INTERNATIONAL COURT OF JUSTICE
DISPUTE CONCERNING THE CONSTRUCTION OF A ROAD IN
COSTA RICA ALONG THE SAN JUAN RIVER
NICARAGUA v. COSTA RICA
REJOINDER OF COSTA RICA
VOLUME I
2 FEBRUARY 2015 INTERNATIONAL COURT OF JUSTICE
DISPUTE CONCERNING THE CONSTRUCTION OF A ROAD IN
COSTA RICA ALONG THE SAN JUAN RIVER
NICARAGUA v. COSTA RICA
REJOINDER OF COSTA RICA
VOLUME I
2 FEBRUARY 2015 TABLE OF CONTENTS
Chapter 1
Introduction
Overview of the matters for response . . . . . . . . . . . . . . . . . . . . . . . . . 1 . . .
The Structure of this Rejoinder . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. . . . . .
Chapter 2
The Absence of Adverse Impact of the Road on the San Juan River
A . Nicaragua’s Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 . . . . . . . . .
B . The Contribution of Sediment from the Road to the River in its
Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 . . . . . . . . . . . .
(1) Impact of the Road on the sediment load of the San Juan River:
before and after construction of the Road . . . . . . . . . . . . . . . .18 . . . . .
(i) Establishing the baseline . . . . . . . . . . . . . . . . . . . . .18 . . . . . . . . .
(ii) Establishing the sediment load post-construction of the road . . . 21
(iii) Nicaragua’s refusal to participate in a joint measurement
exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25. . . . . . . . . . . . . . .
(2) Estimates of sediment eroded from the Road to the River . . . . . . . . . 29
(i) UCR Report: estimates of erosion rates . . . . . . . . . . . . . . 32. . . . .
(ii) Mende Report: measurement of areas subject to erosion . . . . . . 37
(iii) Application of erosion rates to areas subject to erosion . . . . . 39.
(3) Impact of the sediment eroded from the Road on the total
sediment load of the River . . . . . . . . . . . . . . . . . . . . . . . 45. . . . . . . . . .
(4) Impact of the sediment eroded from the Road on the bed in the
Lower San Juan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47. . . . . . . . . . . .
(5) Potential impact of rainfall from a hurricane or tropical storm . . . . . 52
(6) The Road has had no adverse impact on sediment in the River . . . . . 55
C . There is No Risk ofAny OtherAdverse Impact on the San Juan
River . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56 . . . . . . . . . . . . .
(1) Water quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57. . . . . . . . . . . . .
(2) Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57 . . . . . . . . . . . .
(3) Navigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63 . . . . . . . . . . . . .
iii (4) Ecosystem, Tourism and Health . . . . . . . . . . . . . . . . . . . . . .64 . . . . . . .
(5) Remediation of the Road . . . . . . . . . . . . . . . . . . . . . . . . 71. . . . . . . . . .
D . The “Judgment” of the CACJ Should be given No Weight . . . . . . . . 75
E . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75. . . . . . . . . . . .
Chapter 3
Residual Legal Issues
A . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79. . . . . . . . . . . .
B . Nicaragua’s insistence that there has been a breach of the 1858
Treaty of Limits and of its territorial sovereignty . . . . . . . . . . . . . .81
C . There is no obligation to notify the construction of the Border
Road under the 1858 Treaty of Limits or by reason of the Court’s
2009 Judgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .87 . . . . . . . . . .
D . EIAin the Context of an Emergency . . . . . . . . . . . . . . . . . . . . . . .89 . .
(1) The threshold required for an EIA . . . . . . . . . . . . . . . . . . . 90. . . . . . . .
(2) Emergency as an exception to the international obligation to
produce an EIA . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 . . . . . . . . . . . . . .
(3) The existence of a situation of emergency in Costa Rica by
reason of Nicaragua’s actions . . . . . . . . . . . . . . . . . . . .101. . . . . . . . .
(4) The existence of an alternative assessment . . . . . . . . . . . . . . 107 . . . .
E. Notification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110. . . . . . . . . . . .
F . Alleged Breaches of other Treaties . . . . . . . . . . . . . . . . . . . . . .112. . . .
G . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 . . . . . . . . . .
Chapter 4
Remedies
A . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119. . . . . . . . . . .
B . Nicaragua’s claim for re-establishment of the status quo ante . . . . 122
C . Nicaragua’s claims for cessation / guarantees and assurances
of non-repetition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 . . . . . . . . .
D . Late Nicaraguan request for the appointment of an expert by
the Court . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 . . . . . . . . . . .
E . Costa Rican position with regard to the order by the Court
rejecting the provisional measures requested by Nicaragua . . . . . . 128
iv F . Groundless requests for declaratory relief . . . . . . . . . . . . . . . . . . . . 131
(1) Alleged breaches by Costa Rica . . . . . . . . . . . . . . . . . . . . . . .132. . . .
(2) Production of a transboundary EIA . . . . . . . . . . . . . . . . . . . . . .132. . .
(3) Transport of hazardous material . . . . . . . . . . . . . . . . . . . . . . 134 . . . .
(4) Dredging of the San Juan River . . . . . . . . . . . . . . . . . . . . . . .134. . . . .
(5) Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .136. . . . . . . . . . .
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 . . . . . . . . .
Submissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 . . . . . . . . .
Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 . . . . . . . . .
Certification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 . . . . . . . . . .
List of annexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 . . . . . . . . .
v Chapter 1
Introduction
Overview of the matters for response
1.1. In its Reply, Nicaragua has persisted in its attempt to portray the
construction of the 1856 Road (“the Road”) in Costa Rica´s sovereign territory as
a misconceived mega-project, leading to an environmental catastrophe. Thus the
1
Road is depicted as “a project of immense proportions”, by which “Costa Rica
2
has laid waste to a vast stretch of the border area for no comprehensible reason”,
“to the great detriment of Nicaragua and the environment”. 3
1.2. It is for Nicaragua, as Claimant, to make good these allegations of fact
which are central to its claim and occupy the foreground of this dispute . By
contrast, the issues that divide the Parties so far as concerns the applicable
principles of international law on transboundary harm are relatively confined. The
focus of this Rejoinder is thus largely on completing the evidence before the
Court, and responding to the case on the facts as now put by Nicaragua.
1.3. As to that case, Costa Rica makesfive introductory observations.
1.4. First, given that the Road is being constructed solely within Costa Rican
territory, the central issue is whether any sediment reaching the San Juan River
from the construction works has resulted in significant harm to Nicaragua.
4
Despite Nicaragua’s colourful references to large quantities of dump- trucks, the
1
NR, para. 1.5.
2 NR, para. 1.13.
3 NR, para. 1.12.
4
E.g. NR, para. 1.18. Chapter 1
Introduction
Overview of the matters for response
1.1. In its Reply, Nicaragua has persisted in its attempt to portray the
construction of the 1856 Road (“the Road”) in Costa Rica´s sovereign territory as
a misconceived mega-project, leading to an environmental catastrophe. Thus the
1
Road is depicted as “a project of immense proportions”, by which “Costa Rica
2
has laid waste to a vast stretch of the border area for no comprehensible reason”,
“to the great detriment of Nicaragua and the environment”. 3
1.2. It is for Nicaragua, as Claimant, to make good these allegations of fact
which are central to its claim and occupy the foreground of this dispute . By
contrast, the issues that divide the Parties so far as concerns the applicable
principles of international law on transboundary harm are relatively confined. The
focus of this Rejoinder is thus largely on completing the evidence before the
Court, and responding to the case on the facts as now put by Nicaragua.
1.3. As to that case, Costa Rica makesfive introductory observations.
1.4. First, given that the Road is being constructed solely within Costa Rican
territory, the central issue is whether any sediment reaching the San Juan River
from the construction works has resulted in significant harm to Nicaragua.
4
Despite Nicaragua’s colourful references to large quantities of dump- trucks, the
1
NR, para. 1.5.
2 NR, para. 1.13.
3 NR, para. 1.12.
4
E.g. NR, para. 1.18.
1 answer to that question is in the negative because any such sediment is inevitably
insignificant when compared to the sediment load that the River already carries.
Nicaragua’s experts have been unable to rebut this critical fact, which is further
confirmed by the scientific evidence and data presented with this Rejoinder.
1.5. Secondly, while Nicaragua wishes to portray the construction of the Road
as a mega -project, its own experts have f ocused on a length of approximately
41 km of road, of which just a few stretches form the pivot of Nicaragua’s
allegation of significant harm . These stretches appear again and again in the
photographic and other evidence deploye d by Nicaragua. They are in no sense
illustrative of the Road as a whole . Nicaragua has been unable to make out a case
to the contrary , and it has likewise been unable to make out a case that the
construction works on even these limited stretches have caus ed anything
approaching significant harm to its territory.
1.6. Indeed, with respect to the extreme nature of the claims that Nicaragua
makes as to the scale of the construction works, and the harm engendered, it is
recalled that the greater part of the Road i s along terrain that is completely flat,
and/or through land that has been farmed for decades generating no risk of
additional sediment entering the River. Further, this is a narrow road, averaging
less than 10 m in width, not a motorway or highway.
1.7. Costa Rica considers it very important that, in light of Nicaragua’s claims
(see e.g. paragraph 1.1 above), the Court has the best possible understanding of
the real scale of the works on the Road, and of the alleged scope for harm to
Nicaragua. To this end, in the letter accompanying this Rejoinder, Costa Rica has
proposed that a delegation of the Court could take advantage of the postponement
of the oral hearing (scheduled , until recently, for March 2015 ) to conduct a site
2answer to that question is in the negative because any such sediment is inevitably
insignificant when compared to the sediment load that the River already carries.
Nicaragua’s experts have been unable to rebut this critical fact, which is further
confirmed by the scientific evidence and data presented with this Rejoinder.
1.5. Secondly, while Nicaragua wishes to portray the construction of the Road
as a mega -project, its own experts have f ocused on a length of approximately
41 km of road, of which just a few stretches form the pivot of Nicaragua’s
allegation of significant harm . These stretches appear again and again in the
photographic and other evidence deploye d by Nicaragua. They are in no sense
illustrative of the Road as a whole . Nicaragua has been unable to make out a case
to the contrary , and it has likewise been unable to make out a case that the
construction works on even these limited stretches have caus ed anything
approaching significant harm to its territory.
1.6. Indeed, with respect to the extreme nature of the claims that Nicaragua
makes as to the scale of the construction works, and the harm engendered, it is
recalled that the greater part of the Road i s along terrain that is completely flat,
and/or through land that has been farmed for decades generating no risk of
additional sediment entering the River. Further, this is a narrow road, averaging
less than 10 m in width, not a motorway or highway.
1.7. Costa Rica considers it very important that, in light of Nicaragua’s claims
(see e.g. paragraph 1.1 above), the Court has the best possible understanding of
the real scale of the works on the Road, and of the alleged scope for harm to
Nicaragua. To this end, in the letter accompanying this Rejoinder, Costa Rica has
proposed that a delegation of the Court could take advantage of the postponement
of the oral hearing (scheduled , until recently, for March 2015 ) to conduct a site massive and uncontrolled deforestation in Costa Rica”, a s Nicaragua now
8
evidently wishes to re-cast it. The causes of sedimentation in the San Juan River
are many and complex and, as the Court is aware from the Certain Activities case,
the high sediment load long pre-dates the deforestation that Nicaragua now wishes
to bring to the fore. In fact, the high sediment load in the San Juan River is
principally the result of the geology of the region, in particular the tectonic and
9
volcanic activity in the area drained by the San Juan River and its tributaries.
1.10. This is not, however, a matter for the Court to resolve in this case. The
central question here is , and must remain , whether sedimentation from the
construction of the Road has led to significant harm to Nicaragua, alongside
whether alleged risks from the Road led to notification and EIA obligations on the
part of Costa Rica with which it failed to comply. Nicaragua’s belated attempt to
change the tenor of its case merely serves to h ighlight its inability to make good
its contentions on significant adverse impact on which its Application was
founded.
1.11. Fourthly, in the Introduction to its Reply, Nicaragua seeks to weigh up and
to reject Costa Rica’s reasons for the construction of the Road, suggesting that
Costa Rica should have responded differently or not at all to the acts of Nicaragua
that precipitated construction. Nicaragua’s central point here is to contend that it
did nothing to cause a perception of emergencyon the part of Costa Rica.
8
NR, para. 1.28.
9 Appendix A, 2015 Thorne Report, para. 4.140. Moreover, Nicaragua obscures the fact
that its own territory contributes in great proportion to the sediment load of the San Juan
River see the photographs in Figure 4.26 of the Thorne Report, reproduced for the Court’s
convenience in Chapter 2 below.
10 NR, paras. 1.7-1.12.
4massive and uncontrolled deforestation in Costa Rica”, a s Nicaragua now
8
evidently wishes to re-cast it. The causes of sedimentation in the San Juan River
are many and complex and, as the Court is aware from the Certain Activities case,
the high sediment load long pre-dates the deforestation that Nicaragua now wishes
to bring to the fore. In fact, the high sediment load in the San Juan River is
principally the result of the geology of the region, in particular the tectonic and
9
volcanic activity in the area drained by the San Juan River and its tributaries.
1.10. This is not, however, a matter for the Court to resolve in this case. The
central question here is , and must remain , whether sedimentation from the
construction of the Road has led to significant harm to Nicaragua, alongside
whether alleged risks from the Road led to notification and EIA obligations on the
part of Costa Rica with which it failed to comply. Nicaragua’s belated attempt to
change the tenor of its case merely serves to h ighlight its inability to make good
its contentions on significant adverse impact on which its Application was
founded.
1.11. Fourthly, in the Introduction to its Reply, Nicaragua seeks to weigh up and
to reject Costa Rica’s reasons for the construction of the Road, suggesting that
Costa Rica should have responded differently or not at all to the acts of Nicaragua
that precipitated construction. Nicaragua’s central point here is to contend that it
did nothing to cause a perception of emergencyon the part of Costa Rica.
8
NR, para. 1.28.
9 Appendix A, 2015 Thorne Report, para. 4.140. Moreover, Nicaragua obscures the fact
that its own territory contributes in great proportion to the sediment load of the San Juan
River see the photographs in Figure 4.26 of the Thorne Report, reproduced for the Court’s
convenience in Chapter 2 below.
10 NR, paras. 1.7-1.12. 1.14. While not bearing the burden of the proof, Costa Rica has sought to
provide scientifically reliable estimates of water flows and sediment loads of the
San Juan River, as well as ot her evidence, which demonstrate that no significant
harm has been, or risks being, caused to the river by the Road. The evidence that
Nicaragua has submitt ed in its Reply has focused on attempting to cast doubt
(unsuccessfully) on Costa Rica’s scientific reports, but surprisingly it has not
provided the basic measurements that would be expected to be the central plank of
its case on the facts.
The Structure of this Rejoinder
1.15. This Rejoinder is filed in accordance with the Court’s Order of
3 February 2014 setting the date for submission of Costa Rica’s Rejoinder as
2 February 2015.
1.16. The issues are presented as follows.
1.17. In Chapter 2, Costa Rica responds to the e vidence on alleged significant
harm that has been submitted by Nicaragua with its Reply. It is shown that
Nicaragua greatly overestimates the amount of sediment that has come from
construction of the Road, but is anyway unable to make out a case that even its
own estimated quantities of sediment have caused, or have risked causing,
significant harm to the San Juan River (as to which, as noted above, no flow or
sediment measurements have been provided) . Indeed, it is unable to show any
discernible impact to the pre-existing sediment load or bed load of the River. It is
likewise unable to show any other environmental or otherwise adverse impact or
risk of significant harm to Nicaragua.
61.14. While not bearing the burden of the proof, Costa Rica has sought to
provide scientifically reliable estimates of water flows and sediment loads of the
San Juan River, as well as ot her evidence, which demonstrate that no significant
harm has been, or risks being, caused to the river by the Road. The evidence that
Nicaragua has submitt ed in its Reply has focused on attempting to cast doubt
(unsuccessfully) on Costa Rica’s scientific reports, but surprisingly it has not
provided the basic measurements that would be expected to be the central plank of
its case on the facts.
The Structure of this Rejoinder
1.15. This Rejoinder is filed in accordance with the Court’s Order of
3 February 2014 setting the date for submission of Costa Rica’s Rejoinder as
2 February 2015.
1.16. The issues are presented as follows.
1.17. In Chapter 2, Costa Rica responds to the e vidence on alleged significant
harm that has been submitted by Nicaragua with its Reply. It is shown that
Nicaragua greatly overestimates the amount of sediment that has come from
construction of the Road, but is anyway unable to make out a case that even its
own estimated quantities of sediment have caused, or have risked causing,
significant harm to the San Juan River (as to which, as noted above, no flow or
sediment measurements have been provided) . Indeed, it is unable to show any
discernible impact to the pre-existing sediment load or bed load of the River. It is
likewise unable to show any other environmental or otherwise adverse impact or
risk of significant harm to Nicaragua.8 Chapter 2
The Absence of Adverse Impact of the Road on the San Juan River
A . Nicaragua’s Case
2.1. In these proceedings, Nicaragua claims that, as a result of the construction
of the road in Costa Rica, “large amounts of sediment are eroding into the River in
15
amounts sufficient to cause significant environmental harm.” It argues that
“Costa Rica has caused, and is continuing to cause, significant harm to
16
Nicaragua’s San Juan River and its natural environment.” It further argues that
Costa Rica’s construction of the road “has placed Nicaragua at grave risk of
17
continued harm, and nothing Costa Rica has done has mitigated this risk.” It
requests the Court to declare that Costa has breached its “obligation not to damage
Nicaraguan territory” and its “obligations under general international law and the
18
relevant environmental conventions”.
2.2. Nicaragua’s claims of significant harm rest on its case that the R oad is
contributing sediment to the River in quantities which cause harm. In its
Memorial, Nicaragua claimed that the volumes of sediment had a negative impact
upon (a) water quality; (b) morphology of the River; (c) navigation; and (d) the
19
ecosystem (including aquatic life and fishing), tourism and health. Nicaragua
15 NR, para. 2.1. See also NM, paras. 3.3, 3.60 and 5.58.
16
NR, para. 2.137.
17
NR, para 3.59.
18 NR, Submissions, paras. 1(ii) and (iii).
19 NM, para. 3.81.
9 did not however produce any evidence with its Memorial as to the existing
sediment load of the San Juan River. It merely asserted that the contribution of
additional sediment – which it estima tes based on the opinion expressed by Dr
Kondolf (opinion in turn based on his visual observations of the Road from the
River and from the air) – had an adverse impact on the River.
2.3. In its Counter -Memorial, Costa Rica produced e vidence demonstrating
that the Road was not contributing sediment to the River in quantities which cause
or could cause harm. This evidence consisted of comprehensive scientific and
technical reports relating to the impact of the Road on the San Juan River , which
squarely addressed the question whether the Road is contributing sediment to the
River, and if so, how much sediment, and they also considered the relative impact
of this sediment in the context of the existing sediment load of the River. 20 These
reports were assessed in the independent expert report of Professor Colin
21
Thorne.
2.4. In its Reply, Nicaragua submitted several reports responding to Costa
Rica’s evidence concerning the absence of adverse impact on the River. These
are:
(a) Dr G. Mathias Kondolf, “Erosion and Sediment Delivery to the Rio San
Juan from Route 1856”, July 2014, Annex 1 to Nicaragua’s Reply (the
2014 Kondolf Report);
20 See CRCM, Chapter 3.
21
See Professor Thorne, Assessment of the Impact of the Construction of the Border Road in
Costa Rica on t he San Juan River , November 2013, Appendix A to Costa Rica’s
Counter-Memorial (the 2013 Thorne Report).
10did not however produce any evidence with its Memorial as to the existing
sediment load of the San Juan River. It merely asserted that the contribution of
additional sediment – which it estima tes based on the opinion expressed by Dr
Kondolf (opinion in turn based on his visual observations of the Road from the
River and from the air) – had an adverse impact on the River.
2.3. In its Counter -Memorial, Costa Rica produced e vidence demonstrating
that the Road was not contributing sediment to the River in quantities which cause
or could cause harm. This evidence consisted of comprehensive scientific and
technical reports relating to the impact of the Road on the San Juan River , which
squarely addressed the question whether the Road is contributing sediment to the
River, and if so, how much sediment, and they also considered the relative impact
of this sediment in the context of the existing sediment load of the River. 20These
reports were assessed in the independent expert report of Professor Colin
21
Thorne.
2.4. In its Reply, Nicaragua submitted several reports responding to Costa
Rica’s evidence concerning the absence of adverse impact on the River. These
are:
(a) Dr G. Mathias Kondolf, “Erosion and Sediment Delivery to the Rio San
Juan from Route 1856”, July 2014, Annex 1 to Nicaragua’s Reply (the
2014 Kondolf Report);
20 See CRCM, Chapter 3.
21
See Professor Thorne, Assessment of the Impact of the Construction of the Border Road in
Costa Rica on t he San Juan River , November 2013, Appendix A to Costa Rica’s
Counter-Memorial (the 2013 Thorne Report). the dispute before the Court. Insofar a s Costa Rica’s experts have been able to
deal with this matter in the time available, they simply do not agree with the late
22
allegations made by Nicaragua. However, this matter is not a focus in this
Chapter because Costa Rica does not consider it to be properly within the scope of
the dispute before the Court, which concerns the construction of a Road in
Costa Rica.
2.7. Secondly, Nicaragua has persisted in its misplaced emphasis on the
question whether the Road was constructed wit h strict adherence to engineering
standards. It asserts that Costa Rica has “violated ‘the most basic, well accepted
road engineering and road maintenance principles normally applied during road
construction’” 23 and that Costa Rica has “disregarded the simp le but critical
24
principle that a highway construction project must be planned and designed” as
though these “principles” reflect international law obligations binding on
Costa Rica. They do not.
2.8. Nicaragua nonetheless persists in its references to a May 2012 report of
the National Laboratory of the University of Costa Rica (in its Spanish acronym,
25
LANAMME) and a June 2012 report of the Costa Rican Federated Association
26
of Engineers and Architects (the CFIA) , which, as Costa Ri ca explained in its
Counter-Memorial, do not evidence that environmental harm has or will be caused
22
See, e.g, Appendix A, 2015 Thorne Report, paras. 4.120-4.141.
23 NR, para. 3.2. See also paras. 3.3-3.15.
24
NR, para. 3.4.
25
NM, Annex 3, National Laboratory of Materials and Structural Models of the University
of Costa Rica, “Report INF-PITRA-014-12: Report from Inspection of Route 1856 - Juan
Rafael Mora Porras Border Road,” May 2012.
26 NM, Annex 4, Federated Association of Engineers and Architects of Costa Rica, “Report
on Inspection of the on the Border Road, NorthernArea Parallel to the San Juan River
CFIA Report”, 8 June 2012.
12the dispute before the Court. Insofar a s Costa Rica’s experts have been able to
deal with this matter in the time available, they simply do not agree with the late
22
allegations made by Nicaragua. However, this matter is not a focus in this
Chapter because Costa Rica does not consider it to be properly within the scope of
the dispute before the Court, which concerns the construction of a Road in
Costa Rica.
2.7. Secondly, Nicaragua has persisted in its misplaced emphasis on the
question whether the Road was constructed wit h strict adherence to engineering
standards. It asserts that Costa Rica has “violated ‘the most basic, well accepted
road engineering and road maintenance principles normally applied during road
construction’” 23 and that Costa Rica has “disregarded the simp le but critical
24
principle that a highway construction project must be planned and designed” as
though these “principles” reflect international law obligations binding on
Costa Rica. They do not.
2.8. Nicaragua nonetheless persists in its references to a May 2012 report of
the National Laboratory of the University of Costa Rica (in its Spanish acronym,
25
LANAMME) and a June 2012 report of the Costa Rican Federated Association
26
of Engineers and Architects (the CFIA) , which, as Costa Ri ca explained in its
Counter-Memorial, do not evidence that environmental harm has or will be caused
22
See, e.g, Appendix A, 2015 Thorne Report, paras. 4.120-4.141.
23 NR, para. 3.2. See also paras. 3.3-3.15.
24
NR, para. 3.4.
25
NM, Annex 3, National Laboratory of Materials and Structural Models of the University
of Costa Rica, “Report INF-PITRA-014-12: Report from Inspection of Route 1856 - Juan
Rafael Mora Porras Border Road,” May 2012.
26 NM, Annex 4, Federated Association of Engineers and Architects of Costa Rica, “Report
on Inspection of the on the Border Road, NorthernArea Parallel to the San Juan River
CFIA Report”, 8 June 2012. of sediment eroded from the Road to the River, what is in dispute is the addition
of sediment in the range of a very small – even imperceptible – increase in the
sediment load of a River which is naturally adapted to a sediment load that is
“very heavy”. 28 Nicaragua implicitly accepts that the impact of the Road is very
small, as is demonstrated by its belated attempt to depict the existing sediment
load as “excessive” and unnatural, and that the existing sedim ent load is the
responsibility of Costa Rica. 29 Ultimately, the issue for the Court to decide is
whether the Road – which in large part is a track built on existing paths – is
having a significant impact on the San Juan River.
2.10. This leads to a fourth preliminary point. Even accepting the estimates of
sediment eroding from the Road to the River put forward by Nicaragua’s experts
in 2013 (which Costa Rica does not), this sediment would represent only 1% to
2% of the total annual sediment load of the River. 30A contribution of sediment in
this proportion is obviously too small to have any adverse impact on the River,
either in respect of water quality or aggradation of the bed. The new estimates of
sediment eroding from the Road put forward with Nicaragua’s Reply would
indicate an addition of less than 3% to the total annual sediment load of the
28 Navigational Rights , NCM, para 1.1.8: “The sediment load that the San Juan River
receives from rivers originating in Costa Rica is y heavy. Thus, the sediment load
immediately downstream from the Sarapiqui River, measured at the beginning of the
seventies, was 10.2 million metric tons per year” (footnote omitted).
29 NR, paras. 2.75-2.79.
30
See CRCM, para. 3.33. As the Court noted in its Order of 13 December 2013 rejecting
Nicaragua’s Request for Provisional Measures in this case, a contribution of sediment in
the order of 1 to 2 per cent of the total sediment load in the San Juan River “seems too
small a proportion to have a signific ant impact on the river in the immediate future”: see
Construction of a Road in Costa Rica Along the San Juan River (Nicaragua v Costa
Rica), Request presented by Nicaragua for the Indication of Provisional Measures,
Order, 13 December 2013, para. 34.
14of sediment eroded from the Road to the River, what is in dispute is the addition
of sediment in the range of a very small – even imperceptible – increase in the
sediment load of a River which is naturally adapted to a sediment load that is
“very heavy”. 28 Nicaragua implicitly accepts that the impact of the Road is very
small, as is demonstrated by its belated attempt to depict the existing sediment
load as “excessive” and unnatural, and that the existing sedim ent load is the
responsibility of Costa Rica. 29 Ultimately, the issue for the Court to decide is
whether the Road – which in large part is a track built on existing paths – is
having a significant impact on the San Juan River.
2.10. This leads to a fourth preliminary point. Even accepting the estimates of
sediment eroding from the Road to the River put forward by Nicaragua’s experts
in 2013 (which Costa Rica does not), this sediment would represent only 1% to
2% of the total annual sediment load of the River. 30 A contribution of sediment in
this proportion is obviously too small to have any adverse impact on the River,
either in respect of water quality or aggradation of the bed. The new estimates of
sediment eroding from the Road put forward with Nicaragua’s Reply would
indicate an addition of less than 3% to the total annual sediment load of the
28 Navigational Rights , NCM, para 1.1.8: “The sediment load that the San Juan River
receives from rivers originating in Costa Rica is y heavy. Thus, the sediment load
immediately downstream from the Sarapiqui River, measured at the beginning of the
seventies, was 10.2 million metric tons per year” (footnote omitted).
29 NR, paras. 2.75-2.79.
30
See CRCM, para. 3.33. As the Court noted in its Order of 13 December 2013 rejecting
Nicaragua’s Request for Provisional Measures in this case, a contribution of sediment in
the order of 1 to 2 per cent of the total sediment load in the San Juan River “seems too
small a proportion to have a signific ant impact on the river in the immediate future”: see
Construction of a Road in Costa Rica Along the San Juan River (Nicaragua v Costa
Rica), Request presented by Nicaragua for the Indication of Provisional Measures,
Order, 13 December 2013, para. 34. 1856, November 2014 (the 2014 UCR Report ) (Annex 4); and Andreas
Mende, Inventory of Slopes and Water Courses related to the Border Road
No 1856 betwee n Mojón II and Delta Costa Rica: Second Report ,
December 2014 (the 2014 Mende Report) (Annex 3);
(b) issues of the potential impact of extreme weather and other exceptional
events: Juan Carlos Fallas Sojo, Comments on the Report by Dr Kondolf
as it pertains to Hurricanes and Tropical Storms, 2014 (the Fallas Report)
(Annex 9); and Allan Astorga, Extraordinary sediment inputs due to
exceptional events on the San Juan River , 2014 (the Astorga Report )
(Annex 10);
(c) issues of potential ecological impacts: Professor Ian Cowx, Independent
Expert Report concerning Evidence of Impacts on the Aquatic Ecology of
the San Juan River, Nicaragua due to construction of Route 1856 in Costa
Rica, 2014 (the Cowx Report) (Annex 2 ); Arturo Angulo, Fish Fauna in
the San Juan River, 2014 (the Angulo Report ) (Annex 7); Bernald
Pacheco, Answers and Study Analysis, “Ecological Impacts of the Route
1856 on the San Juan River, Nicaragua”, by Dr Rios Touma 2014, 2014
(the Pacheco Report ) (Annex 6 ); PE Gutierrez, Critical statistical
analysis of the report “Ecological Impacts of the Route 1856 on the San
Juan River, Nicaragua” by Blanca Ríos Touma, 2014 (the Gutierrez
Report) (Annex 8); and Centro Científico Tropical (CCT) Follow -up and
Monitoring Study Route 1856 Project - EDA Ecological Component ,
January 2015 (the 2015 CCT Report) (Annex 14).
