient. The uncertainty analysis yields water
fluxes of 3  16 cm day-1 (mean  standard deviation of 901 simulations), with a median
of 6 cm day-1. Figure 3-15 (b) shows the frequency distribution of the water fluxes
obtained with the Monte Carlo simulations. It can be seen that the distribution is
asymmetrical and that the estimated water fluxes are significantly different from zero,
which implies that river-aquifer interactions are occurring. Nonetheless, the uncertainty
in the water fluxes at TR5 is large most likely due to the smaller amplitude of the water
river temperature (compared to the other TR’s), which occurs because of the influence
of the artesian well SPW-DQN that continuously discharges warm groundwater into the
river. Figure 3-15 (c) shows the mean daily values of hydraulic head difference in the
period of study. The estimated water fluxes and the hydraulic head difference display
small temporal variations. In fact, the most significant change in hydraulic head
difference occurs during the time when the sediment transport experiment occurred, i.e.,
during January 10th, 2017 when FCAB stopped the diversion of water into the FCAB
intake (Mao, 2017). The increase in magnitude of the water fluxes directed towards the
groundwater at TR5 during January 10th, 2017 is more noticeable than at TR1, TR2 and
TR3. This more noticeable increase in water flux and in hydraulic head difference is
likely due to the change of a river reach that is controlled by a continuous and nearly
constant supply of warm groundwater (coming from the artesian well SPW-DQN) to a
river reach controlled by a dominant cooler surface water flow (recall that the sediment
transport experiment increased significantly the flow along the river reach).
Annex X
521
40 Figure 3–14. Thermal and hydraulic data collected at TR5: (a) temperature time series at different depths; (b) river stage, groundwater level and daily precipitation; and (c) hydraulic gradient between the river and the aquifer. 522
Annex X
41
Figure 3–15. (a) Daily mean water fluxes (cm3 cm-2 day-1) between the river and the aquifer at 0.14 m
depth, estimated with the average thermal properties at TR5 (continuous black line) and with Monte Carlo
simulations (boxplot). Positive values correspond to downwelling conditions and negative values
correspond to upwelling conditions. (b) Histogram of the mean water flux of each simulation throughout all
the studied period obtained in the Monte Carlo simulations (red columns). The dashed line corresponds to
the mean water flux estimated using the average thermal properties. The numbers above the red columns
are the frequency percentages below 10%. (c) Daily hydraulic head difference between the river and the
aquifer.
Annex X
523
42 3.3.2.6 Comparison of the measurements between the TR’s and limitations of the thermal method used in this investigation Figure 3-16 shows the temperature time series for all the TR’s analyzed in this study during the period November 5th of 2016 to January 31st of 2017. Figure 3-16 (a) shows that the amplitude of the temperature time series measured in the shallow sediments combined with the resolution of the iButtons temperature loggers allows determination of the water exchange using the Hatch et al. (2006) amplitude method. The maximum mean daily amplitude values, measured near the water-sediment interface, were: 3.4 °C at TR1; 1.3 °C at TR2; 1.6 °C at TR3; 4 °C at TR4; and 1 °C at TR5. Note that locations with small thermal amplitude may also indicate groundwater input into the river not necessarily associated with water exchange occurring across the water-sediment interface. For instance, as explained in Chapters 4 and 5, the small thermal amplitude observed at TR5 is the result of a warm groundwater discharge that comes from an artesian well (SPW-DQN). Similar processes to those occurring at SPW-DQN can occur for the springs that discharge into the river, e.g., the small thermal amplitude observed at TR2 and TR3 is likely the result of spring discharge from the walls of the ravine. Nonetheless, our results show, with a high probability, that in the streambed water is flowing from the river towards the aquifer at these locations. The water fluxes estimated using the mean thermal properties at each TR are presented in Figure 3-17. The mean water fluxes were 21, 6, -9, 46, and 9 cm day-1, at TR1, TR2, TR3, TR4, and TR5, respectively. Nonetheless, the Monte Carlo analysis yielded estimates of water fluxes with their uncertainty: 22  6 cm day-1 (mean  standard deviation) (TR1); 5  7 cm day-1 (TR2); -7  21 cm day-1 (TR3); 44  7 cm day-1 (TR4); and 3  16 cm day-1 (TR5). In general, all the estimated water fluxes are smaller than 60 cm day-1. Therefore, the water exchange through the riverbed is relatively small compared to the river flow (see Chapter 5 for more details). The magnitude of the fluxes found in this study is of the same order as other investigations. For instance, Gordon et al. (2012) found positive fluxes of ~86 cm day-1 at 15 and 25 cm depth. On the other hand, the extinction depths are similar to the ones obtained by Briggs et al. (2012); between 30 and 60 cm, with amplitude signals over 0.2-1.2 °C. The river stage data exhibit daily cycles, although the variations in amplitude between the TR’s were different. These differences appear to be related to the cross-section properties of the river at each TR. 524
Annex X
43
Figure 3–16. Temperature time series for all the TR’s. (a) TR1. (b) TR2. (c) TR3. (d) TR4. (e) TR5.
Annex X
525
44 Figure 3-17. Comparison of the daily mean water fluxes between the river and the aquifer estimated using the Hatch et al. (2006) amplitude method and the average thermal properties at each TR. The main limitation of the Hatch et al. (2006) amplitude method is that it uses an analytical solution valid for steady and uniform fluid velocity parallel with the vertical direction. Also, the analytical solution applies when the heat properties of the medium and fluid are constant in space and time, and when there is thermal equilibrium between the porous media and the fluid. The main practical consequence of these limitations is that the Hatch et al. (2006) amplitude method is not sensitive enough when abrupt changes occur in the seepage rates or in the water flux direction. The significance of this limitation will depend on the magnitude and frequency of seepage rate fluctuations. In many natural systems the seepage rates will vary slowly over days, weeks, months or even seasons. Therefore, in these types of systems this limitation is not relevant (Hatch et al., 2006). Nonetheless, rapid changes can occur in natural system as the result of human activities, e.g., in our case we analyzed time periods where pump tests and a sediment transport experiment changed abruptly the river stage. In those cases, this limitation of the analytical method may be more relevant and will result in a different timing of the change in water flux across the sediments. Also in our analysis, because we collected thermal data at 1 h intervals, the analytical method estimates a 2-h constant water flux (Gordon et al., 2012); therefore, this method is not able to capture the variability that was observed in the hydraulic head data that were collected at 15-min time intervals. In addition, the use of the Hatch et al. (2006) amplitude method alone, i.e., without using the phase shift as an additional input, may lead to error in flux estimations when field data are used to populate the model or when three-dimensional flow structures appear in the seepage velocity field (Hatch et al., 2006; Vandersteen et al., 2015). The Hatch et al. (2006) amplitude method also assumes that the porous medium is homogeneous, which typically does not occur in natural systems. This 526
Annex X
45
limitation increases the uncertainty in the estimation of the water fluxes when the
porous medium is stratified.
Despite all the limitations discussed above, as shown by the Monte Carlo simulations, in
this study the water fluxes estimated using thermal methods are significantly different
than zero in most of the TR’s. Therefore, the interactions between groundwater and
surface waters are undeniable. These interactions were also observed using the
hydraulic head difference that was measured with pressure transducers. In general
terms, the results obtained in this work suggest that the water fluxes are smaller than 60
cm day-1 (Figure 3-17) and that water flows from the river to the aquifer. However,
since the observations are point-in-space measurements, it may be possible that the flow
is reversed in other locations of the river reach. In fact, in the hyporheic zone of
mountain streams it is common to find both downwelling and upwelling zones
(Wroblicky et al., 1998; Buffington and Tonina, 2009; Tonina and Buffington, 2009;
Westhoff et al., 2011).
3.4 Conclusions
Temperature time series, combined with the sediment thermal properties and hydraulic
head data, can be used to quantify the fluxes that occur between the river and the
aquifer.
The measured field thermal conductivity values ranged between 0.995 and 1.990 W m-1
K-1, and are smaller than the values found in similar investigations (1.3-2.26 W m-1 K-1).
The estimated volumetric heat capacity was similar to that obtained in other studies
(2.83-3.61 MJ m-3 K-1). The saturated hydraulic conductivity values agree with values
for sand, ranging between 1.7 and 41.5 m day-1.
The change in amplitude of the sediment temperature time series can be used to estimate
water fluxes between groundwater and surface waters for particular locations in a river.