(d) Costa Rica’s completed and ongoing remediation works: CODEFORSA,
Consulting Services for the Development and Implementation of an
16 1856, November 2014 (the 2014 UCR Report ) (Annex 4); and Andreas
Mende, Inventory of Slopes and Water Courses related to the Border Road
No 1856 betwee n Mojón II and Delta Costa Rica: Second Report ,
December 2014 (the 2014 Mende Report) (Annex 3);
(b) issues of the potential impact of extreme weather and other exceptional
events: Juan Carlos Fallas Sojo, Comments on the Report by Dr Kondolf
as it pertains to Hurricanes and Tropical Storms, 2014 (the Fallas Report)
(Annex 9); and Allan Astorga, Extraordinary sediment inputs due to
exceptional events on the San Juan River , 2014 (the Astorga Report )
(Annex 10);
(c) issues of potential ecological impacts: Professor Ian Cowx, Independent
Expert Report concerning Evidence of Impacts on the Aquatic Ecology of
the San Juan River, Nicaragua due to construction of Route 1856 in Costa
Rica, 2014 (the Cowx Report) (Annex 2 ); Arturo Angulo, Fish Fauna in
the San Juan River, 2014 (the Angulo Report ) (Annex 7); Bernald
Pacheco, Answers and Study Analysis, “Ecological Impacts of the Route
1856 on the San Juan River, Nicaragua”, by Dr Rios Touma 2014, 2014
(the Pacheco Report ) (Annex 6 ); PE Gutierrez, Critical statistical
analysis of the report “Ecological Impacts of the Route 1856 on the San
Juan River, Nicaragua” by Blanca Ríos Touma, 2014 (the Gutierrez
Report) (Annex 8); and Centro Científico Tropical (CCT) Follow -up and
Monitoring Study Route 1856 Project - EDA Ecological Component ,
January 2015 (the 2015 CCT Report) (Annex 14).
(d) Costa Rica’s completed and ongoing remediation works: CODEFORSA,
Consulting Services for the Development and Implementation of an a river such as the San Juan is a natural process, and one which is essential to the
33
life of the River. This process is commonly regarded as beneficial. Sediment
could only be regarded as a pollutant if its concentration is elevated compared to
the natural sediment load carried by theRiver. Sediment concentrations in the San
34
Juan River are high and highly variable, as Nicaragua accepts. As Professors
Thorne and Astorga explain, this is because the basin experiences extraordinary
sediment yields associated with earthquakes and volcani c eruptions that are a
35
natural consequence of its geology. In this context, sediment cannot be regarded
as a pollutant.
(1) Impact of the Road on the sediment load of the San Juan River:
before and after construction of the Road
(i) Establishing the baseline
2.15. In order to assess the baseline of the sediment load of the San Juan River
prior to the construction of Route 1856, Costa Rica’s experts initially referred to
the existing records, such as they are. In its Counter -Memorial, Costa Rica’s
experts had recourse to the only records that exist for the pre-construction period,
which date from 1974- 1976. The measurements of Suspended Sediment
Concentration in the San Juan made during this period were recorded jointly by
the two parti es, and were relied upon by Nicaragua in the Navigational Rights
case to assert that the sediment load of the San Juan River is “very heavy”. 36
33
See, eg, CRCM, Annex 81, GM Kondolf, “Hungry water: Effects of dams and gravel
mining on river channels” 21(4) (1997) Environmental Management 533.
34 NR, para. 2.125.
35 Appendix A, 2015 Thorne Report, para . 4.140; see also Vol . III, Annex 10, Astorga
Report, pp. 9-17.
36
Navigational Rights , NCM, para 1.1.8: “The sediment load that the San Juan River
receives from rivers originating in Costa Rica is very heavy. Thus, the sedint load
18a river such as the San Juan is a natural process, and one which is essential to the
33
life of the River. This process is commonly regarded as beneficial. Sediment
could only be regarded as a pollutant if its concentration is elevated compared to
the natural sediment load carried by theRiver. Sediment concentrations in the San
34
Juan River are high and highly variable, as Nicaragua accepts. As Professors
Thorne and Astorga explain, this is because the basin experiences extraordinary
sediment yields associated with earthquakes and volcani c eruptions that are a
35
natural consequence of its geology. In this context, sediment cannot be regarded
as a pollutant.
(1) Impact of the Road on the sediment load of the San Juan River:
before and after construction of the Road
(i) Establishing the baseline
2.15. In order to assess the baseline of the sediment load of the San Juan River
prior to the construction of Route 1856, Costa Rica’s experts initially referred to
the existing records, such as they are. In its Counter -Memorial, Costa Rica’s
experts had recourse to the only records that exist for the pre-construction period,
which date from 1974- 1976. The measurements of Suspended Sediment
Concentration in the San Juan made during this period were recorded jointly by
the two parti es, and were relied upon by Nicaragua in the Navigational Rights
case to assert that the sediment load of the San Juan River is “very heavy”. 36
33
See, eg, CRCM, Annex 81, GM Kondolf, “Hungry water: Effects of dams and gravel
mining on river channels” 21(4) (1997) Environmental Management 533.
34 NR, para. 2.125.
35 Appendix A, 2015 Thorne Report, para . 4.140; see also Vol . III, Annex 10, Astorga
Report, pp. 9-17.
36
Navigational Rights , NCM, para 1.1.8: “The sediment load that the San Juan River
receives from rivers originating in Costa Rica is very heavy. Thus, the sedint load 2.17. In its Reply, Nicaragua also contends that the measurements used by Costa
Rica “cannot support its conclusions because river flows and suspended sediment
loads vary considerably from year to year.” 40As Costa Rica’s experts explain, the
sediment load in a River such as the San Juan, which drains a tectonically acti ve
basin with live volcanoes , does indeed vary widely from year to year, not only
because of varying rainfall and runoff, but also because of extraordinary quantities
of sediment supplied by natural events such as landslides triggered by
41
earthquakes. In order to account for this natural variability, in 2014 Costa Rica’s
experts conducted a thorough analysis of records of the sediment loads measured
at Costa Rican hydrometric gauging stations within the San Juan basin, including
taking account of uncertaint ies in the time series. That analysis resulted in an
estimate of the Suspended Sediment Concentrations of the San Juan of
approximately 12.7 million tonnes per annum . 42 This is actually lower than the
approximation now put forward by Nicaragua’s expert, Dr Andrews, of about
13.7 million tonnes per annum. 43 It is therefore apparent that the difference
between the approximation made by Nicaragua and the estimate given by Costa
Rica’s experts is de minimis . However, adopting Costa Rica’s estimate (which is
lower) is conservative, because the relative contribution of sediment from the
Road will necessarily be higher if the baseline sediment load against which it is
compared is lower. Again, as explained in paragraph 2.10 above, because even on
Nicaragua’s own inflated figures of the Road-derived sediment that contribution is
40
NR, para 2.125, referring to NR, Annex 3, Andrews Report, Sections V(D) and V(E).
41 See Appendix A, 2015 Thorne Report, para. 4.140; and Vol . III, Annex 10, Astorga
Report, pp. 9-17.
42
Appendix A, 2015 Thorne Report, para. 4.77; and Vol . III , Annex 5, 2015 ICE Report,
pp. 15-19.
43 NR, Annex 3, Andrews Report, p. 27.
202.17. In its Reply, Nicaragua also contends that the measurements used by Costa
Rica “cannot support its conclusions because river flows and suspended sediment
loads vary considerably from year to year.” 40As Costa Rica’s experts explain, the
sediment load in a River such as the San Juan, which drains a tectonically acti ve
basin with live volcanoes , does indeed vary widely from year to year, not only
because of varying rainfall and runoff, but also because of extraordinary quantities
of sediment supplied by natural events such as landslides triggered by
41
earthquakes. In order to account for this natural variability, in 2014 Costa Rica’s
experts conducted a thorough analysis of records of the sediment loads measured
at Costa Rican hydrometric gauging stations within the San Juan basin, including
taking account of uncertaint ies in the time series. That analysis resulted in an
estimate of the Suspended Sediment Concentrations of the San Juan of
approximately 12.7 million tonnes per annum . 42 This is actually lower than the
approximation now put forward by Nicaragua’s expert, Dr Andrews, of about
13.7 million tonnes per annum. 43 It is therefore apparent that the difference
between the approximation made by Nicaragua and the estimate given by Costa
Rica’s experts is de minimis . However, adopting Costa Rica’s estimate (which is
lower) is conservative, because the relative contribution of sediment from the
Road will necessarily be higher if the baseline sediment load against which it is
compared is lower. Again, as explained in paragraph 2.10 above, because even on
Nicaragua’s own inflated figures of the Road-derived sediment that contribution is
40
NR, para 2.125, referring to NR, Annex 3, Andrews Report, Sections V(D) and V(E).
41 See Appendix A, 2015 Thorne Report, para. 4.140; and Vol . III, Annex 10, Astorga
Report, pp. 9-17.
42
Appendix A, 2015 Thorne Report, para. 4.77; and Vol . III , Annex 5, 2015 ICE Report,
pp. 15-19.
43 NR, Annex 3, Andrews Report, p. 27. Nicaragua contends that the assumption that the sediment data from the Colorado
River represents 91% of the sediment of the San Juan is not appropriate, because
it is based on a comparison of records collected at one gauge over a two -year
46
period with records collected at another gauge over another two -year period. As
Professor Thorne explains, “[t] he division of flows at the Delta could be
determined with confidence if Nicaragua or its experts measured and made known
47
the discharge of the lower R ío San Juan.” As Nicaragua has not done so, the
only basis on which Costa Rica’s experts are able to estimate the division of flows
is on basis of the available records, while taking due account for uncertainty.
2.21. Further, while Costa Rica maintains that its estimate of the division of
sediment into the Lower San Juan and the Colorado River is reliable, in order to
account for any variance in this flow, in the 2014 ICE Report calculations of the
sediment loads of the San Juan and the Lower San Juan were performed assuming
that 85%, 90% and 95% of the sediment load of the San Juan flows to the
48
Colorado and 15%, 10% and 5% to the Lower San Juan, respectively. As will be
seen below, on any of these assumptions, any contribution of sediment from the
Road to the River is having no adverse impact on the Lower San Juan, let alone
causing any significant harm.
2.22. Nicaragua makes three further criticisms of Costa Rica’s estimate of the
post-Road sediment load of the San Juan. First, it alleges that the samples were
49 50
collected improperly, or may have been collected improperly. These
46 NR, para. 2.126.
47 Appendix A, 2015 Thorne Report, para. 4.75.
48
Appendix A, 2015 Thorne Report, para. 4.76.
49 NR, para. 2.127.
50 NR, para. 2.129.
22Nicaragua contends that the assumption that the sediment data from the Colorado
River represents 91% of the sediment of the San Juan is not appropriate, because
it is based on a comparison of records collected at one gauge over a two -year
46
period with records collected at another gauge over another two -year period. As
Professor Thorne explains, “[t] he division of flows at the Delta could be
determined with confidence if Nicaragua or its experts measured and made known
47
the discharge of the lower R ío San Juan.” As Nicaragua has not done so, the
only basis on which Costa Rica’s experts are able to estimate the division of flows
is on basis of the available records, while taking due account for uncertainty.
2.21. Further, while Costa Rica maintains that its estimate of the division of
sediment into the Lower San Juan and the Colorado River is reliable, in order to
account for any variance in this flow, in the 2014 ICE Report calculations of the
sediment loads of the San Juan and the Lower San Juan were performed assuming
that 85%, 90% and 95% of the sediment load of the San Juan flows to the
48
Colorado and 15%, 10% and 5% to the Lower San Juan, respectively. As will be
seen below, on any of these assumptions, any contribution of sediment from the
Road to the River is having no adverse impact on the Lower San Juan, let alone
causing any significant harm.
2.22. Nicaragua makes three further criticisms of Costa Rica’s estimate of the
post-Road sediment load of the San Juan. First, it alleges that the samples were
49 50
collected improperly, or may have been collected improperly. These
46 NR, para. 2.126.
47 Appendix A, 2015 Thorne Report, para. 4.75.
48
Appendix A, 2015 Thorne Report, para. 4.76.
49 NR, para. 2.127.
50 NR, para. 2.129. but finite suspended sediment load even if it there were no flow whatsoever, or
that the sediment load reduces to zero at low discharges, both of which are
impossible.
2.25. Finally, Nicaragua seeks to challenge Costa Rica’s estimate of the bed
load. This estimate is necessary in order to evaluate Nicaragua’s claim that coarse
sediment added to the sediment load of the River (which would be transported as
bed load) has resulted in aggradation in the lower San Juan River. As no
measurements of bed load are available (and none were presented by Nicaragua),
it was necessary for Costa Rica’s experts to estimate the bed load using a bed load
transport equation.
2.26. Nicaragua challenges Costa Rica’s calculation of the bed load on the basis
that Costa Rica’s assumptions as to the slope of the River were incorrect. In
simple terms, the steeper the slope, the higher the bed load will be. Dr Kondolf
criticizes the slope figures given in Table 1 of the 2013 Thorne Report as being
56
overstated. As Professor Thorne explains in his 2014 Report, this criticism arises
from a mis-labelling of the column headings in Table 1 of his 2013 Report, which
are expressed in degrees. This minor error had no impact on the calculations made
by ICE in estimating the bed load component of the total load of the San Juan
River. 57
2.27. In order to provide as robust bed load estimates as possible (in the time
available), ICE has improved the bed load calculations using the “Engelund-
Hansen approach”, which was recommended by Dr Andrews as an alternative to
56 NR, Annex 1, 2014 Kondolf Report, p. 67; see also NR, paras. 2.132 -2.135.
57
Appendix A, 2015 Thorne Report, para. 4.46.
24but finite suspended sediment load even if it there were no flow whatsoever, or
that the sediment load reduces to zero at low discharges, both of which are
impossible.
2.25. Finally, Nicaragua seeks to challenge Costa Rica’s estimate of the bed
load. This estimate is necessary in order to evaluate Nicaragua’s claim that coarse
sediment added to the sediment load of the River (which would be transported as
bed load) has resulted in aggradation in the lower San Juan River. As no
measurements of bed load are available (and none were presented by Nicaragua),
it was necessary for Costa Rica’s experts to estimate the bed load using a bed load
transport equation.
2.26. Nicaragua challenges Costa Rica’s calculation of the bed load on the basis
that Costa Rica’s assumptions as to the slope of the River were incorrect. In
simple terms, the steeper the slope, the higher the bed load will be. Dr Kondolf
criticizes the slope figures given in Table 1 of the 2013 Thorne Report as being
56
overstated. As Professor Thorne explains in his 2014 Report, this criticism arises
from a mis-labelling of the column headings in Table 1 of his 2013 Report, which
are expressed in degrees. This minor error had no impact on the calculations made
by ICE in estimating the bed load component of the total load of the San Juan
River. 57
2.27. In order to provide as robust bed load estimates as possible (in the time
available), ICE has improved the bed load calculations using the “Engelund-
Hansen approach”, which was recommended by Dr Andrews as an alternative to
56 NR, Annex 1, 2014 Kondolf Report, p. 67; see also NR, paras. 2.132 -2.135.
57
Appendix A, 2015 Thorne Report, para. 4.46. 2.29. Two y ears ago, Costa Rica requested Nicaragua’s agreement “to take
discharge measurements and collect water samples from the San Juan River on a
monthly basis, to establish its chemical quality and to measure the sediment load
that the River carries”. 60A month later Nicaragua responded, suggesting that it
would be willing to take joint measurements, provided Costa Rica suspended all
Road construction works . 61 Given this unacceptable condition, Costa Rica
62
proposed through the Court a joint monitoring program me. Following a lengthy
63
exchange of notes, which caused inordinate delay, and in view of the impending
deadline for submission of Costa Rica’s Counter -Memorial, in September 2013
Costa Rica indicated that it would no longer pursue a joint programme, but instead
64
would encourage Nicaragua to carry out measurements on the River itself. In its
Reply, Nicaraguadid not present any evidence of such measurements.
60 CRCM, Annex 46, Note from the Minister of Foreign Affairs and Worship of Costa Rica
to the Minister of Foreign Affairs of Nicaragua, Reference DM -AM-063-13, 6 February
2013.
61 CRCM, Annex 48 , Note from the Minister of Foreign Affairs of Nicaragua to the
Minister of Foreign Affairs and Worship of Costa Rica, Costa Rica, R eference MRE/DM-
AJ/129/03/13, 5 March 2013.
62 CRCM, Annex 49 , Letter from the Co -Agent of Costa Rica to the Registrar of the
International Court of Justice, Reference ECRPB-013-2013, 7 March 2013.
63
CRCM, Annex 54 , Letter from the Agent of Nicaragua to the Registrar of the
International Court of Justice, Reference HOL-EMB-108, 14 June 2013; CRCM, Annex
55, Letter from the Co -Agent of Costa Rica to the Registrar of the International Court of
Justice, Reference ECRPB-036-13, 24 June 2013; CRCM, Annex 59 , Letter from the
Co-Agent of Costa Rica to the Registrar of the International Court Justice, Reference
ECRPB-052-13, 7 August 2013; CRCM, Annex 64 , Letter from the Agent of Nicaragua
to the Registrar of the International Court of Justice, Reference HOL-EMB-167,
30 August 2013; and CRCM, Annex 65 , Letter from the Co -Agent of Costa Rica to the
Registrar of the International Court of Justice, Reference ECRPB-63-2013, 27 September
2013.
64
CRCM, Annex 65 , Letter from the Co -Agent of Costa Rica to the Registrar of the
International Court of Justice, Reference ECRPB-63-2013, 27 September 2013.
262.29. Two y ears ago, Costa Rica requested Nicaragua’s agreement “to take
discharge measurements and collect water samples from the San Juan River on a
monthly basis, to establish its chemical quality and to measure the sediment load
that the River carries”. 60 A month later Nicaragua responded, suggesting that it
would be willing to take joint measurements, provided Costa Rica suspended all
Road construction works . 61 Given this unacceptable condition, Costa Rica
62
proposed through the Court a joint monitoring program me. Following a lengthy
63
exchange of notes, which caused inordinate delay, and in view of the impending
deadline for submission of Costa Rica’s Counter -Memorial, in September 2013
Costa Rica indicated that it would no longer pursue a joint programme, but instead
64
would encourage Nicaragua to carry out measurements on the River itself. In its
Reply, Nicaraguadid not present any evidence of such measurements.
60 CRCM, Annex 46, Note from the Minister of Foreign Affairs and Worship of Costa Rica
to the Minister of Foreign Affairs of Nicaragua, Reference DM -AM-063-13, 6 February
2013.
61 CRCM, Annex 48 , Note from the Minister of Foreign Affairs of Nicaragua to the
Minister of Foreign Affairs and Worship of Costa Rica, Costa Rica, R eference MRE/DM-
AJ/129/03/13, 5 March 2013.
62 CRCM, Annex 49 , Letter from the Co -Agent of Costa Rica to the Registrar of the
International Court of Justice, Reference ECRPB-013-2013, 7 March 2013.
63
CRCM, Annex 54 , Letter from the Agent of Nicaragua to the Registrar of the
International Court of Justice, Reference HOL-EMB-108, 14 June 2013; CRCM, Annex
55, Letter from the Co -Agent of Costa Rica to the Registrar of the International Court of
Justice, Reference ECRPB-036-13, 24 June 2013; CRCM, Annex 59 , Letter from the
Co-Agent of Costa Rica to the Registrar of the International Court Justice, Reference
ECRPB-052-13, 7 August 2013; CRCM, Annex 64 , Letter from the Agent of Nicaragua
to the Registrar of the International Court of Justice, Reference HOL-EMB-167,
30 August 2013; and CRCM, Annex 65 , Letter from the Co -Agent of Costa Rica to the
Registrar of the International Court of Justice, Reference ECRPB-63-2013, 27 September
2013.
64
CRCM, Annex 65 , Letter from the Co -Agent of Costa Rica to the Registrar of the
International Court of Justice, Reference ECRPB-63-2013, 27 September 2013. 2.32. Nicaragua then engaged in a course of conduct which appears to have been
intended to obstruct the carrying out of such measurements. It suggested a
different date and venue for the meeting, and suggested measurements be carried
out in the area of the new caños. 67 Costa Rica replied , accepting that
measurements be carried out in that area but insisting that the sites proposed by it
be maintained, because in accordance with Ramsar’s request it was necessary to
measure the volume and flow of the San Juan before and after the Delta of the
Colorado River. 68Nicaragua rejected these sites, 69and despite a further exchange
of diplomatic notes 70it proved impossible to reach an agreement.
2.33. In summary, Nicaragua has refused to present measurements of water
volumes and sediment loads of the San Juan of it s own, and when opportunities
were presented to carry out joint measurements of the San Juan and Color ado
Rivers which would have greatly benefited the Court in analysing Nicaragua’s
claims in this case, Nicaragua successfully managed to derail them.
* * *
2.34. As noted in paragraph 2.18 above, based on the available measurements,
Costa Rica’s experts estimate the average annual suspended sediment load for
67 Vol . IV, Annex 41, Note from the Minister of Foreign Affairs of Nicaragua to the
Minister of Foreign Affairs and Worship of Costa Rica, Reference
MRE/DM/AJ/439/10/14, 27 October 2014.
68 Vol . IV, Annex 42, Note from the Acting Minister of Foreign Affairs and Worship of
Costa Rica to the Minister of Foreign Affairs of Nicaraguaference DM-AM-0672-14,
28 October 2014.
69
Vol . IV, Annex 43 , Note from the Minister of Foreign Affairs of Nicaragua to the
Minister of Foreign Affairs and Worship of Costa Rica, Reference MRE/DM-
AJ/448/11/14, 3 November 2014.
70
Vol . IV, Annex 45, Note from the Minister of Foreign Affairs and Worship of Costa Rica
to the Minister of Foreign Affairs of N icaragua, Reference DM -AM-0697-14,
5 November 2014.
282.32. Nicaragua then engaged in a course of conduct which appears to have been
intended to obstruct the carrying out of such measurements. It suggested a
different date and venue for the meeting, and suggested measurements be carried
out in the area of the new caños. 67 Costa Rica replied , accepting that
measurements be carried out in that area but insisting that the sites proposed by it
be maintained, because in accordance with Ramsar’s request it was necessary to
measure the volume and flow of the San Juan before and after the Delta of the
Colorado River. 68Nicaragua rejected these sites, 69and despite a further exchange
of diplomatic notes 70it proved impossible to reach an agreement.
2.33. In summary, Nicaragua has refused to present measurements of water
volumes and sediment loads of the San Juan of it s own, and when opportunities
were presented to carry out joint measurements of the San Juan and Color ado
Rivers which would have greatly benefited the Court in analysing Nicaragua’s
claims in this case, Nicaragua successfully managed to derail them.
* * *
2.34. As noted in paragraph 2.18 above, based on the available measurements,
Costa Rica’s experts estimate the average annual suspended sediment load for
67 Vol . IV, Annex 41, Note from the Minister of Foreign Affairs of Nicaragua to the
Minister of Foreign Affairs and Worship of Costa Rica, Reference
MRE/DM/AJ/439/10/14, 27 October 2014.
68 Vol . IV, Annex 42, Note from the Acting Minister of Foreign Affairs and Worship of
Costa Rica to the Minister of Foreign Affairs of Nicaraguaference DM-AM-0672-14,
28 October 2014.
69
Vol . IV, Annex 43 , Note from the Minister of Foreign Affa irs of Nicaragua to the
Minister of Foreign Affairs and Worship of Costa Rica, Reference MRE/DM-
AJ/448/11/14, 3 November 2014.
70
Vol . IV, Annex 45, Note from the Minister of Foreign Affairs and Worship of Costa Rica
to the Minister of Foreign Affairs of N icaragua, Reference DM -AM-0697-14,
5 November 2014. 2.36. With its Reply, Nicaragua submitted a further report from Dr Kondolf,
based on his visual observations of the Road from the River and the air, and on
75
examination of sat ellite images. Dr Kondolf increased his estimate of the
sediment delivery (from 40% to 60%, 60% being the more conservative estimate
76
adopted by Costa Rica’s experts). He also added an additional 9,960 to
3 -1
19,920 m y for sediment delivery from “access ro ads” to the Road, which he
opines is delivered to the River annually. 77 On this basis, Dr Kondolf estimates
that sediment delivered to the River to between 116,000 and 150,000 m y , which 3 -1
converts to 177,020 and 250,500 tonnes per year . 78 This is a substant ial increase
on his initial estimate of 145,290 to 182,030 tonnes per year.
2.37. With its Counter -Memorial, Costa Rica’s experts considered the
contribution of sediment from the Road to the River based on the following
approach:
(a) Based on field measurements, experts from the Department of Civil
Engineering at the University of Costa Rica estimated land surface
lowering rates for (i) sheet erosion of the road bed and slopes;
(ii) landslides on cut slopes; (iii) gullies on cut slopes ; (iv) gullies on fill
slopes; and (v) rill erosion on cut slopes. 79 Based on the results of this
75
NR, Annex 1, 2014 Kondolf Report, p 59; see also NR, paras. 2.5- 2.6.
76
NR, Annex 1, 2014 Kondolf Report, p. 61. Costa Rica’s experts had assumed the more
conservative estimate of 60% in the evidence submitted with Costa Rica’s Counter -
Memorial.
77
NR, Annex 1, 2014 Kondolf Report, p. 62.
78 Ibid.
79 CRCM, Annex No 1 , University of Costa Rica Centre for Research in Sustainable
Development, Department of Civil Engineering, Report on Systematic Field monitoring
of Erosion and Sediment Yield along Route 1856 , September 2013 (the 2013 UCR
Report); CRCM, Appendix A, 2013 Thorne Report, para. 8.21.
302.36. With its Reply, Nicaragua submitted a further report from Dr Kondolf,
based on his visual observations of the Road from the River and the air, and on
75
examination of sat ellite images. Dr Kondolf increased his estimate of the
sediment delivery (from 40% to 60%, 60% being the more conservative estimate
76
adopted by Costa Rica’s experts). He also added an additional 9,960 to
3 -1
19,920 m y for sediment delivery from “access ro ads” to the Road, which he
opines is delivered to the River annually. 77 On this basis, Dr Kondolf estimates
that sediment delivered to the River to between 116,000 and 150,000 m y , which 3 -1
converts to 177,020 and 250,500 tonnes per year . 78 This is a substant ial increase
on his initial estimate of 145,290 to 182,030 tonnes per year.
2.37. With its Counter -Memorial, Costa Rica’s experts considered the
contribution of sediment from the Road to the River based on the following
approach:
(a) Based on field measurements, experts from the Department of Civil
Engineering at the University of Costa Rica estimated land surface
lowering rates for (i) sheet erosion of the road bed and slopes;
(ii) landslides on cut slopes; (iii) gullies on cut slopes ; (iv) gullies on fill
slopes; and (v) rill erosion on cut slopes. 79 Based on the results of this
75
NR, Annex 1, 2014 Kondolf Report, p 59; see also NR, paras. 2.5- 2.6.
76
NR, Annex 1, 2014 Kondolf Report, p. 61. Costa Rica’s experts had assumed the more
conservative estimate of 60% in the evidence submitted with Costa Rica’s Counter -
Memorial.
77
NR, Annex 1, 2014 Kondolf Report, p. 62.
78 Ibid.
79 CRCM, Annex No 1 , University of Costa Rica Centre for Research in Sustainable
Development, Department of Civil Engineering, Report on Systematic Field monitoring
of Erosion and Sediment Yield along Route 1856 , September 2013 (the 2013 UCR
Report); CRCM, Appendix A, 2013 Thorne Report, para. 8.21. (i) UCR Report: estimates of erosion rates
2.38. In its Reply, Nicaragua criticises UCR’s estimates of erosio n rates on two
grounds. First, Nicaragua suggests that the authors of the 2013 UCR Report
“ignore[d] many of the sites where erosion is the most serious”. 86 The nine sites
examined in the 2013 UCR Report were within the uppermost 15 km of the Road
and Nicaragua claims that the next 26 km of the Road “contains numerous sites
with much more serious erosion than those studied in the 15 km upriver.” 87
2.39. Secondly, Nicaragua criticizes the estimate presented in the 2013 UCR
88
Report for applying erosion rates measured at small features to larger features.
Nicaragua suggests that UCR ought to have made “actual measurements at an
adequate number of genuinely representative sites” instead. 89
2.40. In 2014, UCR added additional sites to its monitoring programme and
made use of more sophisticated technology in order to confirm that its estimates
made in 2013 were reliable. The additional sites included several of the sites
identified by Dr Kondolf as the most severely eroding sites. Their measurements
confirmed that the estimates made in 2013 were reliable – indeed, these additional
sites are in fact lowering the land surface at rates which are either comparable to
or lower than the rates estimated by UCR in 2013. These measure ments therefore
confirm that Dr Kondolf’s higher estimates – made on the basis of photographs
and observations of the Road from the River and the air – are overstated.
86
NM, para 2.100.
87
NM, para. 2.100, citing NR , Annex 1, 2014 Kondolf Report, Section 7 and Inventory of
Severely Eroding Sites, Appendix A.
88 NR, para 2.104, citing NR, Annex 1, 2014 Kondolf Report, Section 7.
89 NR, para. 2.104.
32(i) UCR Report: estimates of erosion rates
2.38. In its Reply, Nicaragua criticises UCR’s estimates of erosio n rates on two
grounds. First, Nicaragua suggests that the authors of the 2013 UCR Report
“ignore[d] many of the sites where erosion is the most serious”. 86 The nine sites
examined in the 2013 UCR Report were within the uppermost 15 km of the Road
and Nicaragua claims that the next 26 km of the Road “contains numerous sites
with much more serious erosion than those studied in the 15 km upriver.” 87
2.39. Secondly, Nicaragua criticizes the estimate presented in the 2013 UCR
88
Report for applying erosion rates measured at small features to larger features.
Nicaragua suggests that UCR ought to have made “actual measurements at an
adequate number of genuinely representative sites” instead. 89
2.40. In 2014, UCR added additional sites to its monitoring programme and
made use of more sophisticated technology in order to confirm that its estimates
made in 2013 were reliable. The additional sites included several of the sites
identified by Dr Kondolf as the most severely eroding sites. Their measurements
confirmed that the estimates made in 2013 were reliable – indeed, these additional
sites are in fact lowering the land surface at rates which are either comparable to
or lower than the rates estimated by UCR in 2013. These measure ments therefore
confirm that Dr Kondolf’s higher estimates – made on the basis of photographs
and observations of the Road from the River and the air – are overstated.
86
NM, para 2.100.
87
NM, para. 2.100, citing NR , Annex 1, 2014 Kondolf Report, Section 7 and Inventory of
Severely Eroding Sites, Appendix A.
88 NR, para 2.104, citing NR, Annex 1, 2014 Kondolf Report, Section 7.
89 NR, para. 2.104. Figure 4.1 (a) Terrestrial LiDAR being set up at monitoring Site 4 on May 27,
2014 (b) LiDAR point cloud (c) contour data for use in erosion measurements and
calculations.
(a)
(b)
(c)
Figure 4.2 (a) Orthophoto, (b) Digital Elevation Model and (c) cross -section
through gully at Site 12 (Dr Kondolf’s site 9.4) derived from photogrammetric
survey on October 28, 2014 and used to estimate gully planform area and eroded
volume.