Our results, based on the thermal analysis combined with Monte Carlo simulations,
show that at four of the locations investigated (TR1, TR2, TR4, and TR5), it is very
likely that water flows from the river into the aquifer through the streambed. At TR3,
the thermal analysis yielded fluxes going from the aquifer towards the river, but the
direction of the hydraulic gradient at that location implies that water flows in the other
direction (from the river to the aquifer). At TR1, we estimated water fluxes of 22  6 cm
day-1 (with a probability of 98% that water is flowing towards the aquifer). At TR2, we
estimated water fluxes of 5  7 cm day-1 (with a probability of 78% that water is
flowing towards the aquifer). At TR3, we estimated water fluxes of -7  21 cm day-1
(with a probability of 56% that water is flowing towards the river). At TR4, we
Annex X
527
46 estimated water fluxes of 44  7 cm day-1 (with a probability of 100% that water is flowing towards the aquifer). And at TR5, we estimated water fluxes of 3  16 cm day-1 (with a probability of 62% that water is flowing towards the aquifer). The measured vertical hydraulic gradient also indicated that water flows from the river to the aquifer. Therefore, it is very probable that in all the investigated locations, water flows from the river streambed to the aquifer (losing river). Nonetheless, in some locations (e.g., TR2, TR3) the hydraulic gradient was small implying that changes in the river stage or in the aquifer water level could reverse the direction of the water fluxes. Even when the analytical method used in this investigation has limitations that can led to differences between the actual and estimated water fluxes, and that the estimated water fluxes were small ( 60 cm day-1), our uncertainty analysis showed that the water exchange between surface water and groundwater is different than zero in most of the TR’s. Therefore, the interactions between groundwater and surface waters are undeniable. These interactions were also supported with the hydraulic head difference that was measured with pressure transducers. It is important to recall that the water exchanges between the river and the aquifer quantified in this study are point-in-space measurements. Therefore, they do not mean that the same behavior occurs along the entire river. 4 RIVER-AQUIFER INTERACTIONS USING DISTRIBUTED TEMPERATURE SENSING ALONG THE RIVER 4.1 Introduction Fiber-optic distributed temperature sensing (FO-DTS) is a method that provides coverage in both space and time that can be used continuously to monitor real-time data in different environments and at spatial scales that range from centimeters to kilometers (Selker et al., 2006). DTS methods offer many advantages over traditional monitoring systems. For instance, due to the nature of this measurement, there is no variability due to different sensors and sensor measurement scale, as all the observations are carried out along the same fiber-optic cable (Tyler et al., 2009; Suárez et al., 2011a). The main objective of this study is to use DTS methods to investigate river-aquifer interactions. The specific objective is to identify potential locations of groundwater discharge into the Silala River using FO-DTS. Note that because we deployed the fiber-optic cable in the river, this approach can only be used to determine locations where water flows into the river (e.g., due to springs or due to groundwater upwelling locations). To determine downwelling sections, fiber-optic cables must be buried at a 528
Annex X
47
specific depth below the water-sediment interface all along the river reach in a similar
way to that presented by Mamer and Lowry (2013), or using vertical high-resolution
systems (Vogt et al., 2010; Briggs et al., 2012, 2016).
The structure of this chapter is as follows: first, the theory of FO-DTS is presented.
Then, we describe the field deployment carried out along the Silala River. Then, the
methods used to calibrate and validate the thermal measurements are presented. Next,
the approach followed to locate the cable along the river is explained and the method to
filter the thermal data that do not correspond to water temperature is described. Later,
the results and discussion are presented. Finally, the main conclusions are drawn.
4.2 Materials and methods
4.2.1 Distributed temperature sensing (DTS) theory
Fiber-optic DTS systems utilize Raman spectra scattering to estimate temperature along
the length of the fiber, achieving temperature resolutions as small as ±0.01 °C, and
spatial and temporal resolutions of 0.25-2 m and 1-60 s, respectively, for cables up to 10
km (Selker et al., 2006). DTS instruments typically use a 10 or 20 ns laser pulse to
illuminate the optical fiber. After emitting the laser pulse, the backscattered signals
return to the DTS instrument, where they are collected in discrete time periods by the
detector unit. The DTS instrument repeats the pulse and data collection continuously for
temporal integration periods specified by the user. This approach delivers averaged
temperature measurements over space and time. For instance, an instrument configured
with a sampling resolution of 0.25 m and a temporal resolution of 5 min delivers the
mean temperature over the 0.25 m section, which is also the temporal average of the
temperature during the time of measurement (this temporal averaging is commonly
referred to as the integration time).
To measure the temperature along the optical path (also referred to as temperature
trace), a DTS instrument emits laser pulses at a known wavelength into the optical fiber.
As light travels along the fiber, a fraction of the incident light is scattered by
interactions between the light and the crystalline structure, and vibration frequency of
the fiber itself (Suárez et al., 2011a). Both elastic and inelastic light scattering occurs,
and fiber-optic DTS systems typically utilize the Raman inelastic scattering to infer the
temperature along the length of the cable (Tyler et al., 2009). The Raman scattering is
due to the interactions between the photons and the vibrating molecules within the
lattice of the glass fiber. This interaction results in a predictable frequency shift. The
scattered light shifted to longer wavelengths than the incident light is known as the
Stokes component, while the scattered light shifted to shorter wavelengths is called the
Annex X
529
48 anti-Stokes component. Although the wavelengths of the Raman signals are predictable, their intensities are not. The intensity of the anti-Stokes backscattering depends strongly on the temperature of the silica molecules of the fiber, while the intensity of the Stokes backscattering depends weakly on this temperature. Because of this differential temperature dependence, the ratio of the anti-Stokes and Stokes signals can be used to determine the temperature, T(z,t) [K], of the fiber at the point of scattering, z [m], at time t [s] (van de Giesen et al., 2012): 0,,ldn,SaSzTztIztCtIztzz (4.1) where IS (z) and IaS (z) [arbitrary units] are the intensities of the Raman Stokes and anti-Stokes backscatter at a distance z along the fiber,  [K] is the shift in energy between a photon at the wavelength of the incident light and the scattered Raman photon, C [-] is a calibration parameter that includes properties of the incident laser and the DTS instrument itself, and  [m-1] is the differential attenuation between the anti-Stokes and Stokes signals within the fiber. Note that equation (4.1) considers that the differential attenuation can be variable along the fiber. The distance of the point of light scattering, z, is calculated by time-domain reflectometry using the speed of light in the glass fiber, c [m s-1], which is dependent on the frequency of the light and the index of refraction of the fiber,  [-] (Suárez et al., 2011a): 2ctz (4.2) where t [s] is the propagation time of the light in the forward and backward directions. For a more detailed description of the DTS theory, the reader is referred to the work of Dakin et al. (1985), Rogers (1999), Suárez et al. (2011a), Hausner et al. (2011), and van de Giesen et al. (2012). 4.2.2 Field deployment With the aim of identifying gaining reaches along the Silala River, DTS measurements were carried out between September 30th, 2016 and January 13th, 2017. The fiber-optic cable used was a multimode loose tube cable with four fibers, a stainless steel armoring and a red polyamide jacket (BRUsens temperature 85 °C, Brugg Kabel AG, Windisch, Switzerland). The cable has an outer diameter of 3.8 mm and a total length of 2 km. An XT-DTS (Silixa, Hertfordshire, UK) instrument was used to perform the DTS 530
Annex X
49
measurements. This instrument has a spatial resolution of ~0.5 m (~0.25 m sampling
resolution) and a temporal resolution of 5 s or more.
Figure 4-1 presents the path of the cable along the river and some pictures of the
deployment. Figure 4-2 shows a photograph of the site where the DTS instrument was
kept. The deployment of the cable was carried out between September 27th and 29th,
2016. The beginning of the cable is located approximately 2 km downstream of the
international border between Chile and Bolivia. The cable was installed going upstream
in the river for an extent of approximately 1.3 km. The cable passed through seven
culverts located along the river (see Figure 4-1). Of the 700 m of cable that were not
deployed in the river, ~150 m were used for calibration purposes and the remainder of
the cable was stored inside a wooden box that contained the cable spool (Figure 4-2). In
some locations along the river it was not possible to avoid obstructions (e.g., due to the
presence of pipes, weirs, and in some places where the vegetation was very dense) so
many sections of the cable were exposed to air (e.g., see Figure 4-1(c)). In some of these
sections the cable was exposed to the sun while in others the cable was shaded by
vegetation. Additionally, two sections along the cable path were monitored with
independent temperature measurements using HOBO Water Temperature Pro v2 Data
Logger (Onset Computer Corporation, Bourne, MA; with an accuracy of 0.2 °C and a
resolution of 0.02 °C). The first of these sections was located 1473 m upstream of the
DTS instrument (distance measured along the cable), and was a 52-m section, which
was coiled and submerged under the water just downstream of culvert 7 (Figure 4-1(e)).
The second section was located at the end of the cable, 1978 m upstream of the DTS
instrument (distance also measured along the cable). This section comprised 8 m of
cable, which were coiled and stored outside the river and beneath the vegetation. Figure
4-3(a) presents a diagram of the DTS deployment, and an example of the raw data, i.e.,
Stokes and anti-Stokes signals and temperature traces, are shown in Figure 4-3(b) and
(c). As shown in Figure 4-3(c), even when the fiber-optic cable was physically installed
in the downstream-upstream direction, we referenced the distance along the cable in the
upstream-downstream direction, i.e., in the flow direction. Thus, the location of the
cable that is at the most upstream point underwater is used as the origin of the distance
along the cable. This convention is used in the rest of this document unless noted.
Annex X
531
50 Figure 4-1. Schematic representation of the fiber-optic cable path and photographs of selected locations along the cable: (a) cable inside culvert; (b) cable in a zone with dense vegetation; (c) cable outside the river to avoid a pipe; (d) cable in the streambed in an open zone; and (e) cable coil that was submerged in the river and typically sank into the river bed. 4.2.3 Calibration and quality analysis of measurements Once the cable was installed and before calibrating the temperatures along the fiber optic, the integrity of the cable was verified through the analysis of the Stokes and anti-Stokes signals (Tyler et al., 2009). These did not show zones with local step losses (Figure 4-3(b)), and demonstrated that the cable was not damaged during the deployment. Also, the intensity of the Raman spectra signals is larger than 500 532
Annex X
51
[arbitrary units], which allows the collection of high-quality data (Personal
Communication with Silixa technicians).
At the end of the cable, two of the four fibers were fused to allow double-ended
measurements (see details in Suárez et al., 2011b; Hausner et al., 2011; van de Giesen et
al., 2012). Double-ended measurements facilitate the calibration of DTS systems in
cables longer than 1 km, improving the thermal resolution of the data collected in the
middle of the fiber path compared to single-ended measurements (Tyler et al., 2009).
The DTS instrument was configured with a sampling resolution of 0.25 m and a
temporal integration of 5 min (measurements of 2.5 min per end of the fiber).
Figure 4-2. Silala River DTS deployment showing some of the features of the installation. The
DTS was powered using solar energy (solar panels and batteries). Two reservoirs (ambient and
cool temperature baths) were used for calibration purposes. The DTS instrument is stored
inside a NEMA box and a section of the cable was kept on its spool.
Annex X
533
52 Figure 4-3. Silala River DTS installation: (a) schematic of the field installation; (b) raw Raman spectra data (recorded by the DTS instrument in arbitrary units linearly related to the power of the scattered signals) and reference locations for a 5-min temperature trace sample (10/02/2016 – 8:00). In this subfigure, the distance along the cable uses the instrument position as the coordinate’s origin; (c) temperatures and the locations of reference zones (Zi) for 5-min temperature trace samples at daily minimum (10/02/2016 – 8:00) and maximum temperatures (10/02/2016 – 15:08). In this subfigure, the distance along the cable uses the most upstream point underwater as the coordinate’s origin. 534
Annex X
53
To calibrate the temperature along the entire cable, the recommendations presented in
the scientific literature were followed (Hausner et al., 2011; Suárez et al., 2011b; van de
Giesen et al., 2012). Two water reservoirs were placed near the DTS instrument (Figure
4-2 and Figure 4-3(a)) and in each reservoir the temperature was monitored using
platinum resistance thermometers (PT100, with a resolution of ±0.01°C) that are
integrated into the XT-DTS. In the reservoirs, aquarium air pumps were used to mix the
water in order to achieve a uniform temperature. At the beginning of the field campaign,
one of the water reservoirs was filled with hot water (this will be referred to as “ambient
temperature bath”, see Figure 4-2 and Figure 4-3(a)) and the other one was filled with a
mixture of water and ice (referred to as “cool temperature bath”, as shown in Figure 4-2
and Figure 4-3(a)). Therefore, a significant temperature difference was observed at the
beginning of the measurements, which allows clear identification of each of the zones in
the thermal profiles logged with the DTS instrument. Two coils of cable were inserted
into the ambient temperature reservoir (Figure 4-3(a)). These coils had lengths of
approximately 30 and 20 m. In the cool temperature bath, one coil of cable of
approximately 25 m was introduced. Thus, a total of approximately 150 m (2 × 75 m)
were used for calibration/validation purposes.
The double-ended measurements used reference sections Z1 and Z3 to calibrate the
temperature along the entire cable (Figure 4-3(a) and (c)). To report the thermal
resolution of this DTS deployment, the calibration metrics used by Hausner et al.
(2011), mean bias (MB) and root mean squared error (RMSE), were employed:
 
1
1

   i re
n
i
f MB T T
n
(4.3)
 2
1
1

   i ref
n
i
RMSE T T
n
(4.4)
where n is the number of temperature observations in the coil at a given time; Ti [°C] is
the calibrated DTS temperature at each observation in the coil; Tref [°C] is the
independently measured temperature at the same point of the coil, used as reference
known temperature. These metrics were calculated for each temperature trace (for a
total of 30039 traces collected at 5-min integration time for 104 days), resulting in a
time series for each metric at each reference section. The mean and standard deviation
of these time series are reported in Table 4-1. In the reference sections that were inside
the uniform temperature water reservoirs (Z1, Z2, Z3, Z4, Z5 and Z6), the RMSE is
smaller than 0.12 °C, while the RMSE in the other reference sections is always smaller
than 0.44 °C. Nonetheless, note that the independent measurement does not necessarily
coincide with the mean temperature along the cable section. For instance, in reference
Annex X
535
54 zones Z9 and Z10 the cable was exposed to air and thus it is probable that thermal stratification created differences between the temperatures measured by the DTS system and the independent temperatures measured by the HOBO sensors. The results presented in Table 4-1 imply that the thermal resolution is of the order of 0.1°C when the data are collected at 5-min integration time. From the data presented by Hausner et al. (2011), this is an expected thermal resolution for a DTS system under environmental conditions and calibrated using the manufacturer’s algorithm. Note also that when the integration time is increased, the thermal resolution can be improved (as described below). Appendix A presents a probabilistic frequency analysis that gives more details of the results presented in Table 4-1. MB (10-3 °C) RMSE (10-3 °C) μ ± σ μ ± σ Z1 83.8 ± 63.8 115 ± 34.2 Z2 96.4 ± 64.2 117 ± 37.6 Z3 0.14 ± 21.4 49.9 ± 16.7 Z4 -56.6 ± 58.4 85.5 ± 36.3 Z5 32.0 ± 86 84.8 ± 51.0 Z6 -72.8 ± 92 113 ± 59.2 Z7 -407 ± 118 409 ± 118 Z8 -431 ± 119 432 ± 119 Z9 146 ± 342 319 ± 228 Z10 129 ± 335 313 ± 220 Table 4-1. Calibration metrics for the Silala River DTS deployment (5-min integration time and sampling resolution of 0.25 m). : mean value; : standard deviation. To perform a more robust analysis with improved resolution, the information can be integrated using longer time periods. To account for the natural variations of the system, two sets of time records were selected, 2 hours around the maximum temperature recorded in the water during each day (4 hours total, highest radiation period, referred to as “day”), and 4 hours prior to the lowest recorded temperature in the water each day (hours with no influence of solar radiation, referred to as “night”). The calibration metrics for the 5250 temperature traces (50 traces per day, collected on 105 days) averaged over the day and the night are reported in Table 4-2, together with the metrics for the global mean temperature (104 complete days, 29701 traces). The results 536
Annex X
55
presented in Table 4-2 show that the thermal resolution (reported as the RMSE) of the
system for the day and night traces is 0.14 and 0.11 °C, respectively, in the worst case.
For the day and night traces, RMSE’s as small as 0.02 and 0.01 °C, respectively, are
observed in the reference sections that were not used to calibrate the temperatures along
the cable. Two important conclusions can be drawn from these results: (1) the thermal
resolution during the night traces improves, which is advantageous for studying riveraquifer
interactions using thermal methods, because during the day the solar radiation
may affect the measurements (e.g., due to solar radiation absorption); and (2) because
the data reported in Table 4-2 are averaged over a period of 105 days and have a
thermal resolution as small as 0.11 °C (night traces), any feature detected from these
observations comes from a physical process that occurs consistently throughout the
monitoring period. In other words, there is a very small probability that thermal features
come from short-period events or anomalies. Appendix A presents more details about
the calibration/validation metrics of the day and night traces.
MB (10-3 °C) RMSE (10-3 °C)
Mean Night Day Mean Night Day
Z1 84.1 90.2 64.9 92.2 102 72.2
Z2 96.7 107 71.6 96.7 108 71.9
Z3 0.4 13.6 -20.5 4.2 18.6 22.4
Z4 -54.3 -2.49 -103 54.4 12.2 104
Z5 34.7 88.6 -24.8 34.9 89.1 25.7
Z6 -69.9 -11.8 -140 77.1 42.5 143
Table 4-2. Calibration metrics for the Silala River DTS deployment using temperature traces
obtained near the solar noon (average of 4 h of data centered at the peak time of solar
radiation) and near the minimum observed temperature (average from 4 h before and up to the
time when the minimum temperatures were observed) and using all the temperature records of
complete days.