2.42. In addition, UCR added three new sites to its monitoring, numbered Sites
11, 12 and 13 in UCR’s 2014 Report and corresponding to Sites 8.1, 9.4 and 9.5
92
in the 2014 Kondolf Report. For these sites, Dr Kondolf and Hagans & Weaver
92 Vol . III, Annex 4 , 2014 UCR Report, section 2.1; see alsoAppendix A, 2015 Thorne
Report, para. 4.3. Dr Kondolf’s Site 9.6 was not included in UCR’s study because it was
being mitigated at the time of study, and UCR took the very co nservative approach of
omitting sites where remediation was taking place, in effect assessing the impact of the
Road before any remediation of it was carried out: see Vol . III, Annex 4 , 2014 UCR
Report, section 2.1; see also Appendix A, 2015 Thorne Report, para. 4.19. As UCR
explained, Dr Kondolf’s Site 8.2 was not included because it did not display a single
34Figure 4.1 (a) Terrestrial LiDAR being set up at monitoring Site 4 on May 27,
2014 (b) LiDAR point cloud (c) contour data for use in erosion measurements and
calculations.
(a)
(b)
(c)
Figure 4.2 (a) Orthophoto, (b) Digital Elevation Model and (c) cross -section
through gully at Site 12 (Dr Kondolf’s site 9.4) derived from photogrammetric
survey on October 28, 2014 and used to estimate gully planform area and eroded
volume.
2.42. In addition, UCR added three new sites to its monitoring, numbered Sites
11, 12 and 13 in UCR’s 2014 Report and corresponding to Sites 8.1, 9.4 and 9.5
92
in the 2014 Kondolf Report. For these sites, Dr Kondolf and Hagans & Weaver
92 Vol . III, Annex 4 , 2014 UCR Report, section 2.1; see alsoAppendix A, 2015 Thorne
Report, para. 4.3. Dr Kondolf’s Site 9.6 was not included in UCR’s study because it was
being mitigated at the time of study, and UCR took the very co nservative approach of
omitting sites where remediation was taking place, in effect assessing the impact of the
Road before any remediation of it was carried out: see Vol . III, Annex 4 , 2014 UCR
Report, section 2.1; see also Appendix A, 2015 Thorne Report, para. 4.19. As UCR
explained, Dr Kondolf’s Site 8.2 was not included because it did not display a single Table 5 .2 Maximum annual erosion rates from the 2014 UCR Report
Erosion type Fill slope Cut slope
erosion rate erosion rate
(m/yr) (m/yr)
a
Rotational landslide 0.40 0.40
Gully 0.76 d 0.27
b
Rill 0.16 0.16
Sheet 0.14 c 0.07
a. As no rotational landslides were measured in fill slopes, the cut slope landslide erosion rate is
recommended.
b. The 2013 report conservatively used the same erosion rate for rills innd fill slopes
and this has been repeated in this report. The estimated erosion rate for rills in fill slopes is
lower (0.07 m/yr.) therefore the higher erosion rate recorded in cut slopes (0.16 m/yr.) has been
conservatively recommended for both sites.
c. Recommended sheet erosion rate is estimated by doubling rate measured for a cut slope to
account for uncompacted condition of soil in fill prisms.
2.44. In addition, UCR provide estimates of road surface erosion, which were
subsequently adjusted by ICE for stretches of dirt and gravel road and stretches
94
where the Road is merely a trail.
2.45. As Professor Thorne explains, UCR’s 2014 measurements confirm that
UCR’s 2013 erosion rates were not “unrepresentative ly low”, as Nicaragua now
claims:
“These measurements reveal that while the gullies at Dr Kondolf’s sites 8.1,
9.4 and 9.5 are indeed larger and have eroded greater volumes of sediment
than gullies formed in fill slopes monitored at UCR Sites 8 and 9, the mean
annual erosion rates (that is their volumes eroded divided by their planform
areas, divided by their age, i.e. how much they have lowered the ground
surface in a year) at Dr Kondolf’s sites 8.1, 9.4 and 9.5 are actually much
lower than that measured at Site 8, and are comparable to those measured at
Site 9.
94
Appendix A, 2015 Thorne Report, para. 4.79 and Table 4.15.
36 Table 5 .2 Maximum annual erosion rates from the 2014 UCR Report
Erosion type Fill slope Cut slope
erosion rate erosion rate
(m/yr) (m/yr)
a
Rotational landslide 0.40 0.40
Gully 0.76 d 0.27
b
Rill 0.16 0.16
Sheet 0.14 c 0.07
a. As no rotational landslides were measured in fill slopes, the cut slope landslide erosion rate is
recommended.
b. The 2013 report conservatively used the same erosion rate for rills in nd fill slopes
and this has been repeated in this report. The estimated erosion rate for rills in fill slopes is
lower (0.07 m/yr.) therefore the higher erosion rate recorded in cut slopes (0.16 m/yr.) has been
conservatively recommended for both sites.
c. Recommended sheet erosion rate is estimated by doubling rate measured for a cut slope to
account for uncompacted condition of soil in fill prisms.
2.44. In addition, UCR provide estimates of road surface erosion, which were
subsequently adjusted by ICE for stretches of dirt and gravel road and stretches
94
where the Road is merely a trail.
2.45. As Professor Thorne explains, UCR’s 2014 measurements confirm that
UCR’s 2013 erosion rates were not “unrepresentative ly low”, as Nicaragua now
claims:
“These measurements reveal that while the gullies at Dr Kondolf’s sites 8.1,
9.4 and 9.5 are indeed larger and have eroded greater volumes of sediment
than gullies formed in fill slopes monitored at UCR Sites 8 and 9, the mean
annual erosion rates (that is their volumes eroded divided by their planform
areas, divided by their age, i.e. how much they have lowered the ground
surface in a year) at Dr Kondolf’s sites 8.1, 9.4 and 9.5 are actually much
lower than that measured at Site 8, and are comparable to those measured at
Site 9.
94
Appendix A, 2015 Thorne Report, para. 4.79 and Table 4.15. the 2013 UCR Report, resulting in a more conservative estimate of the rate of
sediment eroding from the Road to the River. In the updated analysis submitted in
the 2014 UCR Report and the 2014 Mende Report, the same rates are now used,
100
without any variance, so the discrepancy no longer exists. In any event, this
discrepancy was always immaterial because Costa Rica’s approach in its Counter-
Memorial was to show that, even on the estimates of Nicaragu a’s own experts of
Road-derived sediment, there was no adverse impact on the River. This remains
the case so far as concerns Nicaragua’s latest expert evidence, as discussed further
in paragraphs 2.62-2.65 below.
2.49. Secondly, Nicaragua asserts that the areas set out in the 2013 Inventory are
“underestimated” and “based on visual estimates”, rather than actual
101
measurements, resulting in estimates which are unreasonably low. It is correct
that visual estimates were used in 2013, and it is to be noted that the process of
estimating areas in the 2013 Inventory was made difficult by the fact as a result of
inclement weather, making field measurements difficult, and rendering some of
the stretches of the Road inaccessible by vehicle or even on foot, as Dr Mende
102
explains.
2.50. In the 2014 dry season, a new field campaign was carried out by Dr
Mende. Dr Mende was able to inspect every slope and watercourse crossing
between Marker II and Delta Costa Rica, which allow ed him to more closely
scrutinise the condition of the Road. In addition, this field work was done using
more advanced technology than that which was available to Dr Mende in 2013. 103
100
Vol . II, Annex 3, 2014 Mende Report, p. 1.
101 NR, para. 2.110.
102
Vol . II, Annex 3, 2014 Mende Report, p. 3.
103 Vol . II, Annex 3, 2014 Mende Report, p. 4.
38the 2013 UCR Report, resulting in a more conservative estimate of the rate of
sediment eroding from the Road to the River. In the updated analysis submitted in
the 2014 UCR Report and the 2014 Mende Report, the same rates are now used,
100
without any variance, so the discrepancy no longer exists. In any event, this
discrepancy was always immaterial because Costa Rica’s approach in its Counter-
Memorial was to show that, even on the estimates of Nicaragu a’s own experts of
Road-derived sediment, there was no adverse impact on the River. This remains
the case so far as concerns Nicaragua’s latest expert evidence, as discussed further
in paragraphs 2.62-2.65 below.
2.49. Secondly, Nicaragua asserts that the areas set out in the 2013 Inventory are
“underestimated” and “based on visual estimates”, rather than actual
101
measurements, resulting in estimates which are unreasonably low. It is correct
that visual estimates were used in 2013, and it is to be noted that the process of
estimating areas in the 2013 Inventory was made difficult by the fact as a result of
inclement weather, making field measurements difficult, and rendering some of
the stretches of the Road inaccessible by vehicle or even on foot, as Dr Mende
102
explains.
2.50. In the 2014 dry season, a new field campaign was carried out by Dr
Mende. Dr Mende was able to inspect every slope and watercourse crossing
between Marker II and Delta Costa Rica, which allow ed him to more closely
scrutinise the condition of the Road. In addition, this field work was done using
more advanced technology than that which was available to Dr Mende in 2013. 103
100
Vol . II, Annex 3, 2014 Mende Report, p. 1.
101 NR, para. 2.110.
102
Vol . II, Annex 3, 2014 Mende Report, p. 3.
103 Vol . II, Annex 3, 2014 Mende Report, p. 4. Table 4 .7 Example calculation of annual erosion volume for Cut Slope T-8a.
Erosion type Cut slope Slope area Estimated annual
erosion rate affected volume of erosion
(m/y) (m ) (m /y)
Sheet erosion 0.07 185
13
Rills 0.16 554 89
Gullies 0.27 369 100
Land Slides 0.40 739 296
Totals -- 1,847 497
2.53. This calculation was performed for all 201 slopes along t he Road,
3
resulting in an estimate of slope erosion from the Road of 72,000 m /y, which
converts to 120,000 t/y. 107 As Professor Thorne explains, this estimate is more
accurate because it was made using more advanced technology, and it is also very
conservative because it uses the highest average erosion rates for all four possible
erosion processes:
“This volume is based on a scenario in which all four erosion processes
operate at their upper bound rates, simultaneously at every slope along the
3
entire length of the Road. For erosion of 72,000 m actually to occur in one
year, it would require rainfall sufficiently heavy, frequent and widespread to
maximize annual erosion rates along the entire length of the Road, which is
improbable for the meteorological rea sons explained in Section 4D , below.
Hence, I believe this to be a 'worst case' rainfall scenario for slope erosion
along the Road and one that is actually very unlikely to occur, making it a
highly conservative estimate. Also, no account is taken of reductions in
slope erosion resulting from the programme of erosion mitigation performed
by CONAVI and CODEFORSA, which has progressed significantly since
2013 (see Section 7, below). It follows that the annual slope erosion volume
of 72,000 m 3/y produced by D r Mende is very much a ‘worst case’ value,
not a mean annual average value.
The slope erosion volume estimated in 2014 is nearly double that estimated
in 2013, which was 36,590 m /y (or 61,100 t/y). The increase results from
107 Vol . II, Annex 3, 2014 Mende Report, p. 30; see also Appendix A , 2015 Thorne Report,
para. 4.37.
40 Table 4 .7 Example calculation of annual erosion volume for Cut Slope T-8a.
Erosion type Cut slope Slope area Estimated annual
erosion rate affected volume of erosion
(m/y) (m ) (m /y)
Sheet erosion 0.07 185
13
Rills 0.16 554 89
Gullies 0.27 369 100
Land Slides 0.40 739 296
Totals -- 1,847 497
2.53. This calculation was performed for all 201 slopes along t he Road,
3
resulting in an estimate of slope erosion from the Road of 72,000 m /y, which
converts to 120,000 t/y. 107As Professor Thorne explains, this estimate is more
accurate because it was made using more advanced technology, and it is also very
conservative because it uses the highest average erosion rates for all four possible
erosion processes:
“This volume is based on a scenario in which all four erosion processes
operate at their upper bound rates, simultaneously at every slope along the
3
entire length of the Road. For erosion of 72,000 m actually to occur in one
year, it would require rainfall sufficiently heavy, frequent and widespread to
maximize annual erosion rates along the entire length of the Road, which is
improbable for the meteorological rea sons explained in Section 4D , below.
Hence, I believe this to be a 'worst case' rainfall scenario for slope erosion
along the Road and one that is actually very unlikely to occur, making it a
highly conservative estimate. Also, no account is taken of reductions in
slope erosion resulting from the programme of erosion mitigation performed
by CONAVI and CODEFORSA, which has progressed significantly since
2013 (see Section 7, below). It follows that the annual slope erosion volume
of 72,000 m 3/y produced by D r Mende is very much a ‘worst case’ value,
not a mean annual average value.
The slope erosion volume estimated in 2014 is nearly double that estimated
in 2013, which was 36,590 m /y (or 61,100 t/y). The increase results from
107 Vol . II, Annex 3, 2014 Mende Report, p. 30; see also Appendix A , 2015 Thorne Report,
para. 4.37. 110
experts in 2013 and adopted by Dr Kondolf as reasonable in his 2014 Report,
the resulting estimate of Road-derived sediment delivered to the River on a worst
3 111
case conservative basis is 74,949 tonnes (or 44,880 m ) per year. This is
represented in Table 4.16 and Figure 4.16 of the 2015 Thorne Report, reproduced
below for convenience.
110
Appendix A, 2015 Thorne Report, para. 4.80.
111 Appendix A, 2015 Thorne Report, para. 4.80 and Table 4.16.
42 110
experts in 2013 and adopted by Dr Kondolf as reasonable in his 2014 Report,
the resulting estimate of Road-derived sediment delivered to the River on a worst
3 111
case conservative basis is 74,949 tonnes (or 44,880 m ) per year. This is
represented in Table 4.16 and Figure 4.16 of the 2015 Thorne Report, reproduced
below for convenience.
110
Appendix A, 2015 Thorne Report, para. 4.80.
111 Appendix A, 2015 Thorne Report, para. 4.80 and Table 4.16. 2.58. Concerning Dr Kondolf’s suggestion that these access roads are
contributing significant amounts of sediment to the River, Professor Thorne
concludes as follows:
“Bearing in mind the stable condition of the access roads, their remoteness
from the River and the scarcity of streams linking them to the River, in my
opinion it is highly unlikely that sediment from these access roads reaches
113
the Río San Juan in any appreciable quantities.”
2.59. Nicaragua also criticises Costa Rica’s estimates of erosion on the ground
that they do not take into account additional erosion from “failed stream
114
crossings”. As Professor Thorne explains, a high proportion of these sites have
been remediated or are in the process of or scheduled to be remediated, and on
that basis UCR excluded them from their monitoring programme. In any event,
given the very limited nature of these inputs, Professor Thorne explains that they
cannot have any significant or long-lasting effect on the River:
“These yields of Road-derived sediment do not consider erosion from areas
disturbed during construction in 2011. This is because those areas have
subsequently revegetated, either naturally or due to vegetation planting by
CODEFORSA and CONAVI. Neither do the estimates consider erosion at
failed watercourses. This was criticised in relation to the 2013 estimates in
paragraph 2.119 of Nicaragua’s Reply. To explain why it was decided not to
attempt to estimate erosion at failed crossings in 2014, it is only necessary to
examine a typical example, as illustrated by Dr Kondolf in Figure 24, on
page 36 of his 2014 Report, which shows the point where the Road
intersects a small ditch draining an area of pasture. The width of the ditch is
not specified, but as the Road has an average width of 10 m and the ditch is
clearly much narrower than this, it is perhaps 2 m wide. In the vicinity of the
ditch, the channel of the Río San Juan is about 200 m wide. The River in this
3
reach conveys an average annual discharge of the order of 500 m /s and an
annual sediment load of several millions of tonnes. It follows that volume of
113 Appendix A, 2015 Thorne Report, para. 7.32.
114
NR, para. 2.119.
442.58. Concerning Dr Kondolf’s suggestion that these access roads are
contributing significant amounts of sediment to the River, Professor Thorne
concludes as follows:
“Bearing in mind the stable condition of the access roads, their remoteness
from the River and the scarcity of streams linking them to the River, in my
opinion it is highly unlikely that sediment from these access roads reaches
113
the Río San Juan in any appreciable quantities.”
2.59. Nicaragua also criticises Costa Rica’s estimates of erosion on the ground
that they do not take into account additional erosion from “failed stream
114
crossings”. As Professor Thorne explains, a high proportion of these sites have
been remediated or are in the process of or scheduled to be remediated, and on
that basis UCR excluded them from their monitoring programme. In any event,
given the very limited nature of these inputs, Professor Thorne explains that they
cannot have any significant or long-lasting effect on the River:
“These yields of Road-derived sediment do not consider erosion from areas
disturbed during construction in 2011. This is because those areas have
subsequently revegetated, either naturally or due to vegetation planting by
CODEFORSA and CONAVI. Neither do the estimates consider erosion at
failed watercourses. This was criticised in relation to the 2013 estimates in
paragraph 2.119 of Nicaragua’s Reply. To explain why it was decided not to
attempt to estimate erosion at failed crossings in 2014, it is only necessary to
examine a typical example, as illustrated by Dr Kondolf in Figure 24, on
page 36 of his 2014 Report, which shows the point where the Road
intersects a small ditch draining an area of pasture. The width of the ditch is
not specified, but as the Road has an average width of 10 m and the ditch is
clearly much narrower than this, it is perhaps 2 m wide. In the vicinity of the
ditch, the channel of the Río San Juan is about 200 m wide. The River in this
3
reach conveys an average annual discharge of the order of 500 m /s and an
annual sediment load of several millions of tonnes. It follows that volume of
113 Appendix A, 2015 Thorne Report, para. 7.32.
114
NR, para. 2.119. sediment eroded from the Road in the context of the existing sediment load of the
River.
2.63. As explained in subsection (1) above, the best estimate of the average
annual total load in the San Juan being revised to 12,678,000 t/ y, comprising a
suspended load of 9,078,000 t/y and a bedload of 3,600,000 t/y. In the Lower San
Juan, the total load is estimated to be 2,181,000 t, comprising a suspended load of
1,479,000 t plus bedload of 702,000 t. 117
2.64. As noted in paragraph 2.61 above, the average input of sediment from the
Road to the River is estimated to be approximately 75,000 t/y (ignoring the
impacts of the mediation works) . In the context of a sediment load of
12,678,000 t/y, this “is an indiscernible 0.6% of the total load” of the River. 118
This is obviously too small a proportion to have any impact on the River, let alone
any significant impact.
2.65. As noted in paragraph 2.36 above, Dr Kondolf’s estimate of a range of
sediment delivered annually from the Road to the River (116,000 to 150,000 m/y,
-1
which converts to 194,000 to 250,500 t y ) is significantly overstated. But even if
it were an accurate assessment, which Costa Rica does not accept, it would
119
represent only 1% to 2% of the total annual sediment load of the River , or 2%
120
to 3% of the suspended sediment load. A contribution of sediment in this range
121
is similarly too small to have any adverse impact on the River. Finally, if
-1
Dr Andrews’ estimate of the average annual total load (13.7 million t y ) were
117 Appendix A, 2015 Thorne Report, Table 4.17(b).
118 Appendix A, 2015 Thorne Report, para. 4.94.
119
Appendix A, 2015 Thorne Report, para. 4.93.
120 Appendix A, 2015 Thorne Report, para. 4.114.
121 Appendix A, 2015 Thorne Report, para. 4.114.
46sediment eroded from the Road in the context of the existing sediment load of the
River.
2.63. As explained in subsection (1) above, the best estimate of the average
annual total load in the San Juan being revised to 12,678,000 t/ y, comprising a
suspended load of 9,078,000 t/y and a bedload of 3,600,000 t/y. In the Lower San
Juan, the total load is estimated to be 2,181,000 t, comprising a suspended load of
1,479,000 t plus bedload of 702,000 t. 117
2.64. As noted in paragraph 2.61 above, the average input of sediment from the
Road to the River is estimated to be approximately 75,000 t/y (ignoring the
impacts of the mediation works) . In the context of a sediment load of
12,678,000 t/y, this “is an indiscernible 0.6% of the total load” of the River. 118
This is obviously too small a proportion to have any impact on the River, let alone
any significant impact.
2.65. As noted in paragraph 2.36 above, Dr Kondolf’s estimate of a range of
sediment delivered annually from the Road to the River (116,000 to 150,000 m/y,
-1
which converts to 194,000 to 250,500 t y ) is significantly overstated. But even if
it were an accurate assessment, which Costa Rica does not accept, it would
119
represent only 1% to 2% of the total annual sediment load of the River , or 2%
120
to 3% of the suspended sediment load. A contribution of sediment in this range
121
is similarly too small to have any adverse impact on the River. Finally, if
-1
Dr Andrews’ estimate of the average annual total load (13.7 million t y ) were
117 Appendix A, 2015 Thorne Report, Table 4.17(b).
118 Appendix A, 2015 Thorne Report, para. 4.94.
119
Appendix A, 2015 Thorne Report, para. 4.93.
120 Appendix A, 2015 Thorne Report, para. 4.114.
121 Appendix A, 2015 Thorne Report, para. 4.114. of the sediment will settle in the first three kilometres of the Lower San
Juan.
(b) Second, growth of the micro- delta some 30 kilometres downstream of
Delta Colorado indicates that the Lower San Juan has capacity to transport
sand throughout its length. 126
(c) Third, more than 20 sites where Nicaragua has carried out dredging
operations on the Lower San Juan are located downstream of the first three
127
kilometres. If all or even “nearly all” of the coarse sediment entering the
Lower San Juan were deposited in the first three kilometres, as Nicaragua
now suggests, it is unlikely that Nicaragua would have any need to dredge
in these downstream areas, including in the vicinity of the “disputed
territory”.
2.68. As Professor Thorne explains, even accepting the figures of Road- derived
sediment put forward by Nicaragua’s expert (which, for the reasons explained
above, are significantly overstated), and the proposition that all of it is deposited
in the first three kilometres of the Lower San Juan (which, for the reasons
explained above, is an untenable proposition), this would only cause the bed of
the Lower San Juan to rise by 5 to 10 mm per year. As Professor Thorne states:
“Were I to accept Dr Kondolf’s 2014 estimate that the quantity of sediment
derived from the Road plus all the access roads delivered to the River
3
annually is between 116,000 and 150,000 m , which I do not, and applying
Dr Andrews’ assumptions that 10% of that sediment is carr ied into the
Lower Río San Juan and that 12 to 18% of it is relatively coarse, then 1,390
to 2,700 m 3 of sand from Route 1856 plus its access roads would be added
126 Appendix A, 2015 Thorne Report, para. 5.27.
127
See Certain Activities, CRM, Sketch Map 7 .1 ; and Appendix A, 2015 Thorne Report,
para. 5.28.
48 of the sediment will settle in the first three kilometres of the Lower San
Juan.
(b) Second, growth of the micro- delta some 30 kilometres downstream of
Delta Colorado indicates that the Lower San Juan has capacity to transport
sand throughout its length. 126
(c) Third, more than 20 sites where Nicaragua has carried out dredging
operations on the Lower San Juan are located downstream of the first three
127
kilometres. If all or even “nearly all” of the coarse sediment entering the
Lower San Juan were deposited in the first three kilometres, as Nicaragua
now suggests, it is unlikely that Nicaragua would have any need to dredge
in these downstream areas, including in the vicinity of the “disputed
territory”.
2.68. As Professor Thorne explains, even accepting the figures of Road- derived
sediment put forward by Nicaragua’s expert (which, for the reasons explained
above, are significantly overstated), and the proposition that all of it is deposited
in the first three kilometres of the Lower San Juan (which, for the reasons
explained above, is an untenable proposition), this would only cause the bed of
the Lower San Juan to rise by 5 to 10 mm per year. As Professor Thorne states:
“Were I to accept Dr Kondolf’s 2014 estimate that the quantity of sediment
derived from the Road plus all the access roads delivered to the River
3
annually is between 116,000 and 150,000 m , which I do not, and applying
Dr Andrews’ assumptions that 10% of that sediment is carr ied into the
Lower Río San Juan and that 12 to 18% of it is relatively coarse, then 1,390
to 2,700 m 3 of sand from Route 1856 plus its access roads would be added
126 Appendix A, 2015 Thorne Report, para. 5.27.
127
See Certain Activities, CRM, Sketch Map 7 .1 ; and Appendix A, 2015 Thorne Report,
para. 5.28. “Dr Andrews’ opinion might be correct for a delta building into a marine
water body that experiences frequent ‘tidal surges’. But the Caribbean has a
micro-tidal regime, with a diurnal tidal amplitude averaging only about 20
cm (Kjerfve, 1981). This explains why most of the fine sediment carried by
the lower Río San Juan is not deposited within the delta but is carried into
the Caribbean Sea, as I indicated in my 2013 Report and as illustrated in
typical, rainy season satellite images (Figure 5.4), that show plumes of
turbid river water extending into the Bay of San Juan del Norte and the
littoral zone of the Caribbean Sea.” 131
Indeed, this is demonstrated by the satellite images included in Figure 5.4 to
Professor Thorne’s report, which show plumes of turbid river water extending into
the Bay of San Juan del Norte and the littoral zone of the Caribbean Sea
(reproduced below for convenience).
(a)
131
Appendix A, 2015 Thorne Report, para. 5.38.
50 “Dr Andrews’ opinion might be correct for a delta building into a marine
water body that experiences frequent ‘tidal surges’. But the Caribbean has a
micro-tidal regime, with a diurnal tidal amplitude averaging only about 20
cm (Kjerfve, 1981). This explains why most of the fine sediment carried by
the lower Río San Juan is not deposited within the delta but is carried into
the Caribbean Sea, as I indicated in my 2013 Report and as illustrated in
typical, rainy season satellite images (Figure 5.4), that show plumes of
turbid river water extending into the Bay of San Juan del Norte and the
littoral zone of the Caribbean Sea.” 131
Indeed, this is demonstrated by the satellite images included in Figure 5.4 to
Professor Thorne’s report, which show plumes of turbid river water extending into
the Bay of San Juan del Norte and the littoral zone of the Caribbean Sea
(reproduced below for convenience).
(a)
131
Appendix A, 2015 Thorne Report, para. 5.38. amounts of sediment appear to be entering the San Juan from Nicaraguan
territory.
(a) (b)
Figure 4 .26 Turbid water draining to the Río San Juan from Nicaraguan tributaries on
23 December 2012 (a) Río Santa Cruz (b) Río Sábalos.
(5) Potential impact of rainfall from a hurricane or tropical storm
2.71. In its Memorial, relying on the evidence of Dr Kondolf that in the region
of the road “rainfall intensities can be very high, especially during tropical storms
and hurricanes”, 134Nicaragua argued that the volume of sediment being delivered
from the Road to the River will “increase dramatically”. 135
2.72. As Costa Rica explained in its Counter -Memorial, the region in which the
136
Road is located has never been directly hit by a hurricane, and hurricanes
134
NM, Annex 1, 2012 Kondolf Report, para. 4.7.
135 NM, para. 4.19.
136 See CRCM, Annex 13, United States National Oceanic and Atmospheric Administration,
Map of Historical Hurricane Tracks, available at http://csc.noaa.gov/hurricanes.
52amounts of sediment appear to be entering the San Juan from Nicaraguan
territory.
(a) (b)
Figure 4 .26 Turbid water draining to the Río San Juan from Nicaraguan tributaries on
23 December 2012 (a) Río Santa Cruz (b) Río Sábalos.
(5) Potential impact of rainfall from a hurricane or tropical storm
2.71. In its Memorial, relying on the evidence of Dr Kondolf that in the region
of the road “rainfall intensities can be very high, especially during tropical storms
and hurricanes”, 134 Nicaragua argued that the volume of sediment being delivered
from the Road to the River will “increase dramatically”. 135
2.72. As Costa Rica explained in its Counter -Memorial, the region in which the
136
Road is located has never been directly hit by a hurricane, and hurricanes
134
NM, Annex 1, 2012 Kondolf Report, para. 4.7.
135 NM, para. 4.19.
136 See CRCM, Annex 13, United States National Oceanic and Atmospheric Administration,
Map of Historical Hurricane Tracks, available at http://csc.noaa.gov/hurricanes. Pacific than in catchments draining to the Caribbean, such as the San Juan
140
River.
2.75. Based on Professor Fallas’ expert report, Professor Thorne concludes that
Nicaragua’s experts have overstated both the risk of unprecedented rainfall and
the potential impact on sediment loads in the San Juan River of the rainfall
associated with a hurricane or tropical storm. As he states:
“In my opinion as a geomorphologist, the risk of rapid erosion due to intense
rainfall in the area around the Road is probably greater during the localised
thunderstorms associated with a Tropical Wave than would be the case
during a Tropical Cyclone. However, the frequency of localised downpours
is high and their impacts limited because, to restate my position, ‘ the
hydrology, sediment dynamics, morphology and environment of the River
are fully adjusted to the effects o 141 frequent an d heavy rainstorms’ (2013
Thorne Report, paragraph 6.20).”
2.76. In a further attempt to make out its claim of adverse impact, Nicaragua
argues that “[s]evere erosion from the Road can also be expected in the event of
an earthquake”, 142 which Nicaragua considers to be a “very real” risk. 143 Although
there have been earthquakes in the region, as Professor Thorne explains, the area
of slopes on the Road is tiny in comparison to the area of which would be
144
disturbed by landslides triggered by an earthquake. In these circumstances, the
additional risk associated with an earthquake in the vicinity of the Road is
correspondingly tiny.
140 Vol . III, Annex 9, Fallas Report, p. 3.
141
Appendix A, 2015 Thorne Report, para. 4.112.
142 NR, para. 3.55.
143
NR, para. 3.56.
144 Appendix A, 2015 Thorne Report, paras. 4.128-4.129.
54Pacific than in catchments draining to the Caribbean, such as the San Juan
140
River.
2.75. Based on Professor Fallas’ expert report, Professor Thorne concludes that
Nicaragua’s experts have overstated both the risk of unprecedented rainfall and
the potential impact on sediment loads in the San Juan River of the rainfall
associated with a hurricane or tropical storm. As he states:
“In my opinion as a geomorphologist, the risk of rapid erosion due to intense
rainfall in the area around the Road is probably greater during the localised
thunderstorms associated with a Tropical Wave than would be the case
during a Tropical Cyclone. However, the frequency of localised downpours
is high and their impacts limited because, to restate my position, ‘ the
hydrology, sediment dynamics, morphology and environment of the River
are fully adjusted to the effects o 141 frequent an d heavy rainstorms’ (2013
Thorne Report, paragraph 6.20).”
2.76. In a further attempt to make out its claim of adverse impact, Nicaragua
argues that “[s]evere erosion from the Road can also be expected in the event of
an earthquake”, 142 which Nicaragua considers to be a “very real” risk. 143 Although
there have been earthquakes in the region, as Professor Thorne explains, the area
of slopes on the Road is tiny in comparison to the area of which would be
144
disturbed by landslides triggered by an earthquake. In these circumstances, the
additional risk associated with an earthquake in the vicinity of the Road is
correspondingly tiny.
140 Vol . III, Annex 9, Fallas Report, p. 3.
141
Appendix A, 2015 Thorne Report, para. 4.112.
142 NR, para. 3.55.
143
NR, para. 3.56.
144 Appendix A, 2015 Thorne Report, paras. 4.128-4.129. load and deposition calculations using an upper bound estimate of the
amount of Road- derived coarse sediment entering the lower Río San Juan
suggest that this is indiscernible compared to pre -existing coars e load,
especially when allowance is made for uncertainty concerning estimation of
the bedload carried by the River and the proportions in which flow and
sediment are divided when flow bifurcates at the Delta.
Sediment continuity dictates that even if all of the coarse Road -derived
sediment supplied to the lower Río San Juan in one year according to Dr
Andrew’s estimate (which I do not accept) were to be deposited on the bed
of the channel within the first three kilometres downstream of the Delta it
would, on average, raise the bed of the river by less than 5to 10 mm.” 147
2.80. As noted in paragraph 2.2 above, Nicaragua’s claims as to the significant
harm which it alleges is being caused to the River are based on the contribution of
sediment from the Road to the River. In the circumstances Costa Rica’s experts
show that the sediment contribution is insignificant and indiscernible even using
the estimates of that contribution put forward by Nicaragua’s experts, Nicaragua’s
claims must fail.
C . There is No Risk of Any Other Adverse Impact on the San Juan River
2.81. Nicaragua claims that the Road has had an adverse impact on the San Juan
River in respect of (1) water quality; (2) channel morphology; (3) navigation; and
(4) ecosystem, tourism and health. It also claims that (5) Costa Rica has failed and
is failing to remediate the Road’s defects. Each of these claims is addressed in
turn below. A follow up study (the 2015 CCT Report) from the Tropical Science
Centre (CCT, in its Spanish acronym), which validates the findings made in 2013
EDA, further shows that all impacts on Costa Rican territory remain localized. 148
147
Appendix A, 2015 Thorne Report, paras. 8.6-8.7.
148 Vol . III, Annex 14, 2015 CCT Report.
56 load and deposition calculations using an upper bound estimate of the
amount of Road- derived coarse sediment entering the lower Río San Juan
suggest that this is indiscernible compared to pre -existing coars e load,
especially when allowance is made for uncertainty concerning estimation of
the bedload carried by the River and the proportions in which flow and
sediment are divided when flow bifurcates at the Delta.
Sediment continuity dictates that even if all of the coarse Road -derived
sediment supplied to the lower Río San Juan in one year according to Dr
Andrew’s estimate (which I do not accept) were to be deposited on the bed
of the channel within the first three kilometres downstream of the Delta it
would, on average, raise the bed of the river by less than 5to 10 mm.” 147
2.80. As noted in paragraph 2.2 above, Nicaragua’s claims as to the significant
harm which it alleges is being caused to the River are based on the contribution of
sediment from the Road to the River. In the circumstances Costa Rica’s experts
show that the sediment contribution is insignificant and indiscernible even using
the estimates of that contribution put forward by Nicaragua’s experts, Nicaragua’s
claims must fail.
C . There is No Risk of Any Other Adverse Impact on the San Juan River
2.81. Nicaragua claims that the Road has had an adverse impact on the San Juan
River in respect of (1) water quality; (2) channel morphology; (3) navigation; and
(4) ecosystem, tourism and health. It also claims that (5) Costa Rica has failed and
is failing to remediate the Road’s defects. Each of these claims is addressed in
turn below. A follow up study (the 2015 CCT Report) from the Tropical Science
Centre (CCT, in its Spanish acronym), which validates the findings made in 2013
EDA, further shows that all impacts on Costa Rican territory remain localized. 148
147
Appendix A, 2015 Thorne Report, paras. 8.6-8.7.
148 Vol . III, Annex 14, 2015 CCT Report. impacts on morphology, making only a nother very general statement that
153
sediment can impact channel morphology.
2.86. In its Reply, Nicaragua asserts that the Road “has caused undeniable
morphological changes to the River including, most visibly, t he creation of large
deltas of sediment in the River, as well as the deposit of significant quantities of
sediment on the bed of the lower San Juan River.” 154
2.87. Insofar as Nicaragua’s claim of harm on the basis of morphological
changes is based on the deposition of sediment on the bed of the Lower San Juan,
as explained in paragraph 2.68 above, any aggradation in the bed of the Lower
San Juan which has been caused by Road- derived sediment is indiscernible and
cannot have had any adve rse impact. As Professor Thorne explains, even
accepting the figures of Road-derived sediment put forward by Nicaragua’s expert
(which, for the reasons explained above, are significantly overstated), and the
proposition that all of it is deposited in the f irst three kilometres of the Lower San
Juan (which, for the reasons explained above, is an untenable proposition), this
would cause the bed of the Lower San Juan to rise by less than 5 to 10 mm per
year.155 This is not by any measure “significant” and does n ot suffice to establish
a claim for harm based on morphological change.
2.88. Concerning Nicaragua’s claim that the Road has caused the creation of or
addition to “large deltas” of sediment in the River, as Professor Thorne explains,
there is no evidence as to whether the deltas which Nicaragua asserts were created
as a result of the Road did or did not exist before the Road was constructed. Due
153 NM, Annex 1, 2012 Kondolf Report, p. 37, para. 4.9.
154
NR, para. 2.2.
155 Appendix A, 2015 Thorne Report, para. 5.33.
58impacts on morphology, making only a nother very general statement that
153
sediment can impact channel morphology.
2.86. In its Reply, Nicaragua asserts that the Road “has caused undeniable
morphological changes to the River including, most visibly, t he creation of large
deltas of sediment in the River, as well as the deposit of significant quantities of
sediment on the bed of the lower San Juan River.” 154
2.87. Insofar as Nicaragua’s claim of harm on the basis of morphological
changes is based on the deposition of sediment on the bed of the Lower San Juan,
as explained in paragraph 2.68 above, any aggradation in the bed of the Lower
San Juan which has been caused by Road- derived sediment is indiscernible and
cannot have had any adve rse impact. As Professor Thorne explains, even
accepting the figures of Road-derived sediment put forward by Nicaragua’s expert
(which, for the reasons explained above, are significantly overstated), and the
proposition that all of it is deposited in the f irst three kilometres of the Lower San
Juan (which, for the reasons explained above, is an untenable proposition), this
would cause the bed of the Lower San Juan to rise by less than 5 to 10 mm per
year.155 This is not by any measure “significant” and does n ot suffice to establish
a claim for harm based on morphological change.
2.88. Concerning Nicaragua’s claim that the Road has caused the creation of or
addition to “large deltas” of sediment in the River, as Professor Thorne explains,
there is no evidence as to whether the deltas which Nicaragua asserts were created
as a result of the Road did or did not exist before the Road was constructed. Due
153 NM, Annex 1, 2012 Kondolf Report, p. 37, para. 4.9.
154
NR, para. 2.2.
155 Appendix A, 2015 Thorne Report, para. 5.33. Figure 5 .3 Pre- and Post-Road satellite images establishing that at least two of the eight
south bank deltas identified as being formed from sediment derived from the Road were
present prior to construction of the Road.
2.89. In any event, th e eight deltas identified by Dr Kondolf on the southern
bank are morphologically indistinguishable from those which exist on the
northern bank (during an overflight in April 2014, 15 deltas were photographed in
the Nicaraguan bank of the San Juan), and in fact it appears that deltas on the
northern bank are larger than those on the southern bank (see Figure 5.2 to the
2015 Thorne Report, which is partially reproduced below).
60Figure 5 .3 Pre- and Post-Road satellite images establishing that at least two of the eight
south bank deltas identified as being formed from sediment derived from the Road were
present prior to construction of the Road.
2.89. In any event, th e eight deltas identified by Dr Kondolf on the southern
bank are morphologically indistinguishable from those which exist on the
northern bank (during an overflight in April 2014, 15 deltas were photographed in
the Nicaraguan bank of the San Juan), and in fact it appears that deltas on the
northern bank are larger than those on the southern bank (see Figure 5.2 to the
2015 Thorne Report, which is partially reproduced below).
Extract of Figure 5 .2 Fifteen north bank deltas photographed from C osta Rican airspace in
April 2014. These deltas are formed in sediment eroded from Nicaragua and some are
considerably larger than any of those photographed by Dr Kondolf along the south bank. The
size and morphology of these deltas should be compared to those shown in Appendix F of the
2014 Kondolf Report, which were also taken at conditions of low flow in the Río San Juan.
(Partial reproduction Appendix A, 2015 Thorne Report, pp. 92-95) 2.90. Moreover, Dr Kondolf describes the composition of the sediment on the
deltas adjacent to the Costa Rican bank as “angular, friable clasts” and the
sediment on the deltas on the northern (i.e. Nicaraguan) bank as formed in “more
rounded, competent gravels”. 157 As Professor Thorne observes, the fac t that
sediment on the deltas on the Costa Rican bank are friable “indicates that they will
quickly weather down to rounded, gravel -sized particles similar to those forming
deltas at the south bank of the Rio San Juan”, 158 such that “the half -life of their
residence on deltas will be measured in months rather than years” and they will be
“easily absorbed within the existing load of the Rio San Juan.” 159 This process
will be further accelerated by Costa Rica’s mitigation works, which will cut off
the supply of new clasts. 160 In contrast, the deltas on the north bank are formed in
less crumbly gravels which indicates that they “are formed from stream bed
material that has been transported considerable distances from its eroding source,
that these grains do not crumble, and that they will remain too large for the Rio
San Juan to transport away for years or decades”. 161 Professor Thorne concludes
that any impacts of the deltas built or enlarged by Road -derived clasts is local and
transitory: “that any contribution they make to morphological features in the River
is insignificant due to their spatially restricted extent and because their existence
in the channel will be short lived.” 162 In the absence of any significant impact on
the morphology of the River, Nicaragua’s clai m to adverse impact based on the
creation of deltas must be dismissed.
157
NR, Annex 1, 2014 Kondolf Report, p. 70.
158 Appendix A, 2015 Thorne Report, para. 5.12.
159
Appendix A, 2015 Thorne Report, para. 5.14.
160
Appendix A, 2015 Thorne Report, para. 5.16.
161 Appendix A, 2015 Thorne Report, para. 5.12.
162 Appendix A, 2015 Thorne Report, para. 5.17.
622.90. Moreover, Dr Kondolf describes the composition of the sediment on the
deltas adjacent to the Costa Rican bank as “angular, friable clasts” and the
sediment on the deltas on the northern (i.e. Nicaraguan) bank as formed in “more
rounded, competent gravels”. 157 As Professor Thorne observes, the fac t that
sediment on the deltas on the Costa Rican bank are friable “indicates that they will
quickly weather down to rounded, gravel -sized particles similar to those forming
deltas at the south bank of the Rio San Juan”, 158such that “the half -life of their
residence on deltas will be measured in months rather than years” and they will be
“easily absorbed within the existing load of the Rio San Juan.” 159 This process
will be further accelerated by Costa Rica’s mitigation works, which will cut off
the supply of new clasts. 160 In contrast, the deltas on the north bank are formed in
less crumbly gravels which indicates that they “are formed from stream bed
material that has been transported considerable distances from its eroding source,
that these grains do not crumble, and that they will remain too large for the Rio
San Juan to transport away for years or decades”. 161 Professor Thorne concludes
that any impacts of the deltas built or enlarged by Road -derived clasts is local and
transitory: “that any contribution they make to morphological features in the River
is insignificant due to their spatially restricted extent and because their existence
in the channel will be short lived.” 162 In the absence of any significant impact on
the morphology of the River, Nicaragua’s clai m to adverse impact based on the
creation of deltas must be dismissed.
157
NR, Annex 1, 2014 Kondolf Report, p. 70.
158 Appendix A, 2015 Thorne Report, para. 5.12.
159
Appendix A, 2015 Thorne Report, para. 5.14.
160
Appendix A, 2015 Thorne Report, para. 5.16.
161 Appendix A, 2015 Thorne Report, para. 5.12.
162 Appendix A, 2015 Thorne Report, para. 5.17. that Nicaragua’s claim of adverse impact based on navi gation on the River must
be dismissed.
(4) Ecosystem, Tourism and Health
2.94. Nicaragua makes further claims as to the impact of the road infrastructure
works on the San Juan River, as to adverse effects on the ecosystem (in particular,
fish, algae and macroinvertebrates) and tourism. In its Memorial, Nicaragua also
claimed that the construction of the Road has impacted upon the health of the
riparians of the River. In its Request for Provisional Measures, Nicaragua made a
168
similar assertion, but it was not substantiated in any way, and it appeared to
169
abandon it. Nicaragua now appears to have abandoned it in this proceeding
entirely, as it is not referred to in its Reply.
2.95. Concerning impact on the ecosystem , in its Memorial, Nicaragua alleged
harm to the ecosystem of the River, 170 and claimed compensation for losses
allegedly suffered in respect of fishing. 171 Having failed to substantiate this claim
for the reasons explained in Costa Rica’s Counter -Memorial, in its Reply,
Nicaragua has articulated its claim as one for damage caused to aquatic
organisms. 172 Its claim appears to have two aspects: first, the impact of Road -
168
Vol . IV, Annex 24 , Letter from Nicaragua to the Registrar of the International Court of
Justice, Reference HOL-EMB-196, 11 October 2013, p. 3.
169 See CR 2013/31, p. 16, para. 3 (Wordsworth) and p. 33, para. 30 (Wordsworth); see also
CR 2013/29, p. 42, para. 21 (Wordsworth). None of Nicaragua’s counsel mentioned the
word “health” in their oral submissions.
170 NM, para. 3.93. See also paras. 1.9-1.10, 5.61 (referring to Nicaragua’s report to the
Court dated 23 July 2012 in the Certain Activities case), and 5.67.
171 NM, para. 6.33.
172
NR, para. 2.78.
64that Nicaragua’s claim of adverse impact based on navi gation on the River must
be dismissed.
(4) Ecosystem, Tourism and Health
2.94. Nicaragua makes further claims as to the impact of the road infrastructure
works on the San Juan River, as to adverse effects on the ecosystem (in particular,
fish, algae and macroinvertebrates) and tourism. In its Memorial, Nicaragua also
claimed that the construction of the Road has impacted upon the health of the
riparians of the River. In its Request for Provisional Measures, Nicaragua made a
168
similar assertion, but it was not substantiated in any way, and it appeared to
169
abandon it. Nicaragua now appears to have abandoned it in this proceeding
entirely, as it is not referred to in its Reply.
2.95. Concerning impact on the ecosystem , in its Memorial, Nicaragua alleged
harm to the ecosystem of the River, 170 and claimed compensation for losses
allegedly suffered in respect of fishing. 171 Having failed to substantiate this claim
for the reasons explained in Costa Rica’s Counter -Memorial, in its Reply,
Nicaragua has articulated its claim as one for damage caused to aquatic
organisms. 172 Its claim appears to have two aspects: first, the impact of Road -
168
Vol . IV, Annex 24 , Letter from Nicaragua to the Registrar of the International Court of
Justice, Reference HOL-EMB-196, 11 October 2013, p. 3.
169 See CR 2013/31, p. 16, para. 3 (Wordsworth) and p. 33, para. 30 (Wordsworth); see also
CR 2013/29, p. 42, para. 21 (Wordsworth). None of Nicaragua’s counsel mentioned the
word “health” in their oral submissions.
170 NM, para. 3.93. See also paras. 1.9 -1.10, 5.61 (referring to Nicaragua’s report to the
Court dated 23 July 2012 in the Certain Activities case), and 5.67.
171 NM, para. 6.33.
172
NR, para. 2.78. these families may be vulnerable to high loads of sediment, other species are
adapted to high loads. After reviewing the available data and literature, Professor
Cowx concludes that “there is no evidence that the fish and fisheries of the San
179
Juan have or will be impacted by construction of Route 1856.” He notes:
“What the literature actually demonstrates is that Dr Kondolf’s statem ent
is a gross over -generalisation. While some members of the families of
fishes he names are vulnerable to increased in turbidity and suspended
sediment, other members of those families are adapted to high sediment
loading and this is illustrated through the species specific review
summarised herein and reported in detail in the references cited.
Empirical data on the species impacted with specific reference to the San
Juan River are required to justify and substantiate claims of any long -term
impact of construction of Route 1856 on the fish and fisheries of the river.
No such data have been provided by Nicaragua’s experts. The examples
used as evidence are general and unspecific to the San Juan River and the
species that inhabit it.”180
2.98. The second aspect of Nicaragua’s claim concerning impact of the Road on
the ecosystem of the San Juan concerns algae and macroinvertebrates. During the
hearing on Nicaragua’s Request for Provisional Measures, Dr Kondolf reported
results of sampling of periphyton done by Dr Ríos in May 2013. 181 Dr Ríos has
submitted a separate report with Nicaragua’s Rejoinder, in which she seeks to
compare deltas on the southern bank which Nicaragua alleges are composed of
Road-derived sediment with deltas on the northern bank of the River. 182 She
179 Vol . II, Annex 2, Cowx Report, p. 13.
180
Vol . II, Annex 2, Cowx Report, p. 13. These conclusions are also confirmed by a report
produced by a Costa Rican fish expert, Arturo Angulo Sibaja, which concludes that the
relevant species of fish are well adapted to high levels of sediment: see Vol . III, Annex 7,
A Angulo, Fish Fauna in the San Juan River, 2014.
181
Third Kondolf Report, p. 13.
182 NR, Annex 4, Dr Blanca Ríos Touma, “Ecological Impacts of the Route 1856 on the San
Juan River, Nicaragua”, July 2014, (the 2014 Ríos Report).
66these families may be vulnerable to high loads of sediment, other species are
adapted to high loads. After reviewing the available data and literature, Professor
Cowx concludes that “there is no evidence that the fish and fisheries of the San
179
Juan have or will be impacted by construction of Route 1856.” He notes:
“What the literature actually demonstrates is that Dr Kondolf’s statem ent
is a gross over -generalisation. While some members of the families of
fishes he names are vulnerable to increased in turbidity and suspended
sediment, other members of those families are adapted to high sediment
loading and this is illustrated through the species specific review
summarised herein and reported in detail in the references cited.
Empirical data on the species impacted with specific reference to the San
Juan River are required to justify and substantiate claims of any long -term
impact of construction of Route 1856 on the fish and fisheries of the river.
No such data have been provided by Nicaragua’s experts. The examples
used as evidence are general and unspecific to the San Juan River and the
species that inhabit it.”180
2.98. The second aspect of Nicaragua’s claim concerning impact of the Road on
the ecosystem of the San Juan concerns algae and macroinvertebrates. During the
hearing on Nicaragua’s Request for Provisional Measures, Dr Kondolf reported
results of sampling of periphyton done by Dr Ríos in May 2013. 181Dr Ríos has
submitted a separate report with Nicaragua’s Rejoinder, in which she seeks to
compare deltas on the southern bank which Nicaragua alleges are composed of
Road-derived sediment with deltas on the northern bank of the River. 182 She
179 Vol . II, Annex 2, Cowx Report, p. 13.
180
Vol . II, Annex 2, Cowx Report, p. 13. These conclusions are also confirmed by a report
produced by a Costa Rican fish expert, Arturo Angulo Sibaja, which concludes that the
relevant species of fish are well adapted to high levels of sediment: see Vol . III, Annex 7,
A Angulo, Fish Fauna in the San Juan River, 2014.
181
Third Kondolf Report, p. 13.
182 NR, Annex 4, Dr Blanca Ríos Touma, “Ecological Impacts of the Route 1856 on the San
Juan River, Nicaragua”, July 2014, (the 2014 Ríos Report). (a) failure to take into account the comparative size of the drainage areas on
the deltas on the northern and southern banks, which could explain any
differences in ecological health on those deltas;
(b) failure to control for differences in the areas, natural vegetation and land
use in the catchments draining to the deltas, which could give rise to the
differences allegedly found by Ríos in the ecological health of the deltas;
(c) significant deficiencies in the statistical analysis applied to the data
collected, resulting in unreliable conclusions; and
(d) the family level identification of invertebrates sampled in the River, which
does not account for the fact that different species in the same family may
have higher adaptability to adverse conditions, including in respect of
sediment levels. 186
2.102. On the basis of these and other factors, Professor Cowx concludes that:
“Evidence provided in the Ríos Report that compares environmental bio-
indicators for deltas on the northern and southern banks is largely
inconclusive and fails to provide the robust empirical data necessary to
prove that sediment eroded from the Road has adversel y impacted the
187
aquatic ecology of the San Juan River.”
And:
“It is therefore unsound for Dr Kondolf to conclude that Road- derived
sediment has had negative effects [on] invertebrate communities in the San
Juan River.” 188
186
Vol . II, Annex 2, Cowx Report; and Appendix A, 2015 Thorne Report, paras. 6.22-6.35.
187 Vol . II, Annex 2, Cowx Report, p. 19.
188 Vol . II, Annex 2, Cowx Report, p. 18.
68(a) failure to take into account the comparative size of the drainage areas on
the deltas on the northern and southern banks, which could explain any
differences in ecological health on those deltas;
(b) failure to control for differences in the areas, natural vegetation and land
use in the catchments draining to the deltas, which could give rise to the
differences allegedly found by Ríos in the ecological health of the deltas;
(c) significant deficiencies in the statistical analysis applied to the data
collected, resulting in unreliable conclusions; and
(d) the family level identification of invertebrates sampled in the River, which
does not account for the fact that different species in the same family may
have higher adaptability to adverse conditions, including in respect of
sediment levels. 186
2.102. On the basis of these and other factors, Professor Cowx concludes that:
“Evidence provided in the Ríos Report that compares environmental bio-
indicators for deltas on the northern and southern banks is largely
inconclusive and fails to provide the robust empirical data necessary to
prove that sediment eroded from the Road has adversel y impacted the
187
aquatic ecology of the San Juan River.”
And:
“It is therefore unsound for Dr Kondolf to conclude that Road- derived
sediment has had negative effects [on] invertebrate communities in the San
Juan River.” 188
186
Vol . II, Annex 2, Cowx Report; and Appendix A, 2015 Thorne Report, paras. 6.22-6.35.
187 Vol . II, Annex 2, Cowx Report, p. 19.
188 Vol . II, Annex 2, Cowx Report, p. 18. authorized roads, 193 and those authorized roads do not include the Road. 194In the
circumstances, there is no risk of adverse impact to the ecology of the River on
the basis that hazardous substances will be transported on the Road.
2.106. Secondly, Nicaragua claims the Road has adversely impact ed on tourism
in Nicaragua and claims compensation for such harm. This claim is entirely
unsupported by evidence of actual impact, and in any event lacks any sensible
legal foundation.
2.107. As to evidence, Nicaragua relies on the Go lder Report, arguing that the
area’s tourism potential has been “significantly impaired by the Road”, and given
that “tourism in the area ‘is mostly associated with the natural beauty of this
remote and non-highly commercialized region’, the only reasonabl e conclusion is
that Costa Rica’s project detrimentally impacts tourism in Nicaragua.” 195 The
Golder Report refers to the 2013 CCT Report, which identified “landscape
alteration” as an impact of the Road and recommended reforestation in areas
visible to the San Juan River. 196 The Golder Report does not directly address the
conclusion set out in the 2013 CCT Report that “[t]he effect of the construction of
193
Vol . IV, Annex 15, Costa Rica, Executive Decree No 24715 -MOPT-MEIC-S, 6 October
1995, published in the Official Gazette number 207, 1 November 1995, Article 39.
194
Route 1856 is not included in the list of authorized roads: see Vol . IV, Annex 70 ,
Department of Transit Engineering, Ministry of Public Works and Transportation, Costa
Rica, Authorization of Routes for the Transport of Hazardous Materials, 1995; Vol . IV,
Annex 76, Note from the Chief Engineer of the Department of Studies and Designs to the
Chief of the Department of Weights and Dimensions of the Consejo Nacional de Vialidad
(CONAVI) and to the Director General of the Department of Transit Police , Reference
DGIT-ED-4697-2014, 11 June 2014; and Vol . IV, Annex 77 , Internal Communication of
the Costa Rican General Department of Transit Engineering of the Ministry of Public
Works and Transportation, regarding the Authorization of Routes for the Transport of
Hazardous Materials, June 2014.
195 NR, para 2.94, citing NR, Annex 6, Golder Report, Section 7.
196 NR, Annex 6, Golder Report, pp. 42-43.
70authorized roads, 193and those authorized roads do not include the Road. 194 In the
circumstances, there is no risk of adverse impact to the ecology of the River on
the basis that hazardous substances will be transported on the Road.
2.106. Secondly, Nicaragua claims the Road has adversely impact ed on tourism
in Nicaragua and claims compensation for such harm. This claim is entirely
unsupported by evidence of actual impact, and in any event lacks any sensible
legal foundation.
2.107. As to evidence, Nicaragua relies on the Go lder Report, arguing that the
area’s tourism potential has been “significantly impaired by the Road”, and given
that “tourism in the area ‘is mostly associated with the natural beauty of this
remote and non-highly commercialized region’, the only reasonabl e conclusion is
that Costa Rica’s project detrimentally impacts tourism in Nicaragua.” 195 The
Golder Report refers to the 2013 CCT Report, which identified “landscape
alteration” as an impact of the Road and recommended reforestation in areas
visible to the San Juan River. 196 The Golder Report does not directly address the
conclusion set out in the 2013 CCT Report that “[t]he effect of the construction of
193
Vol . IV, Annex 15, Costa Rica, Executive Decree No 24715 -MOPT-MEIC-S, 6 October
1995, published in the Official Gazette number 207, 1 November 1995, Article 39.
194
Route 1856 is not included in the list of authorized roads: see Vol . IV, Annex 70 ,
Department of Transit Engineering, Ministry of Public Works and Transportation, Costa
Rica, Authorization of Routes for the Transport of Hazardous Materials, 1995; Vol . IV,
Annex 76, Note from the Chief Engineer of the Department of Studies and Designs to the
Chief of the Department of Weights and Dimensions of the Consejo Nacional de Vialidad
(CONAVI) and to the Director General of the Department of Transit Police , Reference
DGIT-ED-4697-2014, 11 June 2014; and Vol . IV, Annex 77 , Internal Communication of
the Costa Rican General Department of Transit Engineering of the Ministry of Public
Works and Transportation, regarding the Authorization of Routes for the Transport of
Hazardous Materials, June 2014.
195 NR, para 2.94, citing NR, Annex 6, Golder Report, Section 7.
196 NR, Annex 6, Golder Report, pp. 42-43. urgent need of remediation; 201 and (b) they are insufficient to prevent erosion into
the River, including on the basis that many of the seedlings planted to provide
ground cover have died. 202 Nicaragua claims that Costa Rica’s works have not
203
mitigated the alleged risk of significant harm to the River in the future.
2.111. Contrary to Nicaragua’s assertions based on observations from a distance,
Costa Rica’s remediation works have been effective in reducing erosion. These
ongoing works are detailed in the reports of the CONAVI Department of Costa
204
Rica’s Ministry of Public Works and Transportation, and CODEFORSA, the
Commission contracted to plant and maintain more than 50,000 trees at sites
205
along the Road. They include sites beyond the first 15 km and cover many of
206
the sites identified by Nicaragua’s experts as the most severely eroding.
Explanations of how these works have been effective and photographs
demonstrating that fact are provided with the Reports annexed to this Rejoinder.
These works include the following:
(a) surfacing the Road with gravel to stabilised and protect it from surface
erosion;
201 See NR, paras. 3.21-3.23.
202
See NR, paras. 3.24-3.28.
203
NR, para. 3.33.
204 Vol . III, Annex 11, CONAVI, Works on National Road N° 856: Before and After -
Updated as of December 2014, December 2014 (the 2014 CONAVI Report).
205 Vol . III, Annex 13, CODEFORSA, Consulting Services for the Development and
Implementation of an Environmental Plan for the Juan Rafael Mora Porras Border Road ,
2014 (the 2014 CODEFORSA Report ); and Vol . III, Annex 12, CODEFORSA,
Restoration and rehabilitation of ecosystems affected by the construction of the Juan
Rafael Mora Porras border road, Ruta 1856, November 2014 (the CODEFORSA
Quarterly Report for November 2014).
206 Appendix A, 2015 Thorne Report, paras. 7.20 and 7.26.
72urgent need of remediation; 201 and (b) they are insufficient to prevent erosion into
the River, including on the basis that many of the seedlings planted to provide
ground cover have died. 202 Nicaragua claims that Costa Rica’s works have not
203
mitigated the alleged risk of significant harm to the River in the future.
2.111. Contrary to Nicaragua’s assertions based on observations from a distance,
Costa Rica’s remediation works have been effective in reducing erosion. These
ongoing works are detailed in the reports of the CONAVI Department of Costa
204
Rica’s Ministry of Public Works and Transportation, and CODEFORSA, the
Commission contracted to plant and maintain more than 50,000 trees at sites
205
along the Road. They include sites beyond the first 15 km and cover many of
206
the sites identified by Nicaragua’s experts as the most severely eroding.
Explanations of how these works have been effective and photographs
demonstrating that fact are provided with the Reports annexed to this Rejoinder.
These works include the following:
(a) surfacing the Road with gravel to stabilised and protect it from surface
erosion;
201 See NR, paras. 3.21-3.23.
202
See NR, paras. 3.24-3.28.
203
NR, para. 3.33.
204 Vol . III, Annex 11, CONAVI, Works on National Road N° 856: Before and After -
Updated as of December 2014, December 2014 (the 2014 CONAVI Report).
205 Vol . III, Annex 13, CODEFORSA, Consulting Services for the Development and
Implementation of an Environmental Plan for the Juan Rafael Mora Porras Border Road ,
2014 (the 2014 CODEFORSA Report ); and Vol . III, Annex 12, CODEFORSA,
Restoration and rehabilitation of ecosystems affected by the construction of the Juan
Rafael Mora Porras border road, Ruta 1856, November 2014 (the CODEFORSA
Quarterly Report for November 2014).
206 Appendix A, 2015 Thorne Report, paras. 7.20 and 7.26. 208
Rejoinder. As Professor Thorne explains, the a reas selected for reforestation
(including gentle but not steep slopes) are locations where trees will be effective
in:
(a) reducing the erosivity of rainfall by intercepting precipitation;
(b) reducing the erodibility of the soil by decreas ing soil moisture levels
through evapotranspiration and by providing root reinforcement;
(c) reducing the generation of overland flow by increasing infiltration;
(d) intercepting surface runoff along concentrated flow paths by increasing
surface roughness and ground permeability, to protect the soil and
downslope areas from sheet, rill or gully erosion;
(e) intercepting surface runoff that might otherwise reach the Río San Juan;
and
(f) creating valuable wildlife habitat. 209
2.113. As Professor Thorne explains, based on his inspection of the Road:
“My impression of the Road gained in 2014 is not that erosion has ‘visibly
worsened’ (as Dr Kondolf states on page 11 of his 2014 Report) but, on
the contrary, that it has slowed. This is partly due to the nat ural recovery
of stability that follows disturbance of a landscape: the geomorphic ‘rate
law’ which predicts that rates of change decrease exponentially with time
since disturbance (Graf, 1977), but is also thanks to the concerted efforts
of CONAVI and COD EFORSA in mitigating erosion at multiple sites,
including those between the R ío Infiernito and Boca San Carlos and
208 See Vol . III, Annex 13, 2014 CODEFORSA Report; and Vol . III , Annex 12,
CODEFORSA Quarterly Report for 2014.