4.2.4 Location of the cable along the Silala River
To relate findings from temperature profiles to in-field physical processes, it is
necessary to relate the distance along the cable to its real position along the river, which
is one of the main challenges of DTS methods. In this study, different sections of the
cable that were submerged in the river were exposed to the air for enough time (11 min)
to observe them in the 5-min temperature traces that have been collected. The locations
Annex X
537
56 where the cable was exposed to air were georeferenced using a GPS (Montana 680, Garmin, USA). Additionally, other artificial and natural features (e.g., culverts, springs) were georeferenced to determine the position of the cable along the river. Figure 4-4 presents some of the temperature traces that allowed the determination of the location of the cable along the Silala River. These data were used to build the conceptual model that is presented later (Figure 4-12). Figure 4-4. Temperature traces used to locate different positions of the cable in the field. 4.2.5 Data filtering to remove thermal anomalies not related to water temperature Before investigating river-aquifer interactions using the thermal data from the river, it is important to remove the measurements that do not correspond to water temperature. As described before, many sections were exposed to the air, e.g., see Figure 4-3(c). Therefore, these thermal records must be removed before trying to detect zones of groundwater discharge into the river. To perform this filtering, the following thermal features are relevant (Stonestrom and Constantz, 2003): a) Air-exposed section: cable exposed to air is subject to daily atmospheric oscillations that results in large temporal standard deviation. These sections are subject to extreme conditions. Unshaded cable sections exposed to air during the day can reach temperatures higher than the air temperature because of radiation 538
Annex X
57
absorption (Neilson et al., 2010) and low temperatures during the night, while
shaded cable sections exposed to air exhibit low temperatures during the day and
night.
b) River waters: cable inside the river will be subject to reduced temperature
oscillations compared to the sections exposed to air, i.e., the water acts as a thermal
buffer and a low standard deviation of the temperatures is expected (Stonestrom and
Constantz, 2003). Thermal daily fluctuations are expected in these waters and in
general the temperatures are going to be higher than 0 °C unless the water freezes.
c) Groundwater: sections of the cable that measure the temperature of groundwater
discharge (upwelling) will be subject to lower thermal oscillations than those of the
river waters. This behavior occurs because ground waters generally have seasonal
fluctuations (instead of daily fluctuations), meaning that groundwater typically is
considered to have a constant temperature over the observation periods (Stonestrom
and Constantz, 2003). When groundwater has a different temperature to the river, it
is possible to infer locations of groundwater upwelling. For instance, if there is a
groundwater discharge into a river, a step change in the river thermal profile is
observed at the location of the contribution (e.g., see Selker et al., 2006 and
Westhoff et al., 2007).
Figure 4-5 shows temperature time series collected at different positions along the cable
that illustrate the difference in temperature fluctuations between an air-exposed-to-sun
section and two underwater sections. Regarding river water temperatures, the
temperatures at two important locations are shown. The first location is upstream of a
major groundwater discharge (River upstream SPW-DQN). This location is influenced
by the daily thermal oscillations of the river. The second location is downstream of Well
SPW-DQN, where it is clear how groundwater affects the amplitude of the thermal
signal.
Annex X
539
58 Figure 4-5. Temperature time series at specific points along the cable. To understand how the data filtering was performed, in Figure 4-6 we plotted the profiles of the temperature’s mean and standard deviation in the last 400 m of cable (closest to the DTS instrument). In this section, the cable is measuring the temperature of the river between distances 1570 m (upstream) and 1680 m (downstream). These data correspond to the 5-min data averaged over the 104 days of data collection. The effect of the artesian well SPW-DQN, a warm groundwater source, is clearly observed at 1645 m. From the data presented in Figure 4-6, the following thermal features can be observed:  The fiber-optic cable in the spool is inside a wooden box. In this section, the mean temperature is relatively low (~6-7 °C) but exhibits a large standard deviation (~10 °C). This section is subject to the air temperature inside the box.  The fiber-optic cable located between the ambient and the cool baths (at a distance of 1725 m) is exposed to the sun. Therefore, it has a large variability.  The cable exposed to the river water (at a distance  1628 m) exhibits a low variability.  The thermal profile at SPW-DQN shows an abrupt temperature increase that influences the temperature downstream of the well (an increase of ~6°C in the mean 540
Annex X
59
water temperature) and even lower variability (lowest standard deviation of the
trace).
The thermal features described previously allowed removal of the air temperatures from
the collected data, resulting in a thermal record of the water temperatures along the river
reach. To filter the information, we used the mean temperature profiles during day and
night and the standard deviation profile of the whole dataset. A robust locally weighted
scatterplot smoothing (LOWESS) algorithm (Cleveland, 1979) was used to estimate the
underlying trend of the data (Figure 4-7). We assumed that this trend represents the real
value of temperature and standard deviation in the river waters at each location. Any
spatial deviation from this trend that is larger than the instrument resolution (0.1 °C,
defined by the calibration metrics) was considered an anomaly. We only excluded
anomalies caused by air-exposed sections and thus only the data that fulfill the
following three requirements is kept in consideration:
1) Night temperature equal or higher than its trend.
2) Day temperature equal or lower than its trend.
3) Standard deviation equal or lower than its trend.
This filtering procedure allows consideration of groundwater effects, e.g., lower
standard deviation zones or locations with stable temperatures (typically lower during
the day and higher during the night).
Annex X
541
60 Figure 4-6. Mean (a) and standard deviation (b) profiles of the DTS data averaged over 104 days. It is important to note that errors in the measurement method increase near to the ends of the duplexed fiber-optic cable and decrease towards its middle (Tyler et al., 2009). Therefore, double-ended measurements affect the standard deviations of the temperature at different locations along the cable. Consequently, it is inadequate to use a unique standard deviation to filter the data. Therefore, we used the temperature’s standard deviation profile to filter the data. 542
Annex X
61
Figure 4-7. (a) Mean temperature profile for the day and the night and respective underlying
trend. (b) Temperature standard deviation (SD) profile and trend.
4.3 Results and discussion
The day and night mean temperature profiles shown in Figure 4-8 were used to identify
possible locations of groundwater discharge to the river. In these profiles, the sections
of the cable that were exposed to the air were removed (as described above). Three
types of possible river-aquifer interactions were identified using the following criteria:
1) Type-1 river-aquifer interaction: since groundwater in this basin has a higher
temperature than the river, a step increase in the night temperature of at least 0.1 °C,
with evident effects downstream was identified as a location of groundwater
discharge.
2) Type-2 river-aquifer interaction: a location that exhibits a temperature peak
during the night and a temperature valley during the day was identified as a location
where a river-aquifer interaction occurs. There are two plausible explanations for
this temperature anomaly: 1) if the cable is measuring the temperature at the watersediment
interface, this thermal anomaly can only be explained by a location of
groundwater upwelling. In this case, the groundwater should have a temperature that
is in between the river temperature during the day and night; and 2) if the cable is
buried and measures the temperatures of the riverbed sediments, this thermal
anomaly can be explained by a location where water exchange between the river and
Annex X
543
62 the aquifer occurs (upwelling or downwelling conditions). If the sediments’ thermal conductivity and the depth at which the cable is buried are known, a similar analysis to that performed in Chapter 3 can be used to determine the magnitude and direction of this water exchange. 3) Type-3 river-aquifer interaction: groundwater coming from the SPW-DQN artesian well, which comes from a deeper aquifer than the fluvial aquifer below the Silala River, produces a similar effect to the Type-1 groundwater discharge, but we have separated it since its effect on the thermal dynamics of the river was important during the measurement period. Figure 4-8. Mean temperature profile for the day and the night. The thermal resolution of the “day” data is always smaller than 0.14°C (mean value of 0.07 °C). The thermal resolution of the “night” data is always smaller than 0.11°C (mean value of 0.06 °C). The first case of river-aquifer interaction described previously (Type-1) could only be produced by a local source of heat present during the night since there is no direct radiation. Furthermore, the information considered for the construction of the night profile was selected from approximately 4:00 – 8:00 (4 h before the lowest water temperature in the river). Therefore, any residual heat source from the sediments should also be negligible and could not account for the increase in temperature at any of the locations that exhibit a step increase in temperature. For these reasons, these abrupt temperature increases during the night can only be explained by a warm water discharge that comes from groundwater (from springs or from the streambed). This temperature anomaly can also be seen in the day temperature trace at the same location. This type of 544
Annex X
63
groundwater input was also reported by Westhoff et al. (2007). They detected
temperature changes as large as 2 °C at specific locations in a stream, and used an
energy conservation dynamic model to demonstrate that those thermal anomalies could
only be explained by groundwater inputs. In the case of Westhoff et al. (2007), the
thermal anomalies were more clearly identified during the day because the groundwater
inputs were cooler than the stream temperature, while in the Silala River the sources are
generally warmer than surface waters and are easily identified during the night
(highlighted in green in Figure 4-9).
Figure 4-9. Night mean temperature profile where the different types of river-aquifer
interactions are highlighted.