209
Appendix A, 2015 Thorne Report, para. 7.6.
74 208
Rejoinder. As Professor Thorne explains, the a reas selected for reforestation
(including gentle but not steep slopes) are locations where trees will be effective
in:
(a) reducing the erosivity of rainfall by intercepting precipitation;
(b) reducing the erodibility of the soil by decreas ing soil moisture levels
through evapotranspiration and by providing root reinforcement;
(c) reducing the generation of overland flow by increasing infiltration;
(d) intercepting surface runoff along concentrated flow paths by increasing
surface roughness and ground permeability, to protect the soil and
downslope areas from sheet, rill or gully erosion;
(e) intercepting surface runoff that might otherwise reach the Río San Juan;
and
(f) creating valuable wildlife habitat. 209
2.113. As Professor Thorne explains, based on his inspection of the Road:
“My impression of the Road gained in 2014 is not that erosion has ‘visibly
worsened’ (as Dr Kondolf states on page 11 of his 2014 Report) but, on
the contrary, that it has slowed. This is partly due to the nat ural recovery
of stability that follows disturbance of a landscape: the geomorphic ‘rate
law’ which predicts that rates of change decrease exponentially with time
since disturbance (Graf, 1977), but is also thanks to the concerted efforts
of CONAVI and COD EFORSA in mitigating erosion at multiple sites,
including those between the R ío Infiernito and Boca San Carlos and
208 See Vol . III, Annex 13, 2014 CODEFORSA Report; and Vol . III , Annex 12,
CODEFORSA Quarterly Report for 2014.
209
Appendix A, 2015 Thorne Report, para. 7.6. cause harm. Costa Rica’s evidence demonstrates that this is not the case , and
moreover, even accepting the estimates of contribution of sediment put forward by
Nicaragua’s experts, there is no adverse impact on the River, and there is no, and
never has been any, risk of significant harm. In particular:
(a) The sediment load carried by the San Juan in the period since construction
of the Road is actually lower than it was before the Road was constructed.
Hence, there is no evidence that construction of the Road has increased the
suspended sediment load carried by the San Juan.
(b) The field monitoring undertaken by Costa Rica’s experts on a w orst-case
scenario (including because the upper bound estimates of all erosion rates
are applied at all slopes along the full length of the Road) indicates that the
average input of sediment from the Road to the River is approximately
75,000 tonnes per year (a figure that is all the more conservative as it takes
no account of the mitigation works) . This represents less than 0.6% of the
total sediment load of the River, and is obviously too small a proportion to
have any significant or adverse impact on the River. If the highest estimate
of Road-derived sediment put forward by Nicaragua’s experts is accepted,
it would represent less than 3% of the total sediment load of the River,
which is also obviously too small a proportion to have any significant or
adverse effect on the River.
(c) At Delta Colorado, around 10% of the San Juan enters to the Lower San
Juan River. It is reasonable to assume that around 10% of the additional
sediment would enter the Lower San Juan. Even accepting the figures of
Road-derived sediment put forward by Nicaragua’s expert (which are
significantly overstated), and the proposition that all of it is dep osited in
76cause harm. Costa Rica’s evidence demonstrates that this is not the case , and
moreover, even accepting the estimates of contribution of sediment put forward by
Nicaragua’s experts, there is no adverse impact on the River, and there is no, and
never has been any, risk of significant harm. In particular:
(a) The sediment load carried by the San Juan in the period since construction
of the Road is actually lower than it was before the Road was constructed.
Hence, there is no evidence that construction of the Road has increased the
suspended sediment load carried by the San Juan.
(b) The field monitoring undertaken by Costa Rica’s experts on a w orst-case
scenario (including because the upper bound estimates of all erosion rates
are applied at all slopes along the full length of the Road) indicates that the
average input of sediment from the Road to the River is approximately
75,000 tonnes per year (a figure that is all the more conservative as it takes
no account of the mitigation works) . This represents less than 0.6% of the
total sediment load of the River, and is obviously too small a proportion to
have any significant or adverse impact on the River. If the highest estimate
of Road-derived sediment put forward by Nicaragua’s experts is accepted,
it would represent less than 3% of the total sediment load of the River,
which is also obviously too small a proportion to have any significant or
adverse effect on the River.
(c) At Delta Colorado, around 10% of the San Juan enters to the Lower San
Juan River. It is reasonable to assume that around 10% of the additional
sediment would enter the Lower San Juan. Even accepting the figures of
Road-derived sediment put forward by Nicaragua’s expert (which are
significantly overstated), and the proposition that all of it is dep osited in78 Chapter 3
Residual Legal Issues
A . Introduction
3.1. In Chapter 4 of its Reply, Nicaragua insists upon its position that the 1858
Treaty of Limits “puts the River under Nicarag ua’s sovereignty and, since the
construction of the Road causes serious harm to the River, it is, indeed, crucially
relevant for the present case”. 212 This matter was discussed at length in the
Counter-Memorial. The Treaty of Limits is indeed “crucial” for the relations
between the two countries. However, it is of no relevance to the present
213
proceedings. Costa Rica rejects the interpretation advanced by Nicaragua in its
Reply of the Treaty of Limits, and of the Court’s Judgment of 13 July 2009 in the
case concerning Navigational and Related Rights , for the reasons set out in this
chapter.
3.2. Nicaragua’s Reply acknowledges that Costa Rica is free to make its own
appraisal of its security and communicational needs and of the best means by
which to meet those needs within its territory. 214 It also concedes that this case is
not about the construction of a road on a State’s own territory, but about harm
purportedly caused to a neighbouring State as a result of such construction. 215
Insofar as there is no adverse impact on Nicaragua, the effect of these concessions
is that all the allegations that Nicaragua has made in its pleadings about the
212
NR, para. 4.5.
213 See Chapter 4 of CRCM.
214
NR, para. 4.2.
215 NR, para. 4.6.
79 purported deficiencies in construction work carried out on Costa Rican territory,
about the Emergency Decree which authorised the construction work, as well as
Nicaragua’s extravagant analysis of breaches by Costa Rica of Costa Rican
domestic law, must be disregarded.
3.3. Nicaragua has also asserted that an array of other rules are relevant to the
present proceedings . Indeed, it appears that the main reason for Nicaragua
including in its Memorial allegations premised on its interpretation of the Treaty
of Limits and the related arbitral awards, as well as the Court’s 2009 Judgment,
was to attempt to justify the claim that it is entitled to prevent the exercise of
Costa Rica’s perpetual right of free navigation on the San Juan River as a counter-
216
measure. Nicaragua has not pursued this claim in its Reply , following Costa
217
Rica’s response to it in the Counter -Memorial. Nicaragua also appears to have
abandoned other accusations such as the colourful claim that there was an
218
“invasion of Nicaraguan territory”.
3.4. However, Nicaragua maintains in its Reply that the ( inexistent) harm to the
San Juan River that it alleges also constitute s a breach of both its territorial
sovereignty, and the 1858 Treaty of Limits , including the judicial and arbitral
interpretations thereof. Equally, Nicaragua persists in alleging breaches to its right
of navigation (of which no evidence has been presented after two rounds of
written pleadings) and the Costa Rican obligation to conduct and to notify an
Environment Impact Assessment (“EIA”). This chapter will also discuss these
residual legal issues.
216 NM, para. 4.9 and 6.36.
217
CRCM, paras. 6.24 to 6.25. See also para. 4.4.
218 NM, p. 129, para. 4.13.
80purported deficiencies in construction work carried out on Costa Rican territory,
about the Emergency Decree which authorised the construction work, as well as
Nicaragua’s extravagant analysis of breaches by Costa Rica of Costa Rican
domestic law, must be disregarded.
3.3. Nicaragua has also asserted that an array of other rules are relevant to the
present proceedings . Indeed, it appears that the main reason for Nicaragua
including in its Memorial allegations premised on its interpretation of the Treaty
of Limits and the related arbitral awards, as well as the Court’s 2009 Judgment,
was to attempt to justify the claim that it is entitled to prevent the exercise of
Costa Rica’s perpetual right of free navigation on the San Juan River as a counter-
216
measure. Nicaragua has not pursued this claim in its Reply , following Costa
217
Rica’s response to it in the Counter -Memorial. Nicaragua also appears to have
abandoned other accusations such as the colourful claim that there was an
218
“invasion of Nicaraguan territory”.
3.4. However, Nicaragua maintains in its Reply that the ( inexistent) harm to the
San Juan River that it alleges also constitute s a breach of both its territorial
sovereignty, and the 1858 Treaty of Limits , including the judicial and arbitral
interpretations thereof. Equally, Nicaragua persists in alleging breaches to its right
of navigation (of which no evidence has been presented after two rounds of
written pleadings) and the Costa Rican obligation to conduct and to notify an
Environment Impact Assessment (“EIA”). This chapter will also discuss these
residual legal issues.
216 NM, para. 4.9 and 6.36.
217
CRCM, paras. 6.24 to 6.25. See also para. 4.4.
218 NM, p. 129, para. 4.13. “A simple reading of Article VI shows that the Parties did not intend to
establish any hierarchy as between Nicaragua’s sovereignty over the river
and Costa Rica’s right of free navigation, characterized as “perpetual”, with
each of these affirmations counter -balancing the other. Nicaragua’s
sovereignty is affirmed only to the extent that it does not prejudice the
substance of Costa Rica’s right of free navigation in its domain, the
establishment of which is precisely the point at issue ; the right of free
navigation, albeit “perpetual”, is granted only on condition that it does not
prejudice the key prerogatives of territorial sovereignty.” 223
There is no need to further discuss these points here, since they are not relevant
for this case.
3.7. Nicaragua does not claim that the 1858 Treaty of Limits imposes specific
limitations on works Costa Rica may plan or implement on its territory. The Reply
is more indirect than that. Nicaragua asserts that “the Treaty puts the Riv er under
Nicaragua’s sovereignty and, since the construction of the Road causes serious
harm to the River, it is, indeed, crucially relevant for the present case”. 224 There
has been no serious attempt to rebut what Costa Rica made plain in its Counter-
Memorial: that Nicaragua’s sovereignty over the waters of the San Juan River and
the exercise of this sovereignty are not put in issue by any of the Costa Rican
conduct complained of, even assuming the existence of the alleged harm to those
waters (quod non).
223
Dispute regarding Navigational and Related Rights (Costa Rica v . Nicaragua), Judgment,
I .C .J . Reports 2009, p. 237, para. 48.
224 NR, para. 4.5.
82 “A simple reading of Article VI shows that the Parties did not intend to
establish any hierarchy as between Nicaragua’s sovereignty over the river
and Costa Rica’s right of free navigation, characterized as “perpetual”, with
each of these affirmations counter -balancing the other. Nicaragua’s
sovereignty is affirmed only to the extent that it does not prejudice the
substance of Costa Rica’s right of free navigation in its domain, the
establishment of which is precisely the point at issue ; the right of free
navigation, albeit “perpetual”, is granted only on condition that it does not
prejudice the key prerogatives of territorial sovereignty.” 223
There is no need to further discuss these points here, since they are not relevant
for this case.
3.7. Nicaragua does not claim that the 1858 Treaty of Limits imposes specific
limitations on works Costa Rica may plan or implement on its territory. The Reply
is more indirect than that. Nicaragua asserts that “the Treaty puts the Riv er under
Nicaragua’s sovereignty and, since the construction of the Road causes serious
harm to the River, it is, indeed, crucially relevant for the present case”. 224 There
has been no serious attempt to rebut what Costa Rica made plain in its Counter-
Memorial: that Nicaragua’s sovereignty over the waters of the San Juan River and
the exercise of this sovereignty are not put in issue by any of the Costa Rican
conduct complained of, even assuming the existence of the alleged harm to those
waters (quod non).
223
Dispute regarding Navigational and Related Rights (Costa Rica v . Nicaragua), Judgment,
I .C .J . Reports 2009, p. 237, para. 48.
224 NR, para. 4.5. 3.10. The Reply sets out, again, the same assertions that Nicaragua advanced in
the Memorial, without addressing the arguments made by Costa Rica in the
Counter-Memorial. After citing once again the celebrated arbitral award of Max
Huber in the Island of Palmas (Miangas) case, Nicaragua repeats, quoting once
again from the well known J udgment of the Permanent Court of International
Justice in the Lotus case, that “[a]s a consequence, a State ‘may not exercise its
power in any form in the territory of another State.’” 230 No effort is made to rebut
Costa Rica’s position, which is in any event obvious on the face of the evidence
and pleadings in this case, that there has not been any exercise of Costa Rican
power or authority in Ni caraguan territory, whether this be the waters of the San
Juan River, Nicaraguan airspace or land.
3.11. All that Nicaragua can summon in support of its contention that there have
been “violations to its territorial sovereignty” is the argument that Costa Rica, by
constructing the road, is “voluntarily discharging” sediment in the San Juan River
and “changing the configuration of the river”. 231 In its Counter-Memorial, Costa
Rica explained that even if significant harm were caused to the river as a result of
the construction of the road ( quod non), this would not have constituted a breach
of the obligation to respect the territorial sovereignty of other States, but rather a
breach of the obligation not to cause significant transboundary harm. 232These are
two quite different obligations. Nicaragua has failed to address the significance of
this distinction in general, and in relation to this case in particular.
230 NR, para. 5.12.
231
NR, paras. 5.16-5.17.
232 CRCM, para. 4.9.
843.10. The Reply sets out, again, the same assertions that Nicaragua advanced in
the Memorial, without addressing the arguments made by Costa Rica in the
Counter-Memorial. After citing once again the celebrated arbitral award of Max
Huber in the Island of Palmas (Miangas) case, Nicaragua repeats, quoting once
again from the well known J udgment of the Permanent Court of International
Justice in the Lotus case, that “[a]s a consequence, a State ‘may not exercise its
power in any form in the territory of another State.’” 230 No effort is made to rebut
Costa Rica’s position, which is in any event obvious on the face of the evidence
and pleadings in this case, that there has not been any exercise of Costa Rican
power or authority in Ni caraguan territory, whether this be the waters of the San
Juan River, Nicaraguan airspace or land.
3.11. All that Nicaragua can summon in support of its contention that there have
been “violations to its territorial sovereignty” is the argument that Costa Rica, by
constructing the road, is “voluntarily discharging” sediment in the San Juan River
and “changing the configuration of the river”. 231 In its Counter-Memorial, Costa
Rica explained that even if significant harm were caused to the river as a result of
the construction of the road ( quod non), this would not have constituted a breach
of the obligation to respect the territorial sovereignty of other States, but rather a
breach of the obligation not to cause significant transboundary harm. 232These are
two quite different obligations. Nicaragua has failed to address the significance of
this distinction in general, and in relation to this case in particular.
230 NR, para. 5.12.
231
NR, paras. 5.16-5.17.
232 CRCM, para. 4.9. naturally found by the action of weather or erosion. Causes of erosion are
manifold. Some of them are anthropogenic; by walking along the bank of a River
erosion can occur. But this does not render the soil any less natural a substance, or
the process by which it r eached the River deliberate. Further, as appears from
Chapter 2 above, the fact that sediment enters the River is not unusual or a cause
for concern; fluvial processes are the most common way of transporting sediment,
and the San Juan River is a particular ly sediment-heavy river. There may be
environmental concerns about certain human activities, a point discussed above in
238
this Rejoinder. However, this situation does not in any way amount to a ‘breach
of the territorial sovereignty and integrity’ of States , or a breach of the 1858
Treaty of Limits.
3.14. If Nicaragua’s hypothesis was considered correct, then the sediment that
was dumped into the San Juan River as a result of the construction by Nicaragua
of the Santa Fe bridge would be tantamount to a violation of Costa Rica’s
sovereignty and integrity, as that sediment entered Costa Rica’s Colorado River.
3.15. Shifting its focus regarding its previous general claim about impairment of
239
its right of navigation , Nicaragua further contends “that the formation of
numerous ‘very visible’ and ‘massive’ deltas, resulting from the construction of
the Road changes the very configuration of the River”. 240 A number of points are
made in response to this ludicrous exaggeration. First, there is a dispute between
the parties about whether deltas in the San Juan River are the result of the
construction of the Road. It is undisputed that deltas are present on both banks of
238 See para. 2.70.
239
NM, para. 4.15.
240 NR, para. 5.19 (footnote omitted).
86naturally found by the action of weather or erosion. Causes of erosion are
manifold. Some of them are anthropogenic; by walking along the bank of a River
erosion can occur. But this does not render the soil any less natural a substance, or
the process by which it r eached the River deliberate. Further, as appears from
Chapter 2 above, the fact that sediment enters the River is not unusual or a cause
for concern; fluvial processes are the most common way of transporting sediment,
and the San Juan River is a particular ly sediment-heavy river. There may be
environmental concerns about certain human activities, a point discussed above in
238
this Rejoinder. However, this situation does not in any way amount to a ‘breach
of the territorial sovereignty and integrity’ of States , or a breach of the 1858
Treaty of Limits.
3.14. If Nicaragua’s hypothesis was considered correct, then the sediment that
was dumped into the San Juan River as a result of the construction by Nicaragua
of the Santa Fe bridge would be tantamount to a violation of Costa Rica’s
sovereignty and integrity, as that sediment entered Costa Rica’s Colorado River.
3.15. Shifting its focus regarding its previous general claim about impairment of
239
its right of navigation , Nicaragua further contends “that the formation of
numerous ‘very visible’ and ‘massive’ deltas, resulting from the construction of
the Road changes the very configuration of the River”. 240 A number of points are
made in response to this ludicrous exaggeration. First, there is a dispute between
the parties about whether deltas in the San Juan River are the result of the
construction of the Road. It is undisputed that deltas are present on both banks of
238 See para. 2.70.
239
NM, para. 4.15.
240 NR, para. 5.19 (footnote omitted). 245
of the 2009 Judgment, omitting once again paragraph 93. Nicaragua’s position
is that, since the Court acknowledged a right to regulate navigation on the San
Juan River and the corresponding obligation of Nicaragua to notify such
regulations to Costa Rica, the same obligation of notification would apply to
Costa Rica’s construction of the Road, even if there is no specific provision of this
in the Treaty of Limits.
3.17. The only reasoning supporting this assertion is that the Court’s analysis
would apply mutatis mutandis to the present case, in which there would be not
just regulations but concrete acts, which would be detrimental to the navigation
and the sovereignty of Nicaragua over the waters of the San Juan, “guaranteed by
Article VI of the 1858 Treaty”. 246 Costa Rica has already responded to these
arguments. 247 In summary, the obligation to notify Costa Rica about regulations
concerning navigation on the San Juan River arises by virtue of Costa Rica’s right
of navigation on the River, as established by Article VI of the Treaty of Limits.
There is no Nicaraguan right of any sort pertaining to Costa Rican territory. There
is an obligation of consultation in Article VIII of the Treaty of Limits, which is of
no relevance to the present case because it applies in the event that Nicaragua
envisages canalisation works. None of the factors mentioned by the Court to
determine the existence of the Nicaraguan obligation to notify navigational
regulations apply to the construction of the Road.
3.18. Nicaragua considers that this alleged Costa Rican obligation to notify
245 NR, para. 5.28.
246
NR, para. 5.29.
247 CRCM, paras. 4.13-4.17.
88 245
of the 2009 Judgment, omitting once again paragraph 93. Nicaragua’s position
is that, since the Court acknowledged a right to regulate navigation on the San
Juan River and the corresponding obligation of Nicaragua to notify such
regulations to Costa Rica, the same obligation of notification would apply to
Costa Rica’s construction of the Road, even if there is no specific provision of this
in the Treaty of Limits.
3.17. The only reasoning supporting this assertion is that the Court’s analysis
would apply mutatis mutandis to the present case, in which there would be not
just regulations but concrete acts, which would be detrimental to the navigation
and the sovereignty of Nicaragua over the waters of the San Juan, “guaranteed by
Article VI of the 1858 Treaty”. 246 Costa Rica has already responded to these
arguments. 247 In summary, the obligation to notify Costa Rica about regulations
concerning navigation on the San Juan River arises by virtue of Costa Rica’s right
of navigation on the River, as established by Article VI of the Treaty of Limits.
There is no Nicaraguan right of any sort pertaining to Costa Rican territory. There
is an obligation of consultation in Article VIII of the Treaty of Limits, which is of
no relevance to the present case because it applies in the event that Nicaragua
envisages canalisation works. None of the factors mentioned by the Court to
determine the existence of the Nicaraguan obligation to notify navigational
regulations apply to the construction of the Road.
3.18. Nicaragua considers that this alleged Costa Rican obligation to notify
245 NR, para. 5.28.
246
NR, para. 5.29.
247 CRCM, paras. 4.13-4.17. 253
Consular Staff in Tehran” , as evoked by Nicaragua . Leaving these matters to
one side, this section will address the following points: a) The threshold required
in respect of the obligation to conduct an EIA; b) the emergency exemption under
international law ; c) the emergency situation in the present case; and d) the
existence of an alternative assessment.
(1) The threshold required for an EIA
3.21. The threshold requirement for an EIA issignificant transboundary harm or
impact. The present case concerns the harm that is alleged to have been caused to
Nicaraguan territory by reason of the construction of the Road, and nothing else.
Any analysis by Nicaragua of harm allegedly caused by Costa Rica to Costa Rican
territory is outside the scope of any Nicaraguan claim at the international level and
does not fall within the case that is put before the Court in these proceedings . The
same applies to the new allegations in the Reply concerning different causes of
sedimentation of the San Juan River allegedly originat ing from Costa Rican
territory.
3.22. It is common ground between the parties that the international obligation
to conduct an EIA only arises where there is a risk of significant adverse
transboundary impact or harm . What the parties do not agree on is whether this
threshold was met in the present case.
3.23. As Costa Rica has previously stated , the construction of the Road did not
and does not lead to the discharge of harmful substances or emissions into the San
253
NR, para. 6.18.
90 253
Consular Staff in Tehran” , as evoked by Nicaragua . Leaving these matters to
one side, this section will address the following points: a) The threshold required
in respect of the obligation to conduct an EIA; b) the emergency exemption under
international law ; c) the emergency situation in the present case; and d) the
existence of an alternative assessment.
(1) The threshold required for an EIA
3.21. The threshold requirement for an EIA issignificant transboundary harm or
impact. The present case concerns the harm that is alleged to have been caused to
Nicaraguan territory by reason of the construction of the Road, and nothing else.
Any analysis by Nicaragua of harm allegedly caused by Costa Rica to Costa Rican
territory is outside the scope of any Nicaraguan claim at the international level and
does not fall within the case that is put before the Court in these proceedings . The
same applies to the new allegations in the Reply concerning different causes of
sedimentation of the San Juan River allegedly originat ing from Costa Rican
territory.
3.22. It is common ground between the parties that the international obligation
to conduct an EIA only arises where there is a risk of significant adverse
transboundary impact or harm . What the parties do not agree on is whether this
threshold was met in the present case.
3.23. As Costa Rica has previously stated , the construction of the Road did not
and does not lead to the discharge of harmful substances or emissions into the San
253
NR, para. 6.18. would require an EIA. 258A similar approach is used in the European Community
EIA Directive, 259 and is acknowledged as an appropriate mechanism in the
Guidelines for Biodiversity Inclusive Impact Assessment under the CBD. 260 In
relation to the Espoo list (Appendix 1), Nicaragua notes in its Memorial that the
list includes, “motorways” and “express roads”, as well as “deforestation of large
areas”. 261 However, to be clear, the definition of the road- related terms in
262
Appendix 1 would e xclude a road, such as the one built by Costa Rica.
Evidently, this is not a highway or an “express road”. T he limited clearing
activities associated with the R oad would not amount to “deforestation of large
areas”. In other words, were the Espoo Conventi on to apply – which it does not –
263
Costa Rica would not owe an obligation to conduct an EIA under that treaty.
3.26. As stated above, alleged harm to the ecosystem within Costa Rica is not a
matter to be discussed here, since it would not be transboundary, unle ss there
would be an adverse impact on Nicaraguan territory. It is for Nicaragua to
demonstrate such impact, or at least its likelihood in order to meet the threshold.
But this proof is lacking.
3.27. The only possible environmental consequence of the construct ion of the
Road on the San Juan River is an alleged increase in sediment load. However, the
258
Convention on Environmental Impact Assessment in a Transboundary Context ( Espoo
Convention), Finland, 25 February 1991, Article 2(3) and Appendix 1.
259
European Commision, EIA Directive, Article 4.
260 Conference of the Parties to the C onvention of B iological Diversity (CBD), Decision
VIII/28, “Impact Assessment: Voluntary guidelines on biodiversity -inclusive impact
assessment”, UN Doc. UNEP/CBD/COP/8/31, (CBD EIA Guidelines) Art. 10.
261 NM, footnote 474.
262 Convention on Environmental Impact Assessment in a Transboundary Context ( Espoo
Convention), Finland, 25 February 1991, Appendix 1.
263
Ibid, Article 2(5).
92would require an EIA. 258A similar approach is used in the European Community
EIA Directive, 259 and is acknowledged as an appropriate mechanism in the
Guidelines for Biodiversity Inclusive Impact Assessment under the CBD. 260 In
relation to the Espoo list (Appendix 1), Nicaragua notes in its Memorial that the
list includes, “motorways” and “express roads”, as well as “deforestation of large
areas”. 261 However, to be clear, the definition of the road- related terms in
262
Appendix 1 would e xclude a road, such as the one built by Costa Rica.
Evidently, this is not a highway or an “express road”. T he limited clearing
activities associated with the R oad would not amount to “deforestation of large
areas”. In other words, were the Espoo Conventi on to apply – which it does not –
263
Costa Rica would not owe an obligation to conduct an EIA under that treaty.
3.26. As stated above, alleged harm to the ecosystem within Costa Rica is not a
matter to be discussed here, since it would not be transboundary, unle ss there
would be an adverse impact on Nicaraguan territory. It is for Nicaragua to
demonstrate such impact, or at least its likelihood in order to meet the threshold.
But this proof is lacking.
3.27. The only possible environmental consequence of the construct ion of the
Road on the San Juan River is an alleged increase in sediment load. However, the
258
Convention on Environmental Impact Assessment in a Transboundary Context ( Espoo
Convention), Finland, 25 February 1991, Article 2(3) and Appendix 1.
259
European Commision, EIA Directive, Article 4.
260 Conference of the Parties to the C onvention of B iological Diversity (CBD), Decision
VIII/28, “Impact Assessment: Voluntary guidelines on biodiversity -inclusive impact
assessment”, UN Doc. UNEP/CBD/COP/8/31, (CBD EIA Guidelines) Art. 10.
261 NM, footnote 474.
262 Convention on Environmental Impact Assessment in a Transboundary Context ( Espoo
Convention), Finland, 25 February 1991, Appendix 1.
263
Ibid, Article 2(5). and/or (3) activities that would “result in changes to the access to, and/or rights
268
over biological resources”. These factors are noted in the Report by
Dr W. Sheate annexed to the Reply , which indicates the relevance of a table
269
incorporating these criteria from the CBD Guidelines. But again, the evidence
does not directly address these criteria. Nicaragua has not produced any technical
comparison showing the conditions of habitats or ecosystems in the San Juan river
before and after the construction of the Road, even though Costa Rica requested
Nicaragua evidence in this regard when Nicaragua started publicly complaining
about road construction. 270
3.29. Nicaragua has also denied access to Costa Rican technical experts to carry
271
out measurements on the sediment load carried by the River. The
Environmental Diagnostic Assessment submitted by Costa Rica in CRCM, as well
as the new Follow -up Study, and reports submitted in this Rejoinder have not
disclosed any impacts of the nature that would indicate significant transboundary
adverse impacts in accordance with the identified factors. 272 Notably, if there
would have been serious impacts, they would have taken place in Costa Rican
territory, but evidence shows that impacts were small and localized, are being
mitigated, and that there is certainly no transboundary effect. 273
268 Conference of the Parties to the CBD, Decision VIII/28, UN Doc. UNEP/CBD/COP/8/31,
(CBD EIA Guidelines), section 1.2(a) “Screening”.
269
NR, Vol . II, Annex 5, Sheate Report, p.10.
270 See CRCM, Vol . III, Annex 39, Note from the Minister of Foreign Affairs and Worship
of Costa Rica to the Minister of Foreign Affairs of Nicaragua, Ref DM -601-11,
29 November 2011.
271 See paras. 2.28-2.33 above.
272
CRCM, Vol . III, Annex 10, 2013 CCT Report; and this Rejoinder, Vol . I II, Annex 14,
2015 CCT Report.
273 Vol . III, Annex 14, 2015 CCT Report, p. 83.
94and/or (3) activities that would “result in changes to the access to, and/or rights
268
over biological resources”. These factors are noted in the Report by
Dr W. Sheate annexed to the Reply , which indicates the relevance of a table
269
incorporating these criteria from the CBD Guidelines. But again, the evidence
does not directly address these criteria. Nicaragua has not produced any technical
comparison showing the conditions of habitats or ecosystems in the San Juan river
before and after the construction of the Road, even though Costa Rica requested
Nicaragua evidence in this regard when Nicaragua started publicly complaining
about road construction. 270
3.29. Nicaragua has also denied access to Costa Rican technical experts to carry
271
out measurements on the sediment load carried by the River. The
Environmental Diagnostic Assessment submitted by Costa Rica in CRCM, as well
as the new Follow -up Study, and reports submitted in this Rejoinder have not
disclosed any impacts of the nature that would indicate significant transboundary
adverse impacts in accordance with the identified factors. 272 Notably, if there
would have been serious impacts, they would have taken place in Costa Rican
territory, but evidence shows that impacts were small and localized, are being
mitigated, and that there is certainly no transboundary effect. 273
268 Conference of the Parties to the CBD, Decision VIII/28, UN Doc. UNEP/CBD/COP/8/31,
(CBD EIA Guidelines), section 1.2(a) “Screening”.
269
NR, Vol . II, Annex 5, Sheate Report, p.10.
270 See CRCM, Vol . III, Annex 39, Note from the Minister of Foreign Affairs and Worship
of Costa Rica to the Minister of Foreign Affairs of Nicaragua, Ref DM -601-11,
29 November 2011.
271 See paras. 2.28-2.33 above.
272
CRCM, Vol . III, Annex 10, 2013 CCT Report; and this Rejoinder, Vol . I II, Annex 14,
2015 CCT Report.
273 Vol . III, Annex 14, 2015 CCT Report, p. 83. methodology that allowed it to make its finding in 2010 is apt to determine the
different situation that has to be addressed in the instant case, as Professor Craik’s
275
report states.
3.33. It was not until recently, following the clarification by the Court in the
Pulp Mills case, that the international obligation to produce an EIA in certain
circumstances was determined. The Court came to its conclusion with regard to
the obligation to conduct an EIA on the basis of a practice that “gained in recent
years so much acceptance”. 276 Consequently, the question whether international
law recognises that it is a matter of domestic law whether an emergency situation
is an exception to the obligation to produce an EIA is an issue that must be
approached in a similar manner, namely by determining the existence of
international practice in support.
3.34. International practice demonstrates that the emergency exception has also
“gained … much acceptance”, to use the words of the Court. Both international
and domestic instruments recognise the ability of S tates to exempt specific
projects under particular circumstances, namely activities undertaken in relation to
national security and civil emergencies. In particular:
(1) The Antarctic EIA regime contained in Annex 1 to the Protocol on
275 See Vol . II, Annex 1, Craik Report, para. 2.11.
276
“In this sense, the obligation to protect and preserve, under Article 41 (a) of the Statute
[of the River Uruguay] has to be interpreted in accordance with a practice, which in
recent years has gained so much acceptance that it may now be considered a requirement
under general international law to undertake an environmental impact assessment where
there is a risk that the proposed industrial activity may have a significant adverse impact
in a transboundary co ntext, in particular, on a shared resource.” Pulp Mills on the River
Uruguay (Argentina v . Uruguay), J udgment of 20 April 2010, ICJ Reports 2010, p. 83,
para. 204.
96methodology that allowed it to make its finding in 2010 is apt to determine the
different situation that has to be addressed in the instant case, as Professor Craik’s
275
report states.
3.33. It was not until recently, following the clarification by the Court in the
Pulp Mills case, that the international obligation to produce an EIA in certain
circumstances was determined. The Court came to its conclusion with regard to
the obligation to conduct an EIA on the basis of a practice that “gained in recent
years so much acceptance”. 276 Consequently, the question whether international
law recognises that it is a matter of domestic law whether an emergency situation
is an exception to the obligation to produce an EIA is an issue that must be
approached in a similar manner, namely by determining the existence of
international practice in support.
3.34. International practice demonstrates that the emergency exception has also
“gained … much acceptance”, to use the words of the Court. Both international
and domestic instruments recognise the ability of S tates to exempt specific
projects under particular circumstances, namely activities undertaken in relation to
national security and civil emergencies. In particular:
(1) The Antarctic EIA regime contained in Annex 1 to the Protocol on
275 See Vol . II, Annex 1, Craik Report, para. 2.11.
276
“In this sense, the obligation to protect and preserve, under Article 41 (a) of the Statute
[of the River Uruguay] has to be interpreted in accordance with a practice, which in
recent years has gained so much acceptance that it may now be considered a requirement
under general international law to undertake an environmental impact assessment where
there is a risk that the proposed industrial activity may have a significant adverse impact
in a transboundary co ntext, in particular, on a shared resource.” Pulp Mills on the River
Uruguay (Argentina v . Uruguay), J udgment of 20 April 2010, ICJ Reports 2010, p. 83,
para. 204. (4) Article 4(6) of the Draft Protocol on Environmental Impact Assessment in
a Transboundary Context to the Framework Convention for the Protection
of the Marine Environment of the Caspian Sea reserves the rights of the
Parties to implement laws in the “interests of national security”. 280
(5) Article 1(3) of the EC EIA Directive provides that states may decide on a
case-by-case basis to exempt the Directive from “projects or parts of
projects, having defence as their sole purpose, or to projects having the
281
response to civil emergencies as their sole purpose”.
(6) The United States federal EIA statute, the National Environmental Policy
Act, (the first EIA statute to be enacted globally), contains in its
regulations (40 C.F.R. 1506.11), a provision entitled “Emergencies”,
which states:
“Where emergency circumstances make it necessary to take an action
with significant environmental impact without observing the
provisions of these regulations, the Federal agency taking the action
should consult with the Council about alternative arrangement s.
Agencies and the Council will limit such arrangements to actions
280 Protocol on Environmental Impact Assessment in a Transboundary Context to the
Framework Convention for the Protection of the Marine Environment of the Caspian Sea,
Draft prepared for Conference of the Parties, Fifth Meeting, 28 -30 May 2014, UN Doc.
TC/COP5/4 Rev.1, Art. 4(6).
281 European Commission, E nvironmental Impact Assessment (EIA) Directive
(85/337/EEC), 1985, available at http://ec.europa.eu/environment/eia-legalcontext.htm.
98(4) Article 4(6) of the Draft Protocol on Environmental Impact Assessment in
a Transboundary Context to the Framework Convention for the Protection
of the Marine Environment of the Caspian Sea reserves the rights of the
Parties to implement laws in the “interests of national security”. 280
(5) Article 1(3) of the EC EIA Directive provides that states may decide on a
case-by-case basis to exempt the Directive from “projects or parts of
projects, having defence as their sole purpose, or to projects having the
281
response to civil emergencies as their sole purpose”.
(6) The United States federal EIA statute, the National Environmental Policy
Act, (the first EIA statute to be enacted globally), contains in its
regulations (40 C.F.R. 1506.11), a provision entitled “Emergencies”,
which states:
“Where emergency circumstances make it necessary to take an action
with significant environmental impact without observing the
provisions of these regulations, the Federal agency taking the action
should consult with the Council about alternative arrangement s.
Agencies and the Council will limit such arrangements to actions
280 Protocol on Environmental Impact Assessment in a Transboundary Context to the
Framework Convention for the Protection of the Marine Environment of the Caspian Sea,
Draft prepared for Conference of the Parties, Fifth Meeting, 28 -30 May 2014, UN Doc.
TC/COP5/4 Rev.1, Art. 4(6).
281 European Commission, E nvironmental Impact Assessment (EIA) Directive
(85/337/EEC), 1985, available at http://ec.europa.eu/environment/eia/-legalcontext.htm. international and national instruments, the exemption must be understood as a
standard part of the EIA process. The international obligation should, therefore, be
interpreted to be subject ed to the right of S tates to incorporate an emergency
exemption under domestic law, as Professor Craik equally concludes in his
Report. 291
3.36. While the specific exemption originates in domestic law, in the present
case through Costa Rica’ s Emergency Decree, Nicaragua is incorrect in
characterizing the exemption as an internal law used to excuse an international
obligation, as it is the international obligation itself that includes the right of
States to exempt activities under conditions o f civil emergencies and national
security concerns.
3.37. The presence of an emergency exemption under international law is
consistent with the obligation of S tates to use due diligence, which was also
another element employed by the Court to reach its conclus ion of the existence of
292
an obligation to conduct an EIA. In the case of an emergency, it is not
“reasonable” nor within the degree of care “expected of a good Government” to
require a State to delay urgent activities in order to conduct an EIA. The
exemption is also consistent with the deference that international law provides to
States to determine the contents of their EIA instruments, as it leaves it up to
291
Vol . II, Annex 1, Craik Report, para. 5.3 pp. 17-18.
292 “In this sense, the obligation to protect and preserve, under Article 41 (a) of the Statute
[of the River Uruguay] has to be interpreted in accordance with a practice, which in
recent years has gained so much acceptance that it may now be considered a requirement
under general international law to undertake an environmental impact assessment where
there is a risk that the proposed industrial activity may have a significant adverse impact
in a transboundary context , in particular, on a shared resource.” Pulp Mills on the River
Uruguay (Argentina v . Uruguay), J udgment of 20 April 2010, ICJ Reports 2010, p. 83,
para. 204.
100international and national instruments, the exemption must be understood as a
standard part of the EIA process. The international obligation should, therefore, be
interpreted to be subject ed to the right of S tates to incorporate an emergency
exemption under domestic law, as Professor Craik equally concludes in his
Report. 291
3.36. While the specific exemption originates in domestic law, in the present
case through Costa Rica’ s Emergency Decree, Nicaragua is incorrect in
characterizing the exemption as an internal law used to excuse an international
obligation, as it is the international obligation itself that includes the right of
States to exempt activities under conditions o f civil emergencies and national
security concerns.
3.37. The presence of an emergency exemption under international law is
consistent with the obligation of S tates to use due diligence, which was also
another element employed by the Court to reach its conclus ion of the existence of
292
an obligation to conduct an EIA. In the case of an emergency, it is not
“reasonable” nor within the degree of care “expected of a good Government” to
require a State to delay urgent activities in order to conduct an EIA. The
exemption is also consistent with the deference that international law provides to
States to determine the contents of their EIA instruments, as it leaves it up to
291
Vol . II, Annex 1, Craik Report, para. 5.3 pp. 17-18.
292 “In this sense, the obligation to protect and preserve, under Article 41 (a) of the Statute
[of the River Uruguay] has to be interpreted in accordance with a practice, which in
recent years has gained so much acceptance that it may now be considered a requirement
under general international law to undertake an environmental impact assessment where
there is a risk that the proposed industrial activity may have a significant adverse impact
in a transboundary context , in particular, on a shared resource.” Pulp Mills on the River
Uruguay (Argentina v . Uruguay), J udgment of 20 April 2010, ICJ Reports 2010, p. 83,
para. 204. territory occurred in the northern part of Isla Portillos that Costa Rica must
295
confine its reaction to defend itself spatially to that area. Further, whether the
project of the Border Road was completed or not in a short period of time is also
296
immaterial to the declaration of a situation of emergency. The Reply also
digresses into consideration about whether the measure s that Costa Rica took to
297
respond to the emergency w ere appropriate. The evaluation of the
appropriateness of the measures adopted is a purely domestic matter that only
concerns Costa Rica. The only valid concern Nicaragua may raise is significant
transboundary harm. As seen above, there is no basis for such a claim.
3.41. What is relevant in order to evaluate the construction of the Border R oad
in a situation of emergency is whether , at the time the decision was taken , an
emergency situation was in existence. The Emergency Decree was not issued two
months after the construction work on the Border Road commenced, as Nicaragua
contends. 298 In December 2010, a matter of weeks after the first Nicaraguan
occupation of the northern part of Isla Portillos, the improvement of existing dirt
roads began in order to allow access with and between the b order posts of Delta
Costa Rica and Boca Sarapiquí. It was in that context that the Costa Rica n
Minister of Public Security considered that, in order to e nsure adequate access to
the border posts and to facilitate the mobilisation of its citizens in case o f the
299
aggravation of the situation, the construction of a road was necessary.
Consequently, the Costa Rican Government issued the Emergency Decree on
295
NR, para. 6.14-6.15.
296 NR, para. 6.13.
297 NR, para. 6.16-6.17.
298
NR, para. 6.12.
299 CRCM, paras. 2.25-2.27.
102territory occurred in the northern part of Isla Portillos that Costa Rica must
295
confine its reaction to defend itself spatially to that area. Further, whether the
project of the Border Road was completed or not in a short period of time is also
296
immaterial to the declaration of a situation of emergency. The Reply also
digresses into consideration about whether the measure s that Costa Rica took to
297
respond to the emergency w ere appropriate. The evaluation of the
appropriateness of the measures adopted is a purely domestic matter that only
concerns Costa Rica. The only valid concern Nicaragua may raise is significant
transboundary harm. As seen above, there is no basis for such a claim.
3.41. What is relevant in order to evaluate the construction of the Border R oad
in a situation of emergency is whether , at the time the decision was taken , an
emergency situation was in existence. The Emergency Decree was not issued two
months after the construction work on the Border Road commenced, as Nicaragua
contends. 298 In December 2010, a matter of weeks after the first Nicaraguan
occupation of the northern part of Isla Portillos, the improvement of existing dirt
roads began in order to allow access with and between the b order posts of Delta
Costa Rica and Boca Sarapiquí. It was in that context that the Costa Rica n
Minister of Public Security considered that, in order to e nsure adequate access to
the border posts and to facilitate the mobilisation of its citizens in case o f the
299
aggravation of the situation, the construction of a road was necessary.
Consequently, the Costa Rican Government issued the Emergency Decree on
295
NR, para. 6.14-6.15.
296 NR, para. 6.13.
297 NR, para. 6.16-6.17.
298
NR, para. 6.12.
299 CRCM, paras. 2.25-2.27. 304
these issues.
3.44. These threats and rejections of the 1858 Treaty occurred at a time in which
Nicaragua also persistently prevented or otherwise impeded the exercise of Costa
Rican rights of navigation on the San Juan River, including the enactment of a
discriminatory decree in contravention of the Court’s J udgment in the
305
Navigational and Related Rights case. Nicaragua’s numerous violations of
Costa Rica’s right to navigate the Sa n Juan River are well documented in
correspondence between the parties, 306 affidavits, 307 and press reports. 308 These
304
See Vol . IV Annex 59 , Press Release of 26 October 1976 and Minutes of Liberia
meeting, 25 January 1977, in: Ministry of Foreign Affairs and Worship of Costa Rica
referring to the initiation of discussions of a maritime boundary in the Pacific Ocean,
Annual Report 1976- 1977, Vol. I, pp. 156- 160 see also Vol . IV, Annex 60 , Minutes of
the First Meeting of the Sub-Commission on Limits and Cartography, 7 November 2002.
305 See CRCM, Vol III, Annex 26, Nicaraguan Decree N° 79-2009 is titled “Creation of the
Inter-Institutional Commission to Develop and Implement the Regulations regarding
Navigation on the San Juan River, specifically where the International Court of Justice
Grants Limited Navigation Rights to the Republic of Costa Rica”.
306
See Vol . IV, Annex 16, Note from the Minister of Foreign Affairs and Worship of Costa
Rica to the Minis ter of Foreign Affairs of Nicaragua, Reference DM- 543-09, 27 July
2009; Vol . IV, Annex 17, Note from the Acting Minister of Foreign Affairs and Worship
of Costa Rica, to the Acting Minister of Minister of Foreign Affairs of Nicaragua,
Reference DVM-176-09, 21 August 2009; Vol . IV, Annex 18, Note from the Minister of
Foreign Affairs and Worship of Costa Rica to the Minister of Foreign Affairs of
Nicaragua, Reference DM-674-09, 7 September 2009; CRCM Vol . III, A nnex 34, Note
from the Minister of Foreign Affai rs and Worship of Costa Rica to the Minister of
Foreign Affairs of Nicaragua, Reference DM-AM-816-09, including annex entitled Annex
to Note DM- AM-816-09 . Decree N° 79 -2009 of the President of the Republic of
Nicaragua contravenes the judgment of the Inter national Court of Justice of 13 July
2009, 20 November 2009; CRCM Vol . III A nnex 35, Note from the Minister of Foreign
Affairs of Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica,
Reference MRE/DM-AJST/297/3/2010, 25 March 2010,; See Vol . IV, Annex 71, Note
from the Chief of Post, Police Delegation of Sarapiquí, Costa Rica, to the Regional
Director of the Fourth Region -Heredia, Ref. 1571 -2010-DPS 27 September 2010; CRM
Certain Activities Case, Vol . III, Annex 62, Note from the Ministry of Foreign Affairs
and Worship to the Ministry of Foreign Affairs of Nicaragua, 24 November 2010,
Reference DM-478-10; CRM Certain Activities Case, Vol . III, Annex 84, Note from the
acting Minister of Foreign Affairs of Nicaragua to the Minister of Foreign Affairs and
Worship of Costa Rica , Reference MRE/DVM/AJST/121/04/11, 8 April 2011; CRM
104 304
these issues.
3.44. These threats and rejections of the 1858 Treaty occurred at a time in which
Nicaragua also persistently prevented or otherwise impeded the exercise of Costa
Rican rights of navigation on the San Juan River, including the enactment of a
discriminatory decree in contravention of the Court’s J udgment in the
305
Navigational and Related Rights case. Nicaragua’s numerous violations of
Costa Rica’s right to navigate the Sa n Juan River are well documented in
correspondence between the parties, 306 affidavits,307 and press reports. 308 These
304
See Vol . IV Annex 59 , Press Release of 26 October 1976 and Minutes of Liberia
meeting, 25 January 1977, in: Ministry of Foreign Affairs and Worship of Costa Rica
referring to the initiation of discussions of a maritime boundary in the Pacific Ocean,
Annual Report 1976- 1977, Vol. I, pp. 156- 160 see also Vol . IV, Annex 60 , Minutes of
the First Meeting of the Sub-Commission on Limits and Cartography, 7 November 2002.
305 See CRCM, Vol III, Annex 26, Nicaraguan Decree N° 79-2009 is titled “Creation of the
Inter-Institutional Commission to Develop and Implement the Regulations regarding
Navigation on the San Juan River, specifically where the International Court of Justice
Grants Limited Navigation Rights to the Republic of Costa Rica”.
306
See Vol . IV, Annex 16, Note from the Minister of Foreign Affairs and Worship of Costa
Rica to the Minis ter of Foreign Affairs of Nicaragua, Reference DM- 543-09, 27 July
2009; Vol . IV, Annex 17, Note from the Acting Minister of Foreign Affairs and Worship
of Costa Rica, to the Acting Minister of Minister of Foreign Affairs of Nicaragua,
Reference DVM-176-09, 21 August 2009; Vol . IV, Annex 18, Note from the Minister of
Foreign Affairs and Worship of Costa Rica to the Minister of Foreign Affairs of
Nicaragua, Reference DM-674-09, 7 September 2009; CRCM Vol . III, A nnex 34, Note
from the Minister of Foreign Affai rs and Worship of Costa Rica to the Minister of
Foreign Affairs of Nicaragua, Reference DM-AM-816-09, including annex entitled Annex
to Note DM- AM-816-09 . Decree N° 79 -2009 of the President of the Republic of
Nicaragua contravenes the judgment of the Inter national Court of Justice of 13 July
2009, 20 November 2009; CRCM Vol . III A nnex 35, Note from the Minister of Foreign
Affairs of Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica,
Reference MRE/DM-AJST/297/3/2010, 25 March 2010,; See Vol . IV, Annex 71, Note
from the Chief of Post, Police Delegation of Sarapiquí, Costa Rica, to the Regional
Director of the Fourth Region -Heredia, Ref. 1571 -2010-DPS 27 September 2010; CRM
Certain Activities Case, Vol . III, Annex 62, Note from the Ministry of Foreign Affairs
and Worship to the Ministry of Foreign Affairs of Nicaragua, 24 November 2010,
Reference DM-478-10; CRM Certain Activities Case, Vol . III, Annex 84, Note from the
acting Minister of Foreign Affairs of Nicaragua to the Minister of Forei gn Affairs and
Worship of Costa Rica , Reference MRE/DVM/AJST/121/04/11, 8 April 2011; CRM demonstrates the seriousness of Costa Rica’s concerns regarding Nicaraguan
conduct in the border area.
3.46. Furthermore, Costa Rica’s well rooted concerns have again been
confirmed by the recent Nicaraguan rejection not only of Costa Rican navigational
rights, but also of a practical arrangement informed by both parties to the Court in
Certain Activities case, that would have allowed Costa Rica to fill the last caños
309
constructed by Nicaragua in violation of the Court’s Order of 8 March 2011.
309
See Vol . IV, Annex 32, Note from the Acting Minister of Foreign Affairs and Worship
of Costa Rica to the Minister of Foreign Affairs of Nicaragua, reference DM- AM-348-14,
17 July 2014; Vol . IV , Annex 34 , Note from the Minister of Foreign Affairs of
Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica, Reference
MRE/DM/336/8/14, 4 August 2014; Vol . IV, Annex 36, Note from the Minister of
Foreign Affairs of Nicaragua to the Min ister of Foreign Affairs and Worship of Costa
Rica, Reference MRE/DM-AJ/414/09/19, 19 September 2014; Vol . IV, Annex 37, Note
from the Minister of Foreign Affairs and Worship of Costa Rica to the Minister of
Foreign Affairs of Nicaragua, Reference DM -AM-574-14, 22 September 2014; Vol . IV,
Annex 38, Note from the Agent of Nicaragua to the Registrar of the International Court
of Justice, reference HOL-EMB-124, 23 September 2014; Vol . IV, Annex 39, Note from
the Co -Agent of Costa Rica to the Registrar of the I nternational Court of Justice,
reference ECRPB -103-14, 25 September 2014; Vol . IV, Annex 47, Note from the
Minister of Foreign Affairs and Worship of Costa Rica to the Minister of Foreign Affairs
of Nicaragua, Reference DM -AM-707-14, 7 November 2014; Vol . IV, Annex 49, Note
from the Minister of Foreign Affairs of Nicaragua to the Minister of Foreign Affairs and
Worship of Costa Rica, Reference MRE/DM/DGAJST/456/ 11/14, 11 November 2014;
Vol . IV, Annex 50, Note from the Minister of Foreign Affairs and Worship of Costa Rica
to the Minister of Foreign Affairs of Nicaragua, Reference DM -AM-718-14, 14
November 2014; Vol . IV, Annex 51, Note from the Minister of Foreign Affairs of
Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica, Reference
MRE/DM/677/12/14, 2 December 2014; Vol . IV, Annex 52, Note from the Minister of
Foreign Affairs and Worship of Costa Rica to the Minister of Foreign Affairs of
Nicaragua, Reference DM-AM-774-14 2 December 2014; Vol . IV, Annex 53, Note from
the Acting Minister of Foreign Affairs and Worship of Costa Rica to the Minister of
Foreign Affairs of Nicaragua, Reference DM -AM-789, 4 December 2014; Vol . IV,
Annex 54, Note from the Minister of Foreign Affairs of Nicaragua to the Minister of
Foreign Affairs and Worship of Costa Rica, Reference MRE/DM -AJ/478/12/14,
5 December 2014; Vol . IV, Annex 55, Note from the Minister of Foreign Affairs and
Worship of Costa Rica to the Minister of Foreign Affairs of Nicaragua, Reference DM -
AM-0818-14, 12 December 2014; Vol . IV, Annex 56 , Note from the Minister of Foreign
Affairs of Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica,
Reference MRE/DM-AJ/482/12/14, 15 December 2014; Vol . IV, Annex 57, Note from
106demonstrates the seriousness of Costa Rica’s concerns regarding Nicaraguan
conduct in the border area.
3.46. Furthermore, Costa Rica’s well rooted concerns have again been
confirmed by the recent Nicaraguan rejection not only of Costa Rican navigational
rights, but also of a practical arrangement informed by both parties to the Court in
Certain Activities case, that would have allowed Costa Rica to fill the last caños
309
constructed by Nicaragua in violation of the Court’s Order of 8 March 2011.
309
See Vol . IV, Annex 32, Note from the Acting Minister of Foreign Affairs and Worship
of Costa Rica to the Minister of Foreign Affairs of Nicaragua, reference DM- AM-348-14,
17 July 2014; Vol . IV , Annex 34 , Note from the Minister of Foreign Affairs of
Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica, Reference
MRE/DM/336/8/14, 4 August 2014; Vol . IV, Annex 36, Note from the Minister of
Foreign Affairs of Nicaragua to the Min ister of Foreign Affairs and Worship of Costa
Rica, Reference MRE/DM-AJ/414/09/19, 19 September 2014; Vol . IV, Annex 37, Note
from the Minister of Foreign Affairs and Worship of Costa Rica to the Minister of
Foreign Affairs of Nicaragua, Reference DM -AM-574-14, 22 September 2014; Vol . IV,
Annex 38, Note from the Agent of Nicaragua to the Registrar of the International Court
of Justice, reference HOL-EMB-124, 23 September 2014; Vol . IV, Annex 39, Note from
the Co -Agent of Costa Rica to the Registrar of the I nternational Court of Justice,
reference ECRPB -103-14, 25 September 2014; Vol . IV, Annex 47, Note from the
Minister of Foreign Affairs and Worship of Costa Rica to the Minister of Foreign Affairs
of Nicaragua, Reference DM -AM-707-14, 7 November 2014; Vol . IV, Annex 49, Note
from the Minister of Foreign Affairs of Nicaragua to the Minister of Foreign Affairs and
Worship of Costa Rica, Reference MRE/DM/DGAJST/456/ 11/14, 11 November 2014;
Vol . IV, Annex 50, Note from the Minister of Foreign Affairs and Worship of Costa Rica
to the Minister of Foreign Affairs of Nicaragua, Reference DM -AM-718-14, 14
November 2014; Vol . IV, Annex 51, Note from the Minister of Foreign Affairs of
Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica, Reference
MRE/DM/677/12/14, 2 December 2014; Vol . IV, Annex 52, Note from the Minister of
Foreign Affairs and Worship of Costa Rica to the Minister of Foreign Affairs of
Nicaragua, Reference DM-AM-774-14 2 December 2014; Vol . IV, Annex 53, Note from
the Acting Minister of Foreign Affairs and Worship of Costa Rica to the Minister of
Foreign Affairs of Nicaragua, Reference DM -AM-789, 4 December 2014; Vol . IV,
Annex 54, Note from the Minister of Foreign Affairs of Nicaragua to the Minister of
Foreign Affairs and Worship of Costa Rica, Reference MRE/DM -AJ/478/12/14,
5 December 2014; Vol . IV, Annex 55, Note from the Minister of Foreign Affairs and
Worship of Costa Rica to the Minister of Foreign Affairs of Nicaragua, Reference DM -
AM-0818-14, 12 December 2014; Vol . IV, Annex 56 , Note from the Minister of Foreign
Affairs of Nicaragua to the Minister of Foreign Affairs and Worship of Costa Rica,
Reference MRE/DM-AJ/482/12/14, 15 December 2014; Vol . IV, Annex 57, Note from Costa Rican Government commissioned an Environmental Diagnostic,
which was carried out by a team of experts from the Tropical Science
Center, a well-respected Costa Rican organization established in 1962. The
Center has extensive experience in scientific environmental research in
areas subject to tropical conditions, including environmental impact
assessments. The Environmental Diagnostic covers the entire 108 km of the
Road in the vicinity of the San Juan River, from Boundary Marker 2 to
Delta Colorado. It considers the existing physical environment where the
Road is constructed, including the climate, hydrology, terrestrial and aquatic
flora and fauna, and ecology. It incorporates recommendations for the work
to complete the Road, taking account of any potential risk of environmental
impact. The Environmental Diagnostic fully complies with the guidelines
established by Costa Rican administrative regulations for a project of this
type.”313
3.49. Contrary to what Nicaragua contends, 314the EDA fulfils the functions of
an EIA with regard to works that have not yet been implemented. To insist upon a
“fresh EIA” that Costa Rica should carry out is not only legally wrong , it is a
formalistic approach without any concrete practical distinction and fundamentally
it is materially impossible to carry out, given the “ex ante” character of an EIA.
*
* *
3.50. In summary, this section has shown that the threshold for an EIA with regr ad
to the type of road constructed in Costa Rica along the border has not been met, and
that in any event, in accordance with the emergency situation declared by the Costa
Rican Government, it was not possible to conduct an EIA. Once the works
commenced anEDA was conducted, and the continuing planning and monitoring of
the work on the Road takes into account the environmental concerns raised by
Nicaragua in this case in order to avoid any harm being caused to Nicaragua.
313 CRCM, para. 2.35.
314
NR, para. 6.38 and 6.55.
108 Costa Rican Government commissioned an Environmental Diagnostic,
which was carried out by a team of experts from the Tropical Science
Center, a well-respected Costa Rican organization established in 1962. The
Center has extensive experience in scientific environmental research in
areas subject to tropical conditions, including environmental impact
assessments. The Environmental Diagnostic covers the entire 108 km of the
Road in the vicinity of the San Juan River, from Boundary Marker 2 to
Delta Colorado. It considers the existing physical environment where the
Road is constructed, including the climate, hydrology, terrestrial and aquatic
flora and fauna, and ecology. It incorporates recommendations for the work
to complete the Road, taking account of any potential risk of environmental
impact. The Environmental Diagnostic fully complies with the guidelines
established by Costa Rican administrative regulations for a project of this
type.” 313
3.49. Contrary to what Nicaragua contends, 314the EDA fulfils the functions of
an EIA with regard to works that have not yet been implemented. To insist upon a
“fresh EIA” that Costa Rica should carry out is not only legally wrong , it is a
formalistic approach without any concrete practical distinction and fundamentally
it is materially impossible to carry out, given the “ex ante” character of an EIA.
*
* *
3.50. In summary, this section has shown that the threshold for an EIA with regr ad
to the type of road constructed in Costa Rica along the border has not been met, and
that in any event, in accordance with the emergency situation declared by the Costa
Rican Government, it was not possible to conduct an EIA. Once the works
commenced anEDA was conducted, and the continuing planning and monitoring of
the work on the Road takes into account the environmental concerns raised by
Nicaragua in this case in order to avoid any harm being caused to Nicaragua.
313 CRCM, para. 2.35.
314
NR, para. 6.38 and 6.55. E . Notification
3.53. In the Reply, Nicaragua maintains that Costa Rica breached its obligation
to notify Nicaragua about the construction of the R oad on the basis of general
international law or conventional instruments. This allegation ignores the fact that
the threshold for the application of this obligation of notification was not met, as
well as the fact of the emergency situation under which the Road was constructed,
as stated above. 317Nicaragua further contends that Costa Rica’s declaration of
emergency was not communicated to Nicaragua, and that there were no
consultations and negotiations with Nicaragua. 318 These assertions give rise to
two comments:
(1) First, it is unprecedented for State A to request State B to communicate to
State A not only its declaration of an emergency, but also the measures it
plans to take in response, in circumstances where the emergency has arisen
by reason of State A’s actions, and the measures State B plans to take are
for the purpose of defending itself against State A.
(2) Second, Nicaragua has inaccurately presented the relevant correspondence
between the two countries in its Reply.
3.54. It is not correct that Costa Rica refused to consult with Nicaragua. By note
dated 29 November 2011 from the Minister of Foreign Affairs of Costa Rica t o
his Nicaraguan counterpart, which is reproduced in Nicaragua’s Application as
317 See above paras. 3.21 to 3.27 and paras. 3.40 to 3.47.
318
NR, para. 6.63.
110E . Notification
3.53. In the Reply, Nicaragua maintains that Costa Rica breached its obligation
to notify Nicaragua about the construction of the R oad on the basis of general
international law or conventional instruments. This allegation ignores the fact that
the threshold for the application of this obligation of notification was not met, as
well as the fact of the emergency situation under which the Road was constructed,
as stated above. 317 Nicaragua further contends that Costa Rica’s declaration of
emergency was not communicated to Nicaragua, and that there were no
consultations and negotiations with Nicaragua. 318 These assertions give rise to
two comments:
(1) First, it is unprecedented for State A to request State B to communicate to
State A not only its declaration of an emergency, but also the measures it
plans to take in response, in circumstances where the emergency has arisen
by reason of State A’s actions, and the measures State B plans to take are
for the purpose of defending itself against State A.
(2) Second, Nicaragua has inaccurately presented the relevant correspondence
between the two countries in its Reply.
3.54. It is not correct that Costa Rica refused to consult with Nicaragua. By note
dated 29 November 2011 from the Minister of Foreign Affairs of Costa Rica t o
his Nicaraguan counterpart, which is reproduced in Nicaragua’s Application as
317 See above paras. 3.21 to 3.27 and paras. 3.40 to 3.47.
318
NR, para. 6.63. 3.56. All attempts made by Costa Rica, from the beginning of the dispute until
recently, in order to undertake a joint monitoring of the waters of the San Juan
321
River, have failed. This is a further confirmation that the environmental
concerns raised by Nicaragua are not the real reason for instituting these
proceedings. On the other hand, it shows Costa Rica ’s willingness to address the
issue in a scientific and constructive manner.