According to the scientific literature, there are two possible explanations for the thermal
anomalies associated with the Type-2 river-aquifer interactions (Hatch et al., 2006;
Westhoff et al., 2007; Selker et al., 2006; Krause et al., 2012). The first explanation is
that the temperature anomaly is related to a localized groundwater discharge that has a
temperature that is within the range of the river temperatures (Selker et al., 2006). In
this case, the groundwater discharges are not strong enough as to influence the
temperature of the river water downstream (Krause et al., 2012). Therefore, in this
situation the cable would be located just above a localized groundwater discharge of
relatively low flow (compared to the river flow). Note also that to be able to detect this
type of river-aquifer interaction using DTS methods, the fiber-optic cable should be
Annex X
545
64 measuring the water-sediment interface. The second explanation applies when the cable is buried below the water-sediment interface (Krause et al., 2012). In this case, as depth increases, the thermal amplitude is reduced and a phase change occurs. As described in Chapter 3, the amplitude reduction and phase change are associated with the water travel time through the sediments and thus, they can be used to quantify the magnitude and direction of the water exchange between the river and the aquifer. In this case, water can be flowing from the river into the aquifer or in the opposite direction. Therefore, the Type-2 river-aquifer interaction should be carefully investigated to discriminate between these two causal effects. During a field trip carried out during January, 2017, the Type-2 sections were inspected and it was found that the cable was buried due to sediment deposition. Hence, we performed a similar analysis to that shown in Chapter 3. In this analysis, we used the Hatch et al. (2006) amplitude method to quantify the magnitude and direction of water exchange fluxes from two temperature time series obtained from the DTS measurements: one representing the thermal evolution of the water-sediment interface and the other depicting the thermal evolution of the sediments at a specific depth below the water-sediment interface (Figure 4-10(a)). This analysis was performed at two different locations of the cable identified as Type-2 river-aquifer interaction. The thermal measurements of a buried section of the cable were compared to the records of the closest cable section that was measuring the temperature of the river (Figure 4-10). Additionally, the depths to which the cable was buried and the thermal properties of the sediments at the locations identified as Type-2 river-aquifer interaction were measured in the field. These data were used to estimate the water fluxes between the river and the aquifer that explained the amplitude reduction in the temperature time series. The first buried location was at 1376 m along the cable and the water-sediment interface temperature was estimated from the DTS measurements at 1366 m. This location is ~160 m downstream TR4. The cable at 1376 m is known to be buried by the sediments at depths of ~0.15-0.20 m (determined by visual inspection in the field site), and the in-situ thermal conductivity () and heat capacity (Cp) were 0.853  0.156 W m-1 K-1 (average  standard deviation) and 2.509  0.381 MJ m-3 K-1, respectively. The resulting water fluxes at this location ranged between 7  1.4 cm day-1 (for a cable buried at 0.15 m depth) and 17  1.2 cm day-1 (for a cable buried at 0.20 m depth), flowing from the river to the aquifer. The second buried location was at 683 m along the cable and the water-sediment interface temperature was estimated from the DTS measurements at 664 m. This location is ~7 m upstream TR3. The cable at 683 m is known to be buried by the sediments at depths of ~0.03-0.06 m (measured in the field), and the in-situ  and Cp were 1.196  0.191 W m-1 K-1 and 2.498  0.033 MJ m-3 K-1, 546
Annex X
65
respectively. In this location, the resulting water fluxes ranged between 33  13 cm day-
1 (for a cable buried at 0.03 m depth) and 55  12 cm day-1 (for a cable buried at 0.06 m
depth), flowing from the river to the aquifer. Therefore, Type-2 river-aquifer
interactions that we identified are actually buried sections of cable where it is more
likely that the river-aquifer interactions correspond to locations where water flows from
the river to the aquifer, which is consistent with the TR findings (both in magnitude and
direction, as shown in Chapter 3).
Figure 4-10. Water flux estimation in the locations where Type-2 river-aquifer interactions
were identified: diagram of the hypothetical case used for the analysis (a); and temperature
time series used to determine water fluxes between the river and the aquifer at the first location
(at 1376 m) (b) and at the second location (at 683 m) (c).
The third category of river-aquifer interaction (Type-3) is a point-source groundwater
discharge that results from an artesian well discharging into the river (Figure 4-6, Figure
4-9, Figure 4-11, and Figure 4-12). Note that Figure 4-11 also shows the thermal effect
of a pump test (performed during 11/25/2016 in PW-UQN) that discharged warm
Annex X
547
66 groundwater into the river (the thermal profile of the previous day is also shown). The thermal profiles shown in Figure 4-11 highlight the capacity of DTS methods to observe the spatiotemporal thermal dynamics of these types of environments. Note also that the persistent effect downstream of the pump test is similar to that observed downstream SPW-DQN. Nonetheless, the effect of the latter is about twice the magnitude and is almost unaffected during the pumping test, which means that the water flow coming from SPW-DQN is considerably higher than that of the pump test (~20 l/s) and of the river itself (~50 l/s, as shown in Chapter 5). As described later (Chapter 5), the Silala River has many weirs along the river reach that provides flow measurements. These data can help to infer some features of the pump test and of SPW-DQN. In particular, weir 8 is downstream of the pump test discharge point and upstream of SPW-DQN, whereas weir 9 is downstream of SPW-DQN. The distance between weirs 8 and 9 is approximately 200 m and the flows measured in these weirs are presented in Chapter 5 (Figure 5-4). The previous information combined with the thermal data presented in Figure 4-11 (data summarized in Table 4–3) can be used to determine the temperature of the groundwater during the pump test and the temperature of the water that comes from SPW-DQN. Indeed, an energy balance yields a temperature of ~19.5 °C for the groundwater coming from the pump test. This temperature is in agreement with the thermal profiling performed after drilling the well PW-UQN, which showed temperature of ~19.2 °C for depths deeper than 30 m (Arcadis, 2017). On the other hand, considering that the difference in flow rates measured between weirs 8 and 9 is mainly due to the water input from SPW-DQN, an energy balance results in temperatures of ~20 °C for the groundwater. 548
Annex X
67
Date and
time
Flow weir
8 (l/s)
Flow weir
9 (l/s)
Temperature upstream
SPW-DQN (°C)
Temperature downstream
SPW-DQN (°C)
11/24 07:58 41.8 128.4 8.16 16.16
11/25 08:00 63.3 150.8 9.35 16.40
Table 4-3. Information used to estimate groundwater temperatures.
Figure 4-11. Temperature profile during a pumping test and the day before.
Finally, Figure 4-12 shows a conceptual model that indicates the locations of
groundwater discharge that were identified using DTS methods. These locations are
presented as the distance along the fiber-optic cable, but also are depicted with
geographical locations and physical references (this is the reason why the DTS
temperature profile is inverted). Green circles represent a consistent increase in
temperature with effects downstream (Type-1 river-aquifer interaction). The circles in
magenta represent locations with an evident trend to a stable temperature when
comparing day and night temperatures, and a considerably higher temperature during
night, but with no apparent effect on temperatures downstream (Type-2 river-aquifer
interaction). The orange circle represents a relevant and consistent increase in
Annex X
549
68 temperature that affects downstream conditions (Type-3 river-aquifer interaction). This increase in the river temperature is caused by the artesian well SPW-DQN. The groundwater discharge locations that were identified using DTS methods are coherent with the field observations of springs, as shown in the photograph of Figure 4-12(a) and in the photographs shown in Appendix A. Typically, these springs discharge water from the ravine walls at elevations higher than the river stage (see Figure 4-12 (a)). The location of the springs that have enough thermal mass to change the river temperature by more than ~0.1°C is also indicated in Figure 4-12 (this is the reason why not all the springs that have been documented in the current studies are depicted in this conceptual model). Figure 4-12 also presents the locations of the temperature rods (TR’s), described in Chapter 3, which shows point-in-space estimations of water discharging from the river into the aquifer through the riverbed. 550
Annex X
69
Figure 4-12. Conceptual model of the locations of main groundwater contributions detected
using DTS methods. (a) Photograph depicting SP-FCAB15 entering the Silala River. (b)
Photograph depicting a zone where the DTS cable was buried.
Annex X
551
70 4.4 Conclusions In this study, DTS methods were used to investigate river-aquifer interactions along a ~1.3 km reach of the Silala River. The river reach began at ~750 m from the Chile-Bolivia international border and finished ~2 km downstream of the border. Along this reach, temperature was measured with a spatial resolution of 0.5 m (sampling resolution of 0.25 m) and using an integration time of 5 min. When the data were integrated to 5 min, the thermal resolution of the DTS system was 0.12 °C. The collected temperature data were separated into mean profiles taken during the day and the night. The mean day traces corresponded to thermal data averaged during the 4 h of maximum irradiance of each day, while the mean night traces were data averaged during the 4 h before reaching the minimum temperature along the river. The mean day and night traces were averaged over a time period of ~104 days and resulted in a thermal resolution of 0.14 and 0.11 °C, respectively. Therefore, the thermal resolution during the night traces improves, which is advantageous for studying river-aquifer interactions using thermal methods because during the day the solar radiation may affect the measurements (e.g., due to solar radiation absorption). These results also imply that any thermal feature larger than 0.11 °C detected from the night traces comes from a physical process that occurs consistently throughout the monitoring period (~104 days). Hence, the identified thermal features are indicative of groundwater entering the river. Several locations of groundwater discharge were identified, but the most distinct groundwater input was the Type-3 river-aquifer interaction. This warm discharge corresponds to SPW-DQN, which heavily influences the thermal behavior of the river downstream, producing an increase of ~5 °C in the mean water temperature (during the entire day) and up to ~9 °C at specific times (during the night). Our results show that SPW-DQN greatly affects the thermal dynamics of the river. The Type-1 river-aquifer interaction was the second most important groundwater discharge at the study site. The thermal anomalies caused by the Type-1 inputs are clearly present during the whole measuring period and were consistent in their locations, increasing the downstream temperature of the river. These locations are highly correlated with the positions of the springs that discharge from the walls of the ravine at elevations higher than the river stage, as shown in the photographs. Finally, from field observations, Type-2 river-aquifer interactions are more likely to correspond to locations where the DTS cable is buried. Our results suggest that at these points the water discharges from the river to the aquifer, consistent with the findings using the TR’s. 552
Annex X
71
5 ANALYSIS OF WATER FLOWS AT THE DIFFERENT WEIRS ALONG
THE SILALA RIVER
5.1 Introduction
In this Chapter, the water flow measured at different locations along the Silala River is
reported. The control sections are weirs 4, 5, 6, 7, 8 and 9, and a site within reach A,
which is one of the river reaches that were used to investigate the fluvial processes of
the Silala River (Mao, 2017). The location of the sections where the river flow was
measured is shown in Figure 5-1, which also presents the position of the temperature
rods (TR’s) and the layout of the DTS cable (blue line). The objective of this section is
to analyze if the conceptual model presented in Figure 4-12 agrees with the data
measured at the sections in the Silala River where the water flow is being measured.