3.57. Furthermore, although Nicaragua did not communicate any EIA to Costa
Rica with regard to the dredging and other works on the San Juan R iver, it now
claims a breach by Costa Rica of the same obligation to produce an EIA for the
construction of a two -car carriageway on Costa Rican territory. Nicaragua’s
dredging work, and its attempts to modify the course of the River (not to mention
the construction of artificial caños on Costa Rican territory and the planned
construction of an Interoceanic Canal) are able to affect Costa Rican navigational
rights and produce significant environmental harm within Costa Rica, particularly
with regard to the water flow from the San Juan to the Colorado River, the latter
being an entirely Costa Rican river. Failure to comply with its own prior
obligation to notify Costa Rica in these circumstances also prevents Nicaragu a
from invoking any alleged lack of notification by Costa Rica.
F . Alleged Breaches of other Treaties
3.58. In its Reply, Nicaragua insists that the construction of the Road entails
violations of the Convention on Biological Diversity ( “CBD”), the Ramsar
Convention, the Central American Convention for the Protection of the
Environment and other regional instruments, as well as the bilateral “SI -A-PAZ”
321 See paras. 2.28-2.33.
1123.56. All attempts made by Costa Rica, from the beginning of the dispute until
recently, in order to undertake a joint monitoring of the waters of the San Juan
321
River, have failed. This is a further confirmation that the environmental
concerns raised by Nicaragua are not the real reason for instituting these
proceedings. On the other hand, it shows Costa Rica ’s willingness to address the
issue in a scientific and constructive manner.
3.57. Furthermore, although Nicaragua did not communicate any EIA to Costa
Rica with regard to the dredging and other works on the San Juan R iver, it now
claims a breach by Costa Rica of the same obligation to produce an EIA for the
construction of a two -car carriageway on Costa Rican territory. Nicaragua’s
dredging work, and its attempts to modify the course of the River (not to mention
the construction of artificial caños on Costa Rican territory and the planned
construction of an Interoceanic Canal) are able to affect Costa Rican navigational
rights and produce significant environmental harm within Costa Rica, particularly
with regard to the water flow from the San Juan to the Colorado River, the latter
being an entirely Costa Rican river. Failure to comply with its own prior
obligation to notify Costa Rica in these circumstances also prevents Nicaragu a
from invoking any alleged lack of notification by Costa Rica.
F . Alleged Breaches of other Treaties
3.58. In its Reply, Nicaragua insists that the construction of the Road entails
violations of the Convention on Biological Diversity ( “CBD”), the Ramsar
Convention, the Central American Convention for the Protection of the
Environment and other regional instruments, as well as the bilateral “SI -A-PAZ”
321 See paras. 2.28-2.33. 3.61. Nicaragua does not deny that the condition of reciprocity applies for the
obligation of notification, exchange of information and consultation set out in
Article 14 (1) of the C BD. 328 This condition of reciprocity is not exclusive to the
CBD; it also applies at the general level. In this case, and as early as 29 November
2011, Costa Rica invited Nicaragua to present objective evidence, and to engage
in bilateral dialogue, to review its concerns about the construction of the Road. As
explained above, Nicaragua failed to comply vis -à-vis Costa Rica with its
obligations of notification, exchange of information and consultation with regard
to the dredging, deviation of waters and other major works in the San Juan
329
River. Consequently, the condition of reciprocity is absent.
3.62. In its attempt at justifying the lack of reciprocity, Nicaragua provides an
explanation, set out in full as follows:
“Nicaragua’s dredging was undisputedly conduct ed within its sovereign
territory, the San Juan de Nicaragua River, and could not possibly have
any appreciable effect, on biological diversity or otherwise, in Costa Rica;
and Nicaragua’s cleaning of the caño was conducted in what Nicaragua
believed, and continues to believe, is also part of its sovereign territory and
would thus not affect Costa Rica; Nicaragua has shown that in any event
these activities did not, in fact, cause harm to Costa Rica. Therefore,
reciprocity cannot possibly be found to be lacking.”
3.63. Nicaragua thus employs the same arguments that it strongly criticises
when used by Costa Rica. In both cases (putting aside for the purpose of this
analysis the most important issue of the Certain Activities case: the occupation by
Nicaragua of th e Costa Rican territory) , it is a matter of proof whether the
activities carried out in the respective territories may cause significant
328 NR, para. 6.111.
329 See para. 3.57 above.
1143.61. Nicaragua does not deny that the condition of reciprocity applies for the
obligation of notification, exchange of information and consultation set out in
Article 14 (1) of the C BD. 328 This condition of reciprocity is not exclusive to the
CBD; it also applies at the general level. In this case, and as early as 29 November
2011, Costa Rica invited Nicaragua to present objective evidence, and to engage
in bilateral dialogue, to review its concerns about the construction of the Road. As
explained above, Nicaragua failed to comply vis -à-vis Costa Rica with its
obligations of notification, exchange of information and consultation with regard
to the dredging, deviation of waters and other major works in the San Juan
329
River. Consequently, the condition of reciprocity is absent.
3.62. In its attempt at justifying the lack of reciprocity, Nicaragua provides an
explanation, set out in full as follows:
“Nicaragua’s dredging was undisputedly conduct ed within its sovereign
territory, the San Juan de Nicaragua River, and could not possibly have
any appreciable effect, on biological diversity or otherwise, in Costa Rica;
and Nicaragua’s cleaning of the caño was conducted in what Nicaragua
believed, and continues to believe, is also part of its sovereign territory and
would thus not affect Costa Rica; Nicaragua has shown that in any event
these activities did not, in fact, cause harm to Costa Rica. Therefore,
reciprocity cannot possibly be found to be lacking.”
3.63. Nicaragua thus employs the same arguments that it strongly criticises
when used by Costa Rica. In both cases (putting aside for the purpose of this
analysis the most important issue of the Certain Activities case: the occupation by
Nicaragua of th e Costa Rican territory) , it is a matter of proof whether the
activities carried out in the respective territories may cause significant
328 NR, para. 6.111.
329 See para. 3.57 above. the construction of the Road does not concern wetlands extending over the
territories of both countries. Also, what is at issue in this case is not works related
to a water system shared by two States. Indeed, this case is not related to any
hydraulic endeavour that could affect any water system, but to a road constructed
entirely on Costa Rican territory . The mere fact of proximity to a river does not
render Article 5 of the Ramsar Convention applicable.
3.65. In its analysis of regional treaties, the Reply relies upon the so- called
“judgment” of the Central American Court of Justice of 21 June 2012. 332 As
explained in the Counter-Memorial and also in Chapter 2 above, Costa Rica is not
a party to the Statute of that Court, hence, its alleged “judgment” has no validity at
all.333
3.66. With regard to the regional agreements themselves, the Reply adopts the
same approach as the Memorial of advancing general cons iderations about the
supposedly “grave doubts” Nicaragua has about “Costa Rica’s commitment to the
objectives of these agreements”, and suggesting that the construction of the
334
Border Road would be contrary to the object and purpose of them. Costa Rica
335
has already set out its position on these issues in the Counter-Memorial.
3.67. The only new elements included in the Reply with regard to regional
agreements are the references to “anthropogenic sediment” as “pollution” and the
fact that rains “have brought in sufficiently moored culverts into the river”. Both
332 NR, paras. 6.117.
333 CRCM, para. 3.70, this Rejoinder, paras. 2.115-2.116.
334
NR, para. 6.127.
335 CRCM, paras. 5.34-5.39.
116the construction of the Road does not concern wetlands extending over the
territories of both countries. Also, what is at issue in this case is not works related
to a water system shared by two States. Indeed, this case is not related to any
hydraulic endeavour that could affect any water system, but to a road constructed
entirely on Costa Rican territory . The mere fact of proximity to a river does not
render Article 5 of the Ramsar Convention applicable.
3.65. In its analysis of regional treaties, the Reply relies upon the so- called
“judgment” of the Central American Court of Justice of 21 June 2012. 332 As
explained in the Counter-Memorial and also in Chapter 2 above, Costa Rica is not
a party to the Statute of that Court, hence, its alleged “judgment” has no validity at
all.333
3.66. With regard to the regional agreements themselves, the Reply adopts the
same approach as the Memorial of advancing general cons iderations about the
supposedly “grave doubts” Nicaragua has about “Costa Rica’s commitment to the
objectives of these agreements”, and suggesting that the construction of the
334
Border Road would be contrary to the object and purpose of them. Costa Rica
335
has already set out its position on these issues in the Counter-Memorial.
3.67. The only new elements included in the Reply with regard to regional
agreements are the references to “anthropogenic sediment” as “pollution” and the
fact that rains “have brought in sufficiently moored culverts into the river”. Both
332 NR, paras. 6.117.
333 CRCM, para. 3.70, this Rejoinder, paras. 2.115-2.116.
334
NR, para. 6.127.
335 CRCM, paras. 5.34-5.39. to notify Nicaragua about the construction of a Border R oad on Costa
Rican territory;
(c) The threshold for the international requirement of an EIA has not been met
in the present case;
(d) General international law acknowledges the exemption from an EIA in
situations of emergency;
(e) The emergency declared by Costa Rica as a consequence of the continued
rejection by Nicaragua of the 1858 Treaty of Limits, including the military
occupation of part of its territory by Nicaragua, falls within such
exemption;
(f) It was Nicaragua itself that created that emergency situation;
(g) The EDA in any event addresses the environmental transboundary
concerns for works already completed, the monitoring of these works, as
well as future infrastructure work , and thereby fulfils any applicable
international obligations;
(h) Costa Rica has not breached any obligation of notification, exchange of
information and consul tation. On the contrary, it invited Nicaragua to
engage in such a procedure, with no concrete results;
(i) In any event, Nicaragua failed to fulfil the condition of reciprocity with
regard to the production of an EIA, notification, exchange of information
and consultation while dredging the San Juan R iver, deviating its waters,
constructing caños on Costa Rican territory , amongst other works, which
are activities that occurred before or at the time of the situation of
emergency leading to the construction of the Road.
(j) None of the alleged violations of international obligations stemming from
any of the treaties referred toby Nicaragua are grounded in fact or in law.
118 to notify Nicaragua about the construction of a Border R oad on Costa
Rican territory;
(c) The threshold for the international requirement of an EIA has not been met
in the present case;
(d) General international law acknowledges the exemption from an EIA in
situations of emergency;
4.1.
(e) The emergency declared by Costa Rica as a consequence of the continued
of reparation, namely:
rejection by Nicaragua of the 1858 Treaty of Limits, including the military
occupation of part of its territory by Nicaragua, falls within such
exemption;
(f) It was Nicaragua itself that created that emergency situation;
(g) The EDA in any event addresses the environmental transboundary
concerns for works already completed, the monitoring of these works, as
well as future infrastructure work , and thereby fulfils any applicable
international obligations;
(h) Costa Rica has not breached any obligation of notification, exchange of
information and consul tation. On the contrary, it invited Nicaragua to
engage in such a procedure, with no concrete results;
(i) In any event, Nicaragua failed to fulfil the condition of reciprocity with
regard to the production of an EIA, notification, exchange of information
and consultation while dredging the San Juan R iver, deviating its waters, 4.2.
constructing caños on Costa Rican territory , amongst other works, which
341
are activities that occurred before or at the time of the situation of
342
emergency leading to the construction of the Road. 343
(j) None of the alleged violations of international obligations stemming from 344
345
any of the treaties referred toby Nicaragua are grounded in fact or in law.
346 347
(a) a declaration of unlawful conduct and responsibility;
(b) an order requiring Costa Rica to cease its unlawful conduct, and
prohibiting Costa Rica from undertaking any further development in
the area without an appropriate transboundary EIA; 348and
(c) an order requiring Costa Rica to restore the status quo ante , which it
349
presented together with a claim for compensation.
4.3. It also included new claims for reparation in its Memorial, namely:
(a) an order requiring Costa Rica to provide assurances and guarantees of
non-repetition; 350
(b) a declaration that Nicaragua is entitled to suspend Costa Rica’s right
of free navigation on the San Juan; 351
(c) a declaration that Nicaragua is entitled to execute works to improve
352
navigation on the San Juan River “as it deems suitable”, and to re-
347
NM, paras. 6.10 to 6.12.
348 Ibid, paras. 6.13 to 6.17.
349 Ibid, paras. 6.26 to 6.34.
350
Ibid, paras. 6.18 to 6.25.
351
Ibid, paras. 6.35 to 6.44.
352 Ibid, Submission 3(i).
120 347
(a) a declaration of unlawful conduct and responsibility;
(b) an order requiring Costa Rica to cease its unlawful conduct, and
prohibiting Costa Rica from undertaking any further development in
the area without an appropriate transboundary EIA; 348 and
(c) an order requiring Costa Rica to restore the status quo ante , which it
349
presented together with a claim for compensation.
4.3. It also included new claims for reparation in its Memorial, namely:
(a) an order requiring Costa Rica to provide assurances and guarantees of
non-repetition; 350
(b) a declaration that Nicaragua is entitled to suspend Costa Rica’s right
of free navigation on the San Juan; 351
(c) a declaration that Nicaragua is entitled to execute works to improve
352
navigation on the San Juan River “as it deems suitable”, and to re-
347
NM, paras. 6.10 to 6.12.
348 Ibid, paras. 6.13 to 6.17.
349 Ibid, paras. 6.26 to 6.34.
350
Ibid, paras. 6.18 to 6.25.
351
Ibid, paras. 6.35 to 6.44.
352 Ibid, Submission 3(i). B . Nicaragua’s claim for re-establishment of the status quo ante
4.5. Nicaragua’s claim for re- establishment of the status quo ante 361 is very
confused. It is also groundless, it requires the Court to make an order of no utility,
and is internally inconsistent with its claim for compensation.
4.6. As to the confused nature of the claim, this has evolved from a request for
an order requiring Costa Rica to “restore the status quo ante”, 362to a claim for the
re-establishment of the status quo ante “as far as possible” 363 or “as proximate as
possible”. 364 At the same time, Nicaragua now says that it “does not claim a
complete re-establishment of the status quo ante, which would lead to a complete
destruction of the road”, 365 and that it “does not challenge the right of Costa Rica
to build whatever road it deems useful on its territory.” 366
4.7. It follows that the true position is that Nicaragua doe s not claim, and
recognises that it is not entitled to claim, re -establishment of the status quo ante ,
and that even the qualifiers such as “as far as possible” are inapposite to the relief
really sought under this head, which is ultimately that Costa Rica carry out the
mitigation works according to recomme ndations made by experts appointed by
367
Nicaragua, including as to re-routing of sections of the Road.
361 NR, paras. 7.7 to, 7.10 and Submissions, para. 2(ii).
362
NM, para. 6.31.
363 NR, para. 7.7.
364 NR, para. 7.8.
365
NR, para. 7.8.
366 NR, para. 7.19.
367 NR, para. 7.10.
122 B . Nicaragua’s claim for re-establishment of the status quo ante
4.5. Nicaragua’s claim for re- establishment of the status quo ante 361 is very
confused. It is also groundless, it requires the Court to make an order of no utility,
and is internally inconsistent with its claim for compensation.
4.6. As to the confused nature of the claim, this has evolved from a request for
an order requiring Costa Rica to “restore the status quo ante”, 362to a claim for the
re-establishment of the status quo ante “as far as possible” 363or “as proximate as
possible”. 364 At the same time, Nicaragua now says that it “does not claim a
complete re-establishment of the status quo ante, which would lead to a complete
destruction of the road”, 365 and that it “does not challenge the right of Costa Rica
to build whatever road it deems useful on its territory.” 366
4.7. It follows that the true position is that Nicaragua doe s not claim, and
recognises that it is not entitled to claim, re -establishment of the status quo ante ,
and that even the qualifiers such as “as far as possible” are inapposite to the relief
really sought under this head, which is ultimately that Costa Rica carry out the
mitigation works according to recomme ndations made by experts appointed by
367
Nicaragua, including as to re-routing of sections of the Road.
361 NR, paras. 7.7 to, 7.10 and Submissions, para. 2(ii).
362
NM, para. 6.31.
363 NR, para. 7.7.
364 NR, para. 7.8.
365
NR, para. 7.8.
366 NR, para. 7.19.
367 NR, para. 7.10. showing of actionable damage. The water continues to flow from the south basin
on Costa Rican territory to the San Juan River, as it has always done. There is no
evidence whatsoever that there has been any change in this pattern resulting from
construction of the Road, and no evidence of any form of compensable loss or
damage as a result of such construction. Nicaragua cannot avoid this point by
seeking to delay assessment of damages to a further phase. Even at the current
phase, it must make a colourable showing of loss and damage. It is wholly unable
to do so.
C . Nicaragua’s claims for cessation / guarantees and assurances of
non-repetition
371
4.11. As Costa Rica has already explained, an order requiring a State to cease
its wrongful conduct, and/or requiring guarantees and assurances of non-repetition
– like a claim for compensation – must necessarily be premised on an
internationally wrongful act. As Nicaragua has failed to demonstrate that any such
internationally wrongful acts have been committed by Costa Rica , there is no
basis on which to ground its requests for cessation and for assurances and
guarantees of non-repetition. Without prejudice to the foregoing, Costa Rica sets
out below why, in any event, Nicaragua has failed to meet the additional
requirements of these forms of reparation.
4.12. As to Nicaragua’s claim for cessation, even if Costa Rica’s conduct were
declared unlawful ( quod non), Nicaragua has not demonstrated why the Court
371 CRCM, para. 6.8.
124showing of actionable damage. The water continues to flow from the south basin
on Costa Rican territory to the San Juan River, as it has always done. There is no
evidence whatsoever that there has been any change in this pattern resulting from
construction of the Road, and no evidence of any form of compensable loss or
damage as a result of such construction. Nicaragua cannot avoid this point by
seeking to delay assessment of damages to a further phase. Even at the current
phase, it must make a colourable showing of loss and damage. It is wholly unable
to do so.
C . Nicaragua’s claims for cessation / guarantees and assurances of
non-repetition
371
4.11. As Costa Rica has already explained, an order requiring a State to cease
its wrongful conduct, and/or requiring guarantees and assurances of non-repetition
– like a claim for compensation – must necessarily be premised on an
internationally wrongful act. As Nicaragua has failed to demonstrate that any such
internationally wrongful acts have been committed by Costa Rica , there is no
basis on which to ground its requests for cessation and for assurances and
guarantees of non-repetition. Without prejudice to the foregoing, Costa Rica sets
out below why, in any event, Nicaragua has failed to meet the additional
requirements of these forms of reparation.
4.12. As to Nicaragua’s claim for cessation, even if Costa Rica’s conduct were
declared unlawful ( quod non), Nicaragua has not demonstrated why the Court
371 CRCM, para. 6.8. 376
order of such a measure by the Court. Nicaragua’s claim for guarantees and
assurances of non- repetition is grounded on the same facts as its claim for
cessation, namely the ongoing erosion Nicaragua alleges the road works are
causing. Ongoing erosion – even if established – does not constitute actionable
damage when it is on the scale alleged by Nicaragua, still less that disclosed by
the evidence; in the present case there is no showing of actual transboundary harm
and no risk of such harm. W ithout prejudice to its position in the present
proceedings, the fact is that Costa Rica has undertaken extensive mitigation works
on the Road, and continues to undertake such works , in a manner which it has
fully documented. This conduct further demonstrates that Costa Rica has always
acted, and continues to act, in good faith. Even if its conduct were declared
unlawful in some respect (quod non), the Court must presume that Costa Rica will
continue to act in good faith in the future , including with respect to mitigation
works and completion of the Road. 377 Consequently, there are no “special
circumstances” warranting an order requiring Costa Rica to provide guarantees
and assurances of non-repetition.
D . Late Nicaraguan request for the appointment of an expert by the Court
4.15. The present proceedings were initiated by Nicaragua in December 2011. In
accordance with the timetable agreed by the Court, Nicaragua presented its
Memorial on 19 December 2012, and Costa Rica its Counter -Memorial on
376 Dispute Regarding Navigational and Related Rights (Costa Rica v . Nicaragua) ,
Judgment, I .C .J . Reports 2009, p. 267, para. 150.
377
Factory at Chorzów , Merits, P.C.I.J. Judgment N° 13, 1927, Series A, No. 17 , p. 63;
Nuclear Tests (Australia v . France), Judgment, I .C .J Reports 1974, p. 272, para. 60;
Nuclear Tests (New Zealand v . France), Judgment, I .C .J . Reports 1974, p. 477, para. 63;
Military and Paramilitary Activitiesin and against Nicaragua (Nicaragua v . United
States of America), Jurisdiction and Admissibility, Judgment, I .C .J . Reports 1984, p. 437,
para. 101.
126 376
order of such a measure by the Court. Nicaragua’s claim for guarantees and
assurances of non- repetition is grounded on the same facts as its claim for
cessation, namely the ongoing erosion Nicaragua alleges the road works are
causing. Ongoing erosion – even if established – does not constitute actionable
damage when it is on the scale alleged by Nicaragua, still less that disclosed by
the evidence; in the present case there is no showing of actual transboundary harm
and no risk of such harm. W ithout prejudice to its position in the present
proceedings, the fact is that Costa Rica has undertaken extensive mitigation works
on the Road, and continues to undertake such works , in a manner which it has
fully documented. This conduct further demonstrates that Costa Rica has always
acted, and continues to act, in good faith. Even if its conduct were declared
unlawful in some respect (quod non), the Court must presume that Costa Rica will
continue to act in good faith in the future , including with respect to mitigation
works and completion of the Road. 377 Consequently, there are no “special
circumstances” warranting an order requiring Costa Rica to provide guarantees
and assurances of non-repetition.
D . Late Nicaraguan request for the appointment of an expert by the Court
4.15. The present proceedings were initiated by Nicaragua in December 2011. In
accordance with the timetable agreed by the Court, Nicaragua presented its
Memorial on 19 December 2012, and Costa Rica its Counter -Memorial on
376 Dispute Regarding Navigational and Related Rights (Costa Rica v . Nicaragua) ,
Judgment, I .C .J . Reports 2009, p. 267, para. 150.
377
Factory at Chorzów , Merits, P.C.I.J. Judgment N° 13, 1927, Series A, No. 17 , p. 63;
Nuclear Tests (Australia v . France), Judgment, I .C .J Reports 1974, p. 272, para. 60;
Nuclear Tests (New Zealand v . France), Judgment, I .C .J . Reports 1974, p. 477, para. 63;
Military and Paramilitary Activitiesin and against Nicaragua (Nicaragua v . United
States of America), Jurisdiction and Admissibility, Judgment, I .C .J . Reports 1984, p. 437,
para. 101. 4.17. There is no good basis for Nicaragua waiting until the close of the written
proceedings before making a request for the appointment of an independent
expert. The fact that Nicaragua proposes a road engineer, rather than a river
expert, shows that it has no case, other than to claim that it does not like the road.
The obvious explanations for this late request are that either it is a dilatory tactic
or it constitutes a belated recognition that it has been unable to make out its case
by reference to its own evidence. Either way, the proposal should be rejected.
E . Costa Rican position with regard to the order by the Court rejecting the
provisional measures requested by Nicaragua
4.18. In its Reply, Nicaragua contends that Costa Rica has not complied with
commitments as to mitigation made in the course of the 2013 provisional
measures hearing, 382 and also that it was primed to restart construction works 383
on or around 4 August 2014 (the date Nicaragua filed its Reply with the Court),
contrary to representations made during the oral hearings on provisional measures
in 2013.
4.19. In its order of 13 December 2013, rejecting Nicaragua’s request for
provisional measures, the Court stated:
“It [Costa Rica] explained that, under the updated version of the schedule,
the resumption of construction works on the section of the road along the
south bank of the San Juan River would not begin “before late 2014 or
382
NR, para. 7.18.
383 NR, para. 5.31.
1284.17. There is no good basis for Nicaragua waiting until the close of the written
proceedings before making a request for the appointment of an independent
expert. The fact that Nicaragua proposes a road engineer, rather than a river
expert, shows that it has no case, other than to claim that it does not like the road.
The obvious explanations for this late request are that either it is a dilatory tactic
or it constitutes a belated recognition that it has been unable to make out its case
by reference to its own evidence. Either way, the proposal should be rejected.
E . Costa Rican position with regard to the order by the Court rejecting the
provisional measures requested by Nicaragua
4.18. In its Reply, Nicaragua contends that Costa Rica has not complied with
commitments as to mitigation made in the course of the 2013 provisional
measures hearing, 382 and also that it was primed to restart construction works 383
on or around 4 August 2014 (the date Nicaragua filed its Reply with the Court),
contrary to representations made during the oral hearings on provisional measures
in 2013.
4.19. In its order of 13 December 2013, rejecting Nicaragua’s request for
provisional measures, the Court stated:
“It [Costa Rica] explained that, under the updated version of the schedule,
the resumption of construction works on the section of the road along the
south bank of the San Juan River would not begin “before late 2014 or
382
NR, para. 7.18.
383 NR, para. 5.31. 4.22. Once the designs are ready, another public tendering process will begin
and one or several companies will be selected to resume the construction proc ess.
As at the date of this Rejoinder, construction works have not resumed, although
mitigation work has continued to be implemented. T he situation on the ground
accurately reflects what Costa Rica said during the oral hearings on provisional
measures in 2013, and there is no basis for Nicaragua’s contention that Costa Rica
has failed to act in accordance with the statements it made on carrying out
386
remedial works.
4.23. As Costa Rica stated during the oral hearings in 2013, it has no intention
to suspend the construction works. As a matter of courtesy, it informed the Court
that the construction process was delayed, and that construction works would not
recommence until late 2014 at the earliest. It also made it clear that the design
387
process would continue and that installation of bridges and other remedial
388
works would continue, as indeed has happened.
4.24. The aim of the works being carried out on the Road, and of the final
designs for the segments of the Road, is to improve the quality of the Road,
including through preventing environmental harm taking place in the territory of
Costa Rica. A natural consequence of carrying out these works is that, in addition
to preventing environmental harm occurring in Costa Rica, the works would also
prevent any possible harm being potentially caused in the surrounding area,
including to Nicaragua. As stated in Chapter 2 above, Nicaragua had not
386 Cf. NR, para. 7.18.
387 Construction of a Road in Costa Rica along the San Juan River (Nicaragua v . Costa
Rica), Oral Hearing, 6 November 2013, CR 2013/29, p. 18, para. 18. (Parlett).
388
Ibid, p. 20, para. 24. (Parlett).
1304.22. Once the designs are ready, another public tendering process will begin
and one or several companies will be selected to resume the construction proc ess.
As at the date of this Rejoinder, construction works have not resumed, although
mitigation work has continued to be implemented. T he situation on the ground
accurately reflects what Costa Rica said during the oral hearings on provisional
measures in 2013, and there is no basis for Nicaragua’s contention that Costa Rica
has failed to act in accordance with the statements it made on carrying out
386
remedial works.
4.23. As Costa Rica stated during the oral hearings in 2013, it has no intention
to suspend the construction works. As a matter of courtesy, it informed the Court
that the construction process was delayed, and that construction works would not
recommence until late 2014 at the earliest. It also made it clear that the design
387
process would continue and that installation of bridges and other remedial
388
works would continue, as indeed has happened.
4.24. The aim of the works being carried out on the Road, and of the final
designs for the segments of the Road, is to improve the quality of the Road,
including through preventing environmental harm taking place in the territory of
Costa Rica. A natural consequence of carrying out these works is that, in addition
to preventing environmental harm occurring in Costa Rica, the works would also
prevent any possible harm being potentially caused in the surrounding area,
including to Nicaragua. As stated in Chapter 2 above, Nicaragua had not
386 Cf. NR, para. 7.18.
387 Construction of a Road in Costa Rica along the San Juan River (Nicaragua v . Costa
Rica), Oral Hearing, 6 November 2013, CR 2013/29, p. 18, para. 18. (Parlett).
388
Ibid, p. 20, para. 24. (Parlett). (d) Nicaragua is entitled to dredge the San Juan de Nicaragua River as
deemed necessary (section 4 below).
(1) Alleged breaches by Costa Rica
4.27. For the reasons set out in Costa Rica’s Counter -Memorial, and this
Rejoinder, Nicaragua has not established any of the breaches of international law
alleged to have been committed by Costa Rica. For its part, Nicaragua does not
recognise that it is bound by th ese same obligations in relation to Cost a Rica, for
390
its deeply alarming conductaddressed in the Certain Activities case.
4.28. Nicaragua invokes the allegation that Costa Rica has denied any
wrongdoing as a basis for declaratory relief . Nicaragua interprets the mere fact
that a country robustly defends baseless claims madeagainst it in the context of an
artificially conceived case as grounds for declaratory relief. This is untenable.
(2) Production of a transboundary EIA
4.29. Costa Rica has presented comprehensive evidence in support of its case
that no significant transboundary damage has been caused to Nicaragua and/or
that there was any significant risk of the same. Further, under the circumstances of
the declared emergency, there was no obligation to prepare a transboundary EIA,
particularly when the emergency was caused by the very State that now demands
390 Dispute Concerning Certain Activities carried out by Nicaragua in the Border Area
(Costa Rica v . Nicaragua), NCM, Vol. I, paras 3.21-3.52.
132 (d) Nicaragua is entitled to dredge the San Juan de Nicaragua River as
deemed necessary (section 4 below).
(1) Alleged breaches by Costa Rica
4.27. For the reasons set out in Costa Rica’s Counter -Memorial, and this
Rejoinder, Nicaragua has not established any of the breaches of international law
alleged to have been committed by Costa Rica. For its part, Nicaragua does not
recognise that it is bound by th ese same obligations in relation to Cost a Rica, for
390
its deeply alarming conductaddressed in the Certain Activities case.
4.28. Nicaragua invokes the allegation that Costa Rica has denied any
wrongdoing as a basis for declaratory relief . Nicaragua interprets the mere fact
that a country robustly defends baseless claims madeagainst it in the context of an
artificially conceived case as grounds for declaratory relief. This is untenable.
(2) Production of a transboundary EIA
4.29. Costa Rica has presented comprehensive evidence in support of its case
that no significant transboundary damage has been caused to Nicaragua and/or
that there was any significant risk of the same. Further, under the circumstances of
the declared emergency, there was no obligation to prepare a transboundary EIA,
particularly when the emergency was caused by the very State that now demands
390 Dispute Concerning Certain Activities carried out by Nicaragua in the Border Area
(Costa Rica v . Nicaragua), NCM, Vol. I, paras 3.21-3.52. (3) Transport of hazardous material
4.32. The belated case put forward in respect of transport of hazardous materials
398
has been considered in Chapter 2 above . As Costa Rica has shown, the
transport of fuels and any hazardous material is rigorously regulated by Costa
Rican regulations. 399 In circumstances where Nicaragua has put forward no
evidence (or even argument) to suggest that these would not be applied so far as
concerns the Road, and that there is some significant risk of harm to Nicaragua
arising out of non-application, there is no conceivable basis for the relief sought.