Annex X
553
72 Figure 5-1. Image of the study site that shows the location where the water flow is measured (weirs 4, 5, 6, 7, 8 and 9, and flow gauge at a site within reach A). The locations of the temperature rods (TR1-TR5) are also depicted. 5.2 Materials and methods 5.2.1 Weirs and stage-discharge curves The Antofagasta (Chili) and Bolivia Railway Company Ltd. (FCAB) has several trapezoidal weirs installed along the Silala River. These weirs were used to determine the flow rate in different reaches of the Silala River. Figure 5-2 shows photographs of the weirs used in this analysis. The inset in Figure 5-2(c) shows the geometrical properties of the weirs. 554
Annex X
73
Figure 5-2. Photograph of the weirs used to estimate the flowrates in the Silala River: (a) weir
4; (b) weir 8; and (c) weir 9. The geometrical properties of the weirs are depicted in the inset of
subfigure (c).
To determine the flowrates at the weirs, the following stage-discharge curve was used:
𝑄𝑄 = 1000(1.32 tan 𝛼𝛼 ℎ2.47 + 1.691 𝑏𝑏1.02 ℎ1.47) (5.1)
where Q [L s-1] is the flowrate, h [m] is the stage, tan  = (L - b)/2h0 [-], and L [m], b
[m] and h0 [m] are the geometrical dimensions of the weirs (see Figure 5-2(c) and Table
5-1). The stage-discharge curve presented in equation (5.1) and the geometrical values
shown in Table 5-1 were provided by FCAB.
Annex X
555
74 Dimension Units Weir 4 Weir 8 Weir 9 L m 0.700 0.700 0.700 b m 0.500 0.500 0.502 h0 m 0.500 0.495 0.500 tan() - 0.200 0.202 0.198 Table 5-1. Geometrical information related to the weirs located along the study site. Additionally, at a specific site within reach A (Figure 5-1) the water stage was monitored with the aim of investigating fluvial processes in the Silala River (Mao, 2017). As in this reach a stage-discharge curve was also constructed and the water stage was monitored, we also used this site to analyze the water flow at this location. The details of the construction of this stage-discharge curve are presented elsewhere (Mao, 2017). 5.2.2 Water level monitoring at the weirs At weir 4, which is located near the Chile-Bolivia international border, two pressure transducers were installed (HOBO Water Level (4 m) Data Logger, Onset Computer Corporation, Bourne, MA). These instruments have a typical error of 4 mm, a maximum error of 8 mm, and a resolution of 1.4 mm. One of the pressure transducers measured the absolute pressure of the water column and the other measured atmospheric pressure for barometric compensation. Barometric compensation was performed using the HOBOware Pro software (Onset Computer Corporation, Bourne, MA). Weirs 8 and 9 also were instrumented with pressure transducers (HOBO Water Level (4 m) Data Logger, Onset Computer Corporation, Bourne, MA) to measure the absolute pressure of the water column. The barometric compensation for these transducers was achieved by the use of a pressure transducer located at a meteorological station located at ~150 m from weir 8 and at ~60 m from weir 9 (Suárez, Muñoz et al., 2017). The data from these pressure transducers were collected at time intervals of 15 min. Every time prior to data retrieval, the water level at the weir measurement location was recorded to confirm that the data from the pressure transducers were correct. Then, equation (5.1) was used to estimate the water flow time series at each control section (weirs or at reach A). These flowrates were computed for the dates between November 24th, 2016 and January 17th, 2017. 556
Annex X
75
5.2.3 Anthropogenic water use in the study site
FCAB takes water from their intake that is located between weirs 4 and 5, near the
Chile-Bolivia international border (Figure 5-1). Historical data provided by FCAB
shows that between June 2010 and September 2016, a flow of 125.8  9.4 l/s (average 
standard deviation) was collected at the FCAB intake (Figure 5-3).
Figure 5-3. Flow diverted at the FCAB intake between June 2010 and September 2016.
5.3 Results
The flowrates measured at weirs 4, 8 and 9, and at reach A are shown in Figure 5-4. In
general, the flowrates measured at each control section display a periodic behavior with
a daily time scale. An increase in the flowrates at weirs 8 and 9 during the end of
November and early December is related to pump tests that discharged water into the
river. The large increase in flowrates observed in January 10th, 2017 was due to a
sediment transport experiment where FCAB did not divert waters into the FCAB intake
(Mao, 2017). Thus, an additional water flow of ~120 l/s traveled through the river.
Annex X
557
76 Figure 5-4. Flowrates measured at weirs 4, 8 and 9, and at reach A between November 24th, 2016 and January 17th, 2017. The data displayed in Figure 5-4 can be used to quantify the net water gains or losses that occur due to river-aquifer interactions. Indeed, at weir 4 a flowrate of 140.4  8.7 l/s (average  standard deviation) was observed, while at reach A the flowrate was of 22.1  5.1 l/s. Thus, between these locations there is a net water loss of 118.3  10.1 l/s. Because the FCAB intake collects ~125.8 l/s, there is a net gain of 7.5 l/s (8.9 l/s coming from the springs that surround this reach and a water loss of 1.4 l/s towards the aquifer through the riverbed). In this river reach, the flow of one spring is monitored in weir 7 (Figure 5-1). Historical data of water flows at weir 7 (from May 1993 to December 2016), which were provided by FCAB, show a flow of 2.3  0.2 l/s. These results are in agreement with the identification of groundwater inputs using DTS methods that showed many locations of groundwater discharge, i.e., springs discharging into the river, between the beginning of the cable and reach A (as shown in Figure 4-12). Nonetheless, it is important to emphasize that even when there is a net gain of water towards the river, there are zones where the river discharges waters into the aquifer, i.e., losing sections, such as those observed at the TR’s. As shown in Figure 5-4, the reach between reach A and weir 8 also exhibits a net water gain of 25.5  9.4 l/s (27.0 l/s coming from the springs that surround this reach and a water loss of 1.5 l/s towards the aquifer through the riverbed). In the upper part of this reach there are many springs that discharge groundwater into the river (e.g., see Figure 4-9). In terms of the results obtained with the DTS system, as shown in Figure 4-12, in this river reach there are also many locations where warm groundwater inputs (Type-1) influence the thermal dynamics of the river. Nonetheless, there are also zones where the river discharges water into the aquifer, such as near TR4, where the groundwater level 558
Annex X
77
was deeper than 1.0 m below the water-sediment interface. These results highlight the
natural variability that is observed in the river.
Finally, in the river reach between weirs 8 and 9 there is an important net water gain
that is dominated by the influence of well SPW-DQN (Type-3 river-aquifer interaction).
Between these two weirs, a net water gain of 91.2  6.6 l/s was estimated (91.6 l/s
coming from well SPW-DQN and a water loss of 0.4 l/s towards the aquifer through the
riverbed). The effect of SPW-DQN on the thermal dynamics of the river is very clear,
increasing the water temperature and reducing the amplitude of the thermal signal (see
Figure 3-14(a) and Figure 4-6(a)). However, it is important to recall that in this reach
there are also river waters discharging to the aquifer, as shown in the analysis of TR5
(Figure 3-15).
Using the average water flux that drained towards the aquifer as estimated by the TR’s,
and assuming a river with a mean width of ~1 m and a river reach of ~2 km, the water
flow discharged from the river into the aquifer is 3.3  2.7 l/s. This water flow is much
smaller than the net water gains determined from the weir data. An approximate
calculation using the data from the weirs shows that if FCAB does not divert water to
their intake, the study reach should gain ~124 l/s. i.e., the water loss through the
streambed is ~3% of the water gains. But performing the same analysis using the
maximum water fluxes estimated in the TR’s, the water loss through the streambed
would be ~8% of the water gained along the study reach.
5.4 Conclusions
The water flow monitored at the different weirs that are located along the study site
supports the conceptual model of river-aquifer interactions (Figure 4-12). Between the
Chile-Bolivia international border and reach A, a net gain of ~7.5 l/s was estimated,
while a net gain of ~25.5 l/s was estimated between reach A and weir 8. On the other
hand, between weirs 8 and 9, a localized warm groundwater discharge (SPW-DQN)
results in a net water gain of ~91.2 l/s (where 91.6 l/s comes from well SPW-DQN and
0.4 l/s flow towards the aquifer through the riverbed). As described in the previous
Chapters, along these reaches there are also sections of the river in which water is
discharged into the aquifer (TR1-TR5), although these discharges are much smaller than
the water gains from the springs. These results highlight the variability and the
heterogeneity of the surface/groundwater interactions in this natural hydrogeological
system.
Annex X
559
78 6 CONCLUSIONS In this study, the interactions between the Silala River and its fluvial aquifer were investigated using thermal methods and data collected from weirs. Two thermal methods were used: (1) using temperature time series of the sediments beneath the rivers to find water fluxes across the water-sediment interface; and (2) using DTS methods to identify thermal anomalies that could be due to groundwater discharge into the river. Our results show that it is likely that at the five locations where the temperature of the sediments beneath the river was monitored, water flows from the river into a shallow aquifer composed of the alluvial sediments lying beneath the bed/river water interface. The estimated mean water fluxes varied between -9 and 46 cm day-1 (when using the average thermal properties of the sediments). In all of these locations, the variability was very small and ranged between 6 and 21 cm day-1. Using the average water flux that drained to the aquifer, and assuming a river with a mean width of ~1 m and a river reach of ~2 km, the water flow discharged from the river into the aquifer would be ~3.3 l/s. This water flow is much smaller than the net water gains from spring inflows determined from the weir data along the study sections. In fact, an approximate calculation using the data from the weirs shows that if FCAB does not divert water to their intake, the study reach would gain ~124 l/s. Therefore, the water loss, in the case, through the streambed would be ~3% of the net water gain. When using the maximum water fluxes estimated in our study, the water loss through the streambed would be ~8% of the water gained along the study reach. The DTS approach allowed identification of several locations of groundwater discharge to the river along the river reach. The most distinct groundwater input corresponded to the SPW-DQN well, which heavily influences the thermal behavior of the river downstream, producing an increase of ~5 °C in the mean water temperature (during the entire day) and up to ~9 °C at specific times (during the night). Our results show that SPW-DQN affects greatly the thermal dynamics of the river. The other important locations of groundwater discharge that were identified with DTS methods correspond to the springs that discharge warm water from the walls of the ravine at elevations higher than the river stage. The thermal anomalies identified using DTS methods were spatially correlated to the location where these springs entered the Silala River. This information is consistent with the net water gains estimated with the weir data. From our results we can conclude that three types of river-aquifer interactions occur at the Silala River. The first one corresponds to the water exchange that goes from the river into the aquifer through the streambed. This water loss is of the order of 3.3 l/s along the study reach. The second type of river-aquifer interaction corresponds to 560
Annex X
79
springs that discharge water from the walls of the ravine at elevations higher than the
river stage. From the analysis of the weir data, a water gain of the order of 35.9 l/s could
be due to these springs along the study reach. The third type of interaction corresponds
to the artesian well SPW-DQN discharging 91.6 l/s of warm waters into the Silala
River.
The data collected in this study reveals that river-aquifer interactions in the Silala River
are undeniable and that water gains in the river occur due to water discharged through
the springs located all along the study zone. The SPW-DQN well corresponds to a
significant groundwater discharge from a deeper groundwater source that not only
increases significantly the water flows, but also increases the temperature of the river
downstream. The findings of this study support the hypothesis that the Silala River is a
perennial river supported by groundwater.
Annex X
561
80 7 REFERENCES Anderson, M.P., 2005. Heat as a ground water tracer. Groundwater, 43(6), 951-968. Anibas, C., Fleckenstein, J.H., Volze, N., Buis, K., Verhoeven, R., Meire, P., Batelaan, O., 2009. Transient or steady-state? Using vertical temperature profiles to quantify groundwater-surface water exchange. Hydrological Processes, 23(15), 2165–2177, doi:10.1002/hyp.7289. Arcadis, 2017. Detailed Hydrogeological Study of the Silala River. (Vol. 4, Annex II). Briggs, M.A., Lautz, L.K., McKenzie, J.M., Gordon, R.P., Hare, D.K., 2012. Using high-resolution distributed temperature sensing to quantify spatial and temporal variability in vertical hyporheic flux. Water Resources Research, 48, W02527, doi:10.1029/2011WR011227. Briggs, M.A., Lautz, L.K., Buckley, S.F., Lane, J.W., 2014. Practical limitations on the use of diurnal temperature signals to quantify groundwater upwelling. Journal of Hydrology, 519, 1739-1751. Briggs, M.A., Buckley, S.F., Bagtzoglou, A.C., Werkema, D.D., Lane, J.W., 2016. Actively heated high-resolution fiber-optic-distributed temperature sensing to quantify streambed flow dynamics in zones of strong groundwater upwelling. Water Resources Research, 52, 5179–5194, doi:10.1002/2015WR018219. Bristow, K.L., Kluitenberg, G.J., Horton, R., 1994. Measurement of soil thermal properties with a dual-probe heat-pulse technique. Soil Science Society of America Journal, 58(5), 1288-1294. Brunner, P., Cook, P. G., Simmons, C. T., 2009. Hydrogeologic controls on disconnection between surface water and groundwater. Water Resources Research, 45(1), W01422. doi: 10.1029/2008WR006953. Buffington, J. M., Tonina, D., 2009. Hyporheic exchange in mountain rivers II: Effects of channel morphology on mechanics, scales, and rates of exchange. Geography Compass, 3(3), 1038-1062. Cleveland, W., 1979. Robust locally weighted regression and smoothing scatterplots. Journal of the American Statistical Association, 74(368), 829-836. doi:10.2307/2286407. Constantz, J., 2008. Heat as a tracer to determine streambed water exchanges. Water Resources Research, 44, W00D10. doi:10.1029/2008WR006996. 562
Annex X
81
Dakin, J.P., Pratt, D.J., Bibby, G.W., Ross, J.N., 1985. Distributed optical fiber raman
temperature sensor using a semiconductor light source and detector. Electronics Letters,
21(13), 569-570.
Darcy, H., 1856. Les Fontaines Publiques de la Ville de Dijon, Dalmont, Paris.
Gordon, R.P., Lautz, L.K., Briggs, M.A., McKenzie, J.M., 2012. Automated calculation
of vertical pore-water flux from field temperature time series using the VFLUX method
and computer program. Journal of Hydrology, 420-421, 142-158.
Hatch, C.E., Fisher, A.T., Revenaugh, J.S., Constantz, J., Ruehl, C., 2006. Quantifying
surface water–groundwater interactions using time series analysis of streambed thermal
records: method development. Water Resources Research, 42, W10410,
doi:10.1029/2005WR004787.
Hausner, M.B., Suárez, F., Glander, K.E., van de Giesen, N., Selker, J.S., Tyler, S.W.,
2011. Calibrating single-ended fiber-optic Raman spectra distributed temperature
sensing data. Sensors, 11(11), 10859-10879. doi:10.3390/s111110859.
Healey, R.W., Ronan, A.D., 1996. Documentation of computer program VS2DH for
simulation of energy transport in variably saturated porous media: modification of the
U.S. Geological Survey’s computer program VS2DT. U.S. Geol. Surv. Water Resour.
Invest. Rep., 96–4230.
Irvine, D.J., Briggs, M.A., Lautz, L.K., Gordon, R.P., McKenzie, J.M., Cartwright, I.,
2016. Using diurnal temperature signals to infer vertical groundwater-surface water
exchange. Groundwater, 55(1), 10-26. doi:10.1111/gwat.12459
Jury, W.A., Horton, R., 2004. Soil Physics. John Willey & Sons. New Jersey.
Keery, J., Binley, A., Crook, N., Smith, J.W.N., 2007. Temporal and spatial variability
of groundwater–surface water fluxes: Development and application of an analytical
method using temperature time series. Journal of Hydrology, 336, 1–16.
Krause, S., Blume, T. and Cassidy, N. J., 2012. Investigating patterns and controls of
groundwater up-welling in a lowland river by combining fiber-optic distributed
temperature sensing with observations of vertical hydraulic gradients. Hydrology and
Earth System Sciences, 16(6), 1775–1792. doi: 10.5194/hess-16-1775-2012.
Lapham, W.W., 1989. Use of temperature profiles beneath streams to determine rates of
vertical ground-water flow and vertical hydraulic conductivity. U.S. Geological Survey
Water Supply Paper, 2337.
Annex X
563
82 Lautz, L.K., 2010. Impacts of nonideal field conditions on vertical water velocity estimates from streambed temperature time series. Water Resources Research, 46, W01509, doi:10.1029/2009WR007917. Luce, C.H., Tonina, D., Gariglio, F., Applebee, R., 2013. Solutions for the diurnally forced advection–diffusion equation to estimate bulk fluid velocity and diffusivity in streambeds from temperature time series. Water Resources Research, 49, 488–506. doi:10.1029/2012WR012380. Lundquist, J.D., Cayan, D.R., 2002. Seasonal and spatial patterns in diurnal cycles in streamflow in the western United States. Journal of Hydrometeorology, 3, 591-603. Mamer, E.A., Lowry, C.S., 2013. Locating and quantifying spatially distributed groundwater/surface water interactions using temperature signals with paired fiber-optic cables. Water Resources Research, 49, 7670–7680, doi:10.1002/2013WR014235. Mao, L., 2017. Fluvial Geomorphology of the Silala River, Second Region, Chile. (Vol. 5, Annex V). McCallum, A.M., Andersen, M.S., Rau, G.C., Acworth, R.I., 2012. A 1-D analytical method for estimating surface water–groundwater interactions and effective thermal diffusivity using temperature time series. Water Resources Research, 48, W11532, doi:10.1029/2012WR012007. Muñoz, J.F., Suárez, F., Fernández, B., Maass T., 2017. Hydrology of the Silala River Basin. (Vol. 5, Annex VII). Naranjo, R.C., Turcotte, R., 2015. A new temperature profiling probe for investigating groundwater-surface water interaction. Water Resources Research, 51, 7790–7797, doi:10.1002/2015WR017574. Neilson, B.T., Hatch, C.E., Ban, H., Tyler, S.W., 2010. Solar radiative heating of fiber-optic cables used to monitor temperatures in water. Water Resources Research, 46, W08540, doi: 10.1029/2009WR008354. Rawls, W.J., Brakensiek, D.L., Saxton, K.E., 1982. Estimation of Soil Water Properties. Transactions of the ASAE, 25(5), 1316-1320. Rogers, A., 1999. Distributed optical-fibre sensing. Measurement Science and Technology, 10(8), R75-R99. Selker, J.S., Thevenaz, L., Huwald, H., Mallet, A., Luxemburg, W., van de Giesen, N.C., Stejskal, M., Zeman, J., Westhoff, M., Parlange, M.B., 2006. Distributed fiber-optic temperature sensing for hydrologic systems. Water Resources Research, 42, W12202, doi:10.1029/2006WR005326. 564
Annex X
83
Shanafield, M., Hatch, C., Pohll, G., 2011. Uncertainty in thermal time series analysis
estimates of streambed water flux. Water Resources Research, 47, W03504,
doi:10.1029/2010WR009574.
Stallman, R.W., 1965. Steady one-dimensional fluid flow in a semi-infinite porous
medium with sinusoidal surface temperature. Journal of Geophysical Research, 70(12),
2821– 2827.
Stonestrom, D.A., Constantz, J., 2003. Heat as a Tool for Studying the Movement of
Ground Water near Streams. Circular 1260 U.S. Geological Survey. Available at:
https://pubs.usgs.gov/circ/2003/circ1260/.
Suárez, F., Hausner, M.B., Dozier, J., Selker, J.S., Tyler, S.W., 2011a. Heat transfer in
the environment: development and use of fiber-optic distributed temperature sensing.
dos Santos Bernardes, M.A. (Ed.), Developments in Heat Transfer, 611-636, InTech,
Rijeka, Croatia.
Suárez, F., Aravena, J.E., Hausner, M.B., Childress, A.E., Tyler, S.W., 2011b.
Assessment of a vertical high-resolution distributed-temperature-sensing system in a
shallow thermohaline environment. Hydrology and Earth System Sciences, 15(3), 1081-
1093. doi:10.5194/hess-15-1081-2011.
Suárez, F., Muñoz, J.F., Maass, T., Mendoza, M., 2017. Evapotranspiration Estimation
in the Silala River Basin - Methods Review and Estimation of Wetland Evaporation.
(Vol. 5, Annex IX).
Tonina, D., Buffington, J.M., 2009. Hyporheic exchange in mountain rivers I:
mechanics and environmental effects. Geography Compass, 3(3), 1063-1086.
Tyler, S.W., Selker, J.S., Hausner, M.B., Hatch, C.E., Torgersen, T., Thodal, C.E.,
Schladow, S.G., 2009. Environmental temperature sensing using Raman spectra DTS
fiber-optic methods. Water Resources Research, 45, W00D23,
doi:10.1029/2008WR007052.
van de Giesen, N., Steele-Dunne, S.C., Jansen, J., Hoes, O., Hausner, M.B., Tyler, S.,
Selker, J., 2012. Double-ended calibration of fiber-optic Raman spectra distributed
temperature sensing data. Sensors, 12(5), 5471-5485.
Vandersteen, G., Schneidewind, U., Anibas, C., Schmidt, C., Seuntjens, P., Batelaan,
O., 2015. Determining groundwater-surface water exchange from temperature-time
series: Combining a local polynomial method with a maximum likelihood estimator.
Water Resources Research, 51, 922–939, doi:10.1002/2014WR015994.
Annex X
565
84 Vogt, T., Schneider, P., Hahn-Woernle, L., & Cirpka, O.A., 2010. Estimation of seepage rates in a losing stream by means of fiber-optic high-resolution vertical temperature profiling. Journal of Hydrology, 380(1), 154-164. Westhoff, M.C., Savenije, H.H.G., Luxemburg, W.M.J., Stelling, G.S., van de Giesen, N.C., Selker, J.S., Pfister, L., Uhlenbrook, S., 2007. A distributed stream temperature model using high resolution temperature observations. Hydrology and Earth System Sciences, 11(4), 1469-1480. Westhoff, M.C., Gooseff, M.N., Bogaard, T.A., & Savenije, H.H.G., 2011. Quantifying hyporheic exchange at high spatial resolution using natural temperature variations along a first order stream. Water Resources Research, 47(10), W10508. doi:10.1029/2010WR009767. Wroblicky, G. J., Campana, M. E., Valett, H. M., Dahm, C. N., 1998. Seasonal variation in surface-subsurface water exchange and lateral hyporheic area of two stream-aquifer systems. Water Resources Research, 34(3), 317-328. Williamson, D. F., Parker, R. A., Kendrick, J. S., 1989. The box plot: A simple visual method to interpret data. Annals of Internal Medicine, 110(11), 916-921. 566
Annex X
1
APPENDIX A
The following figures show the minimum precision of the TR’s temperature sensors that
enable to quantify correctly the water exchange between the river and the aquifer
(Gordon et al., 2012).
Figure A. 1. Amplitude of the thermal observations at different depths (in meters) and thermal
resolution (°C) required by the sensors to estimate correctly the water fluxes at TR1 using the
Hatch et al. (2006) amplitude method.
Figure A. 2. Amplitude of the thermal observations at different depths (in meters) and thermal
Annex X Appendix A
567
2 resolution (°C) required by the sensors to estimate correctly the water fluxes at TR2 using the Hatch et al. (2006) amplitude method. Figure A. 3. Amplitude of the thermal observations at different depths (in meters) and thermal resolution (°C) required by the sensors to estimate correctly the water fluxes at TR3 using the Hatch et al. (2006) amplitude method. Figure A. 4. Amplitude of the thermal observations at different depths (in meters) and thermal resolution (°C) required by the sensors to estimate correctly the water fluxes at TR4 using the Hatch et al. (2006) amplitude method. 568
Annex X Appendix A
3
Figure A. 5. Amplitude of the thermal observations at different depths (in meters) and thermal
resolution (°C) required by the sensors to estimate correctly the water fluxes at TR5 using the
Hatch et al. (2006) amplitude method.
The following figures show the calibration metrics obtained in the DTS system.
Figure A. 6. Temporal distribution of the mean bias (MB).
Annex X Appendix A
569
4 Figure A. 7. Temporal distribution of the root mean squared error (RMSE). Figure A. 8. Empirical distribution of the RMSE (Weibull). 570
Annex X Appendix A
5
Temperatures measured in the reference sections of the fiber-optic cable and calibration
metrics for the day and night temperature traces.
Figure A. 9. Temperature in the reference sections used to calibrate and validate the mean
temperature profiles for day and night and global mean profile.
Annex X Appendix A
571
6 Photographs: Figure A. 10. Photograph depicting spring SP-FCAB15 entering the Silala River. 572
Annex X Appendix A
7
Figure A. 11. Photograph depicting spring SP-FCAB21 entering the Silala River. The
photograph also shows the TR3 and the DTS fiber-optic cable.
Annex X Appendix A
573
8 Figure A. 12. Photograph depicting spring SP-FCAB19 entering the Silala River. 574
Annex X Appendix A
9
Figure A. 13. Photograph depicting a zone where the DTS cable was buried (Type-2 riveraquifer
interaction).
Annex X Appendix A
575
576
Annex X
Data CD
CD-ROM containing supporting data to Annexes I–X
577