4.33. Costa Rica does emphasise here that, although the issue of the transport of
hazardous material does not arise in relation to the Road, Nicaragua’s
announcement that it intends to build an inter -oceanic canal raises very
considerable (extra) concern because it intends to use the canal for the
transportation of hazardous materials, such as oil and liquefied gas. The transport
of hazardous materials in this context, does pose a real danger of true significant
transboundary damage to Costa Rica and to Nicaragua, unlike allegations of
humble farmers carrying out a couple of gallons of fuel.
(4) Dredging of the San Juan River
4.34. Nicaragua’s request that “it is entitled … to execute works to improve
navigation on the San Juan River as it deems suitable”, is duplicative of
submissions made in the Certain Activities case. Indeed, this is the third time that
Nicaragua makes the same request before the Court. The first time was in the
398
See paras. 2.104 and 2.105 above.
399 See para. 2.105 above.
134(3) Transport of hazardous material
4.32. The belated case put forward in respect of transport of hazardous materials
398
has been considered in Chapter 2 above . As Costa Rica has shown, the
transport of fuels and any hazardous material is rigorously regulated by Costa
Rican regulations. 399 In circumstances where Nicaragua has put forward no
evidence (or even argument) to suggest that these would not be applied so far as
concerns the Road, and that there is some significant risk of harm to Nicaragua
arising out of non-application, there is no conceivable basis for the relief sought.
4.33. Costa Rica does emphasise here that, although the issue of the transport of
hazardous material does not arise in relation to the Road, Nicaragua’s
announcement that it intends to build an inter -oceanic canal raises very
considerable (extra) concern because it intends to use the canal for the
transportation of hazardous materials, such as oil and liquefied gas. The transport
of hazardous materials in this context, does pose a real danger of true significant
transboundary damage to Costa Rica and to Nicaragua, unlike allegations of
humble farmers carrying out a couple of gallons of fuel.
(4) Dredging of the San Juan River
4.34. Nicaragua’s request that “it is entitled … to execute works to improve
navigation on the San Juan River as it deems suitable”, is duplicative of
submissions made in the Certain Activities case. Indeed, this is the third time that
Nicaragua makes the same request before the Court. The first time was in the
398
See paras. 2.104 and 2.105 above.
399 See para. 2.105 above. 4.36. For the reasons explained above, all of Nicaragua’s remedies must be
rejected.
(5) Conclusion
4.37. Nicaragua’s constant changes in its request for remedies in the written
phase of these proceedings underscores the fact that it has not built a credible
case. The continuous adding up to the litany of reparations it demands only
reinforces Costa Rica’s position as to the artificial nature of the present case. A
general declaration of unlawful conduct warranting remedies requires that the
applicant state formulate a credible case grounded on verified facts. Nicaragua’s
case has revolved around criticisms of Costa Rica’s evidence, yet, it has not built
a case of its own. The lack of verifiable evidence in two rounds of written and oral
pleadings, even after Costa Rica proposed joint efforts to that end, raises concerns
as to the chimerical character of Nicaragua’s legal strategy. It is Costa Rica’s view
not only that the Court must reject wholly the remedies thus requested, but that it
should seriously consider the bringing up of such gratuitous and wasteful
proceedings as un acceptable show of poor judgment by a country that is
constantly complaining about the expense of treasure and time that thes e cases
demand.
1364.36. For the reasons explained above, all of Nicaragua’s remedies must be
rejected.
(5) Conclusion
4.37. Nicaragua’s constant changes in its request for remedies in the written
phase of these proceedings underscores the fact that it has not built a credible
case. The continuous adding up to the litany of reparations it demands only
reinforces Costa Rica’s position as to the artificial nature of the present case. A
general declaration of unlawful conduct warranting remedies requires that the
applicant state formulate a credible case grounded on verified facts. Nicaragua’s
case has revolved around criticisms of Costa Rica’s evidence, yet, it has not built
a case of its own. The lack of verifiable evidence in two rounds of written and oral
pleadings, even after Costa Rica proposed joint efforts to that end, raises concerns
as to the chimerical character of Nicaragua’s legal strategy. It is Costa Rica’s view
not only that the Court must reject wholly the remedies thus requested, but that it
should seriously consider the bringing up of such gratuitous and wasteful
proceedings as un acceptable show of poor judgment by a country that is
constantly complaining about the expense of treasure and time that thes e cases
demand. everything possible to render those rights empty of content by other means ). It
also appears to have abandoned the idea of having the Road stopped altogether.
5. Nicaragua has nonetheless sought to portray the Road in the worst
possible light , by reference to a very few limited stretches, which appear
repeatedly in the photographs submitted with Nicaragua’s Reply and in its
technical reports. But even in the part of the Road which Nicaragua considers to
be problematic, remediation works are either now complete or underway. Further,
when Nicaragua’s claims are considered in their correct context, even on their
inflated figures of sediment eroded from the Road to the River, it becomes clear
that what is in dispute is no more than a very small – even imperceptible –
increase in the sediment load of a River which is naturally adapted to a sediment
load that is “very heavy”. Indeed, Nicaragua implicitly accepts that the impact of
the Road is very small, as is demonstrated by its belated attempt to depict the
existing sediment load as “excessive” and unnatural, alongside its new claims that
the existing sediment load is the re sponsibility of Costa Rica. Ultimately, the
issue for the Court to decide is whether the Road – which in large part is a track
built on existing paths – is having a significant impact on the San Juan River.
6. Costa Rica submits that the a nswer to that question is no. Indeed, e ven
accepting the estimates of sediment eroding from the Road to the River put
forward by Nicaragua’s experts in its Reply (which Costa Rica most certainly
does not accept), this sediment would represent an addition of less than 3% to the
total annual sediment load of the River. On this basis, while there remain a
number of issues in dispute between the Parties’ exper ts in terms of the extent of
the alleged impacts of the Road, including if any impacts have actually occurred,
ultimately these issues are immaterial, because even on Nicaragua’s own
estimates, the Road is having no adverse impact on the River . These matters are
138everything possible to render those rights empty of content by other means ). It
also appears to have abandoned the idea of having the Road stopped altogether.
5. Nicaragua has nonetheless sought to portray the Road in the worst
possible light , by reference to a very few limited stretches, which appear
repeatedly in the photographs submitted with Nicaragua’s Reply and in its
technical reports. But even in the part of the Road which Nicaragua considers to
be problematic, remediation works are either now complete or underway. Further,
when Nicaragua’s claims are considered in their correct context, even on their
inflated figures of sediment eroded from the Road to the River, it becomes clear
that what is in dispute is no more than a very small – even imperceptible –
increase in the sediment load of a River which is naturally adapted to a sediment
load that is “very heavy”. Indeed, Nicaragua implicitly accepts that the impact of
the Road is very small, as is demonstrated by its belated attempt to depict the
existing sediment load as “excessive” and unnatural, alongside its new claims that
the existing sediment load is the re sponsibility of Costa Rica. Ultimately, the
issue for the Court to decide is whether the Road – which in large part is a track
built on existing paths – is having a significant impact on the San Juan River.
6. Costa Rica submits that the a nswer to that question is no. Indeed, e ven
accepting the estimates of sediment eroding from the Road to the River put
forward by Nicaragua’s experts in its Reply (which Costa Rica most certainly
does not accept), this sediment would represent an addition of less than 3% to the
total annual sediment load of the River. On this basis, while there remain a
number of issues in dispute between the Parties’ exper ts in terms of the extent of
the alleged impacts of the Road, including if any impacts have actually occurred,
ultimately these issues are immaterial, because even on Nicaragua’s own
estimates, the Road is having no adverse impact on the River . These matters are (b) For its part, Costa Rica has shown that the San Juan River has not been
significantly damaged; indeed, taking Nicaragua’s own technical case at
its highest, the Road hypothetically contributes an insignificant quantity of
sediment to the River, less than the annual variability of the sediment load
in what is, and alwa ys has been, a highly sediment laden and variable
river. In short, the impact on the San Juan River as a result of the
construction of the Road is imperceptible . In these circumstances, it is
apparent that the Road is not causing harm to the San Juan River, let alone
significant harm.
(c) In these circumstances, Costa Rica’s sovereign right to develop its own
territory and to construct roads anywhere in its territory must be fully
respected.
9. Costa Rica is confident that the Court is now able to appreciate th e
artificiality of the present case. All Nicaraguan claims and submissions must
consequently be rejected.
140 Submissions
For these reasons, and reserving the right to supplement, amplify or amend
the present submissions, Costa Rica requests the Court to dismiss all of
Nicaragua’s claims in this proceeding.
Ambassador Sergio Ugalde
Co-Agent of Costa Rica
2 February 2015
141142 APPENDIX A
Professor Colin Thorne
Assessment of the Impact of the Construction of the Border Road in Costa
Rica on the San Juan River: Reply Report
February 2015
143144 Appendix A
DISPUTE CONCERNING
C ONSTRUCTION OF AROAD INC OSTAR ICA ALONG THS ANJUAN R IVER
(NICARAGUA V COSTA RICA)
Assessment of the Impact of the Construction of the Border Road in Costa
Rica on the San Juan River: Reply Report
Prepared by
Colin Thorne
Nottingham, UK
Professor and Chair of Physical Geography
University of Nottingham
February 2015
1
145 TABLE OF CONTENTS
TABLE OF CONTENTS............................................................................................................2
LIST OF FIGURES....................................................................................................................3
LIST OF TABLES......................................................................................................................7
1. Introduction ...............................................................................................................8
2. My Qualifications....................................................................................................10
3. Terms of Reference and Methodology............................................................11
4. Has Sediment from Route 1856 had any significant impact on
Sediment loads in the Río San Juan?................................................................15
5. Has Construction of Route 1856 had any significant impacts on
channel morphology in the Río San Juan?......................................................87
6. Has Route 1856 had any significant impacts on ecology or fishery of
the Río San Juan, or any impact on navigation? ........................................108
7. What effect have Costa Rica’s Mitigation Works had, and how have
they progressed since late-2013?..................................................................126
8. Conclusions............................................................................................................155
9. References..............................................................................................................160
10. Statement of Independence and Truth........................................................164
2
146 LIST OF FIGURES
Figure 4.1 (a) Terrestrial LiDAR being set up at monitoring Site 4 on May 27, 2014
(b) LiDAR point cloud (c) contour data for use in erosion measurements and
calculations.......................................................................................................................................... 16
Figure 4.2 (a) Orthophoto, (b) Digital Elevation Model and (c) cross-section
through gully at Site 12 (Dr Kondolf’s site 9.4) derived from photogrammetric
survey on October 28, 2014 and used to estimate gully planform area and eroded
volume................................................................................................................................................... 16
Figure 4.3 Location map showing UCR erosion monitoring and measurement sites.
.................................................................................................................................................................. 20
Figure 4.4 (a) Sediment trap at Site 5 on October 1, 2014. (b) Survey points and
contour map of sediment surface elevations on October 22, 2014............................... 21
Figure 4.5 Current condition of the Road between Marker II and Delta Costa Rica
derived from on the ground inspection by Dr Mende. ....................................................... 30
Figure 4.6 (a) Study area of the Río San Juan Basin (b) tributary basins (c) digital
elevation model (d) mean annual precipitation (e) land cover map (f) soil
classification map.............................................................................................................................. 40
Figure 4.7 (a) land cover factor, C (b) soil erodibility factor, K (c) rainfall erosivity,
R and (d) slope length factor LS. ................................................................................................. 43
Figure 4.8 Distribution of potential soil erosion in the Río San Juan basin between
Lake Nicaragua and Delta Costa Rica. ....................................................................................... 43
Figure 4.9 Spatial distribution of the delivery index in the study area. .....................44
Figure 4.10 Calibrated specific sediment yields in the study area............................... 45
Figure 4.11 Hydrometric stations with sediment records in the Río San Juan basin.
.................................................................................................................................................................. 46
Figure 4.12 Suspended sediment rating curve for the Río Colorado at Delta Costa
Rica (Station 11-04) between 2010 and 2014................................................................... 47
Figure 4.13 Bed load rating curve for the Delta Colorado (11-04) station...............50
Figure 4.14 Mean annual suspended loads in the Río Colorado and lower San Juan
as percentages of suspended load in the Río San Juan assuming that (a) 95%, (b)
90% and (c) 85% of the discharge flows to the Río Colorado......................................... 54
Figure 4.15 Mean annual bedloads in the Río Colorado and lower Río San Juan as
percentages of bedload in the Río San Juan assuming that (a) 95%, (b) 90% and (c)
85% of the discharge flows to the Río Colorado................................................................... 55
Figure 4.16 Annual yields of Road-derived sediment from the basins of major
Costa Rican tributaries between Marker II and Delta Costa Rica under a ‘worst
case’ rainfall scenario...................................................................................................................... 56
Figure 4.17 Suspended and bed sediment load budgets for the Río San Juan
between Marker II and the Delta (values in t yr )............................................................... 60
Figure 4.18 Suspended sediment budget for the San Juan-Colorado System with
the contribution of fine-grained sediment from Route 1856 indicated in red at the
lower edge of the diagram. Additional loads entering the lower San Juan and
3
147 Colorado are based on the assumption that 5% of sediment derived from the Road
is coarse-grained (values in parenthesis correspond to a 10% fraction of coarse
material)............................................................................................................................................... 60
Figure 4.19 Distribution of wind and rain in Costa Rica due to a tropical cyclone
near the Caribbean coast of Nicaragua..................................................................................... 68
Figure 4.20 Rainfall distribution in Costa Rica recorded during Hurricane Mitch..
.................................................................................................................................................................. 68
Figure 4.21 Tectonic map of South Central America, indicating the main tectonic
and neotectonic elements affecting the San Juan basin (indicated by the blue line).
The lower basin lies in an area of the central America back-arc.................................... 74
Figure 4.22 Volcanoes with heights of between 2,000 and 3,000 metres that
periodically contribute extraordinary amounts of sediment to the San Juan
drainage system................................................................................................................................. 75
Figure 4.23 Historical record of earthquakes and volcanic eruptions inputting
extraordinary amounts of sediment to the San Juan basin from the Costa Rican part
of the basin. ......................................................................................................................................... 75
Figure 4.24 Map of landslides triggered by the 2009 Cinchona earthquake (from
Alvarado, 2010)................................................................................................................................. 76
Figure 4.25 (a) tributary basins (b) digital elevation model (c) slope length factor
LS (d) mean annual precipitation (e) rainfall erosivity factor, R (f) land cover (g)
land cover factor, C and (h) calibrated specific sediment yields (E) in the study
area......................................................................................................................................................... 79
Figure 4.26 Turbid water draining to the Río San Juan from Nicaraguan tributaries
on 23 December 2012 (a) Río Santa Cruz (b) Río Sábalos................................................ 82
Figure 5.1 Designation of reaches of the Río San Juan according to the
Montgomery-Buffington classification (Figure 12 in the 2013 Thorne Report)......87
Figure 5.2 Fifteen north bank deltas photographed from Costa Rican airspace in
April 2014. These deltas are formed in sediment eroded from Nicaragua and some
are considerably larger than any of those photographed by Dr Kondolf along the
south bank. The size and morphology of these deltas should be compared to those
shown in Appendix F of the 2014 Kondolf Report, which were also taken at
conditions of low flow in the Río San Juan....................................................................... 92-95
Figure 5.3 Pre- and Post-Road satellite images establishing that at least two of the
eight south bank deltas identified as being formed from sediment derived from the
Road were present prior to construction of the Road........................................................ 96
Figure 5.4 Satellite images showing that flow from the lower Río San Juan carries
turbid water with a high concentration of fine sediment into both the Bay of San
Juan del Norte and the littoral sediment system of the Caribbean Sea. Image dates
(a) 13 December 1997 (b) 26 November 2013………………………………………………..107
Figure 6.1 Sampling Points along the San Juan River between El Castillo and Boca
San Carlos. Each point corresponds to a delta formed by a creek draining to Río
San Juan (This is Figure 1 in the Ríos Report).................................................................... 115
4
148Figure 6.2 Suspended sediment concentration as a function of discharge for 2,409
samples taken from the Río Colorado, Río San Juan and its Costa Rican tributaries.
Note: Station 11-04 is the Delta Colorado (Station) which receives about 90% of
the flow in the Río San Juan immediately upstream. ....................................................... 121
Figure 7.1 Examples of slope preparation and planting to provide surface
protection by CODEFORSA along the Road between Marker II and Boca San Carlos
............................................................................................................................................................... 131
Figure 7.2 Slope 9 (near Tiricias) in February 2014, following treatment by
CODEFORSA for surface protection (example taken from 2014 CODEFORSA
Report)............................................................................................................................................... 132
Figure 7.3 Photographs representative of conditions along the Road between Boca
Sarapiquí and the Delta observed from the air by the author on 17 November,
2014 .................................................................................................................................................... 139
Figure 7.4 Photographs representative of conditions along the Road between Boca
San Carlos and Boca Sarapiquí observed from the air by the author on 17
November 2014.............................................................................................................................. 140
Figure 7.5 Condition of some of the CODEFORSA reforestation sites visited in
2014. Photographs by author.................................................................................................... 141
Figure 7.6 Condition of some of the slopes revegetated by CODEFORSA that were
visited in 2014. Photographs by the author. ....................................................................... 142
Figure 7.7 Condition of additional CONAVI erosion mitigation works inspected by
the author in 2014: left watercourse crossings, and right slopes. Photographs by
author................................................................................................................................................. 143
Figure 7.8 Large scale mitigation works by CONAVI on-going at sites around and
Caño Cureñita, inspected in November 2014. Photographs by author..................... 144
Figure 7.9 Natural landslide at the north (Nicaragua) bank of the Río San Juan
o
observed from Costa Rican airspace at coordinates W 084o 03’ 58.5’’ N 10 45’
31.5’’ on 17 November, 2014. Note the temporary delta formed by sediment and
fallen trees delivered directly to the river by the landslide. Sediment and trees
enter the river due to natural processes to give the Río San Juan naturally high
sediment and debris loads and high turbidity, especially during the rainy season.
Four other similar landslides were also observed during a single overflight that
day. Photographs by author....................................................................................................... 145
Figure 7.10 Typical views of access roads traversed during the field visit on 29
August 2014. Photographs by author.................................................................................... 146
Figure 7.11 The Road near Marker II (a) prior to mitigation work on 15 February 2013
(b) on 7 May 2013 with mitigation measures in place: note in-board drainage channel
and extensive biodegradable, erosion control matting (c) on 23 April showing that
all mitigation measures survived the rainy season and (d) on 29 August showing
that vegetation has stabilized both margins of the road bed and was spreading
across the areas protected by coconut matting. Photographs by author................. 147
Figure 7.12 View down a large gully in a fill prism created by concentrated runoff
from the Road draining to Costa Rican territory to the west of Marker II (a) in
5
149 February when it was actively eroding and (b) in May when the gully had been
back-filled and stabilized using a culvert and concrete drainage channel, with
coconut matting used to protect the surrounding fill slope from sheet and rill
erosion. Subsequent visits in (c) April and (d) August showed the culverted
crossing to be intact after the rainy season and vegetation to be recolonizing the
surrounding area. Photographs by the author. .................................................................. 148
Figure 7.13 Road at East 497867, North 325463 about 6.4 km east of Marker II (a)
on 15 February when failure of geotextile slope protection had allowed
concentrated out-board runoff from the Road to create two gullies and in-board
runoff was undercutting a cut slope (b) on 7 May 2013 after construction of
concrete-lined out-board and in board ditches (c) and (d) in 2014 erosion has been
effectively mitigated, the gullies have healed and both cut and fill slopes are
revegetating. Photographs by author..................................................................................... 149
Figure 7.14 Road at East 498072, North 325345, about 6.6 km east of Marker 2 (a)
on 15 February 2013 showing a network of gullies on an outboard slope and
sediment accumulated as a run-out deposit on the flat terrace surface (b) on 7 May
2013 showing mitigation works (concrete channels, drop structures, silt fences
and sediment trap to prevent sediment reaching the River (c) by April 2014 local
erosion had ceased and the slopes had revegetated and (d) in August the
undersized culvert and fill prism beneath the road at the watercourse crossing had
been replaced by a larger culvert with head and exit works, covered by a
compacted soil-cement mixture. Photographs by author.............................................. 151
Figure 7.15 Road at East 502480, North 321561, close to the Río Infiernito (a) on
15 February when surface unmanaged runoff from the road bed and surrounding
slopes disturbed during construction had caused sheet and rill erosion of bare soil
surfaces. (b) The same stretch of road on 7 May 2013 after protection of the road
surface using crushed rock, installation of silt fences to prevent sheet and rill
erosion while directing down-slope surface runoff into concrete-lined outboard
and inboard ditches (c) in April silt fences were showing wear and tear, but
vegetation was spreading fast and by August (d) erosion mitigation had been
successful at this site. Photographs by author.................................................................... 152
Figure 7.16 Path cleared for the Road near Crucitas, just east of the Río Infiernito
(a) on 15 February when unmanaged runoff from the path cleared in preparation
for construction of the road bed had caused sheet and rill erosion (b) The same
area on 7 May 2013 after installation integrated measures to manage runoff
involving regrading, silt fences, and concrete-lined outboard ditch (c) in April
2014, nothing had changed significantly and in August 2014 it was apparent that
erosion mitigation continued to be successful and that no sediment from the site
was reaching the Río San Juan. Photographs by author ................................................. 154
6
150 LIST OF TABLES
Table 4.1 UCR erosion monitoring and measurement sites. .......................................... 19
Table 4.2 Maximum annual erosion rates from the 2014 UCR Report....................... 21
Table 4.3 Summary erosion monitoring results in the 2013 UCR Report.................22
Table 4.4 Measured data for fill slope gullies at UCR Sites 8, and 11-13...................23
Table 4.5 Types of slope erosion observed along the entire length of the Road
between Marker II and Delta Costa Rica (from the Mende report)............................... 31
Table 4.6 Example calculation of annual erosion volume for Cut Slope T-8a..........32
Table 4.7 Potential erosion based on catchment-aggregated and fully-distributed
applications of the USLE................................................................................................................. 45
Table 4.8 Soil erosion and sediment yields in the Study Area of the San Juan Basin
............................................................................................................................................................... 466
Table 4.9 Mean annual suspended sediment loads for the 14 hydrometric stations
.................................................................................................................................................................. 48
Table 4.10 Confidence intervals as normalized anomalies for the mean annual
suspended sediment loads at the hydrometric stations................................................. 488
Table 4.11 Mean annual bedload transport in the Río Colorado at Station 11-04 50
Table 4.12 Hydrological data, hourly and daily coefficients of variation and mean
annual suspended sediment loads at the mouths of the Río Sarapiquí and Río San
Carlos..................................................................................................................................................... 52
Table 4.13 Mean Annual suspended and bedloads in the Río San Juan for a range
of percentage discharges flowing to the Río Colorado....................................................... 53
Table 4.14 Mean Annual suspended and bedloads in the lower Río San Juan for a
range of percentage discharges flowing from the Río San Juan to the Río Colorado
.................................................................................................................................................................. 53
Table 4.15 Erosion rates for the Road surface..................................................................... 55
Table 4.16 Annual yields of Road-derived sediment from the basins of major Costa
Rican tributaries between Marker II and Delta Costa Rica under a ‘worst case’
rainfall scenario................................................................................................................................. 56
Table 4.17(a) Initial and adjusted suspended sediment yields for Study Area (b)
Summary of estimated sediment loads and sediment inputs from the Road into the
Río San Juan......................................................................................................................................... 58
Table 4.18 Specific sediment yields in Nicaraguan and Costa Rican Basins ............78
Table 4.20 Mean Annual bedloads in the lower Río San Juan for a range of
percentage discharges flowing from the Río San Juan to the Río Colorado 104 Table
Table 7.1 Mortality report for the CODEFORSA reforestation programme……….129
Table 7.1 Distribution of slopes and watercourse crossings………………………..…..136
Table 7.3 Mitigation status of slopes ……………………………………………………………..137
Table 7.4 Mitigation status of watercourse crossings……………………………………...138
7
151 1. Introduction
1.1. I am Colin Thorne, Professor of Physical Geography at the University of
Nottingham. I have been requested by Costa Rica to prepare an
independent expert report for the International Court of Justice (the
Court) in connection with the claim brought against Costa Rica by
Nicaragua concerning the construction of a road in Costa Rica near the San
Juan River (the Road). I provided an independent expert report entitled
“Assessment of the Impact of the Construction of the Border Road in Costa
Rica on the San Juan River” in December 2013, which was submitted to the
Court as Appendix A to Costa Rica’s Counter -Memorial in the Road case.
This Report responds to Nicaragua’s Reply in the Road Case, and evidence
annexed thereto, dated 4 August 2014.
1.2. I am instructed to form an independent expert opinion on the matters set
out in the Terms of Reference below. In that regard I understand that I
have an obligation to the Court to express my honest opinion.
1.3. Pursuant to those instructions, I have reviewed Nicaragua’s Reply of
4 August 2014 in the Road Case, and in particular have focused on the
following documents:
(a) Report prepared by Dr G. Mathias Kondolf entitled “Erosion and
Sediment Delivery to the R ío San Juan from Rout e 1856”, July 2014
(the 2014 Kondolf Report), which is Annex 1 to Nicaragua’s Reply;
(b) Mr Danny Hagans and Dr Bill Weaver, “Evaluation of Erosion,
Environmental Impacts and Road Repair Efforts at Selected Sites
along Juan Rafael Mora Route 1856 in Costa Rica, Adjacent the R ío
San Juan, Nicaragua”, July 2014 (the Hagans and Weaver Report ),
which is Annex 2 to Nicaragua’s Reply;
8
152(c) Dr Edmund D. Andrews, “An Evaluation of the Methods, Calculations,
and Conclusions Provided by Costa Rica Regarding the Yield and
Transport of Sediment in the R ío San Juan Basin”, July 2014 (the
Andrews Report) which is Annex 3 to Nicaragua’s Reply;
(d) Dr Blanca Rios Touma, “Ecological Impacts of the Route 1856 on the
San Juan River, Nicaragua”, July 2014 (the Rios Report ), which is
Annex 4 to Nicaragua’s Reply; and
(e) Golder Associates Inc., “The Requirements of Impact Assessment for
Large-Scale Road Construction Project in Costa Rica Along the San
Juan River, Nicaragua”, July 2014 (the Golder Report ), which is
Annex 6 to Nicaragua’s Reply.
9
153 2. My Qualifications
2.1. My relevant qualifications are set out in Section 2 of my report of
December 2013 (Appendix A to Costa Rica’s Counter -Memorial). My
curriculum vitaeis included as Attachment 1 to that Report.
10
154 3. Terms of Reference and Methodology
A. Terms of Reference
3.1. I have been asked to provide an independent expert opinion on the
environmental impacts of the Road o n the Río San Juan i n Nicaragua. In
this context, I have been asked to review and assess the information and
opinions given in the reports listed in paragraph 1.3 above and the claims
made by Nicaragua in its Reply in the Construction of a Road case relating
to harm or potential harm to Nicaraguan territory.
3.2. As in my report annexed to Costa Rica’s Counter -Memorial I have been
instructed to consider the potential environmental impacts of the Road on
Nicaragua. Therefore, I have not addressed any impacts of the Road within
Costa Rican territory. Nor do I express any opinion on any question of law.
B. Methodology
3.3. In preparing this Report, my approach has been as follows:
(a) I have reviewed the reports listed in paragraph 1.3 above and I have
reviewed Nicaragua’s Reply insofar as it deals with harm or
potential harm to the San Juan River and makes statements that rely
on the Reports listed above. I have also reviewed the letter from
Dr Andrews dated 12 December 2014 and attached Letter from the
Court to Costa Rica, 16 December 2014, Reference 144543.
(b) I have conducted a review of the published academic literature on
the sediment and environmental impacts of sediment on rivers,
focusing on sources mentioned in the 2014 Kondolf and Andrews
Reports.
11
155 (c) I have participated in three further site visits to the Road, on 23
April, 29 August and 17 November 2014. On each occasion I drove
along stretches of the Road and made detailed observations at sites I
selected as well as viewing those sites and longer stretches of the
Road from the air. I also spoke first hand to engineers and scientists
engaged in works along the Road and took photographs from the
ground and the air;
(d) I have requested, formulated and supervised continuing scientific
and technical studies performed by qualified Costa Rican scientists
and engineers, to elicit the data and information needed to evaluate
the potential for construction of the Road to impact the Río San Juan;
(e) I have participated in meetings with the technical team of scientists
and engineers in San Jose in April, August and November 2014,
during which we discussed approaches and methodologies to be
employed in performing the work, reviewed progress and discussed
the results of investigations employing archi ve-based, field, remote-
sensing and GIS-based research, and computer modelling; and
(f) I have reviewed the preliminary findings of the team, requesting
additional clarification where appropriate.
(g) The technical reports have been produced and provided to me as the
outcomes of this supervised research process are:
Eng. Rafael Oreamuno Vega, M. Eng. and Eng. Roberto Villalobos
Herrera 2014. Second Report on Systematic Field monitoring of
Erosion and Sediment Yield along Route 1856, University of Costa
Rica, CIEDES, Sa n José, Costa Rica, November, 2014, 37 pages [the
2014 UCR Report]
Mende, A. 2014. Inventory of Slopes and Watercourses related to the
Border Road Nº 1856 between Mojón II and Delta Costa Rica: Second
12
156Report, presented to the Ministry of Foreign Affairs - Costa Rica, San
José, Costa Rica, December 2014, 42 pages, plus Appendix B:
Inventory of Slopes (402p) and Appendix C: Inventory of
Watercourses (142p) [the Mende Report]
Institute of Electricity (ICE) 2014. Second Report on Hydrology and
Sediments for the Costa Rican River Basins Draining to the San Juan
River, Center for Basic Engineering Studies, Department of
Hydrology, San José, Costa Rica, December 2014, 34 pages [the 2014
ICE Report]
Fallas, J. C. 2014. Comments on the Report By Dr Kondolf (as it
pertains to Hurricanes And Tropical Storms) In: Section 1.2 - Risks
Of Large Contributions From Rte. 1856 [Annex I, pages 71 -74], Costa
Rican National Meteorological Institute, San José, Costa Rica, 4 pages
[the Fallas Report].
Astorga, A. 2014. Extraordinary sediment inputs due to exceptional
events on the San Juan River, Central American School of Geology at
the University of Costa Rica, San José, Costa Rica, 21 pages [the
AstorgaReport].
Angulo, A. 2014. Fish Fauna in the San Juan River, Tropical Science
Center, 7 pages (the Angulo Report).
Pacheco, B. 2014. Answers and Study Analysis, “Ecological Impacts
of the Route 1856 on the San Juan River, Nicaragua”, by Dr Rios
Touma 2014), Tropical Science Center , 7 pages (the Pacheco
Report).
Gutierrez, P.E. 2014. Critical statistical analysis of the report
“Ecological Impacts of the Route 1856 on the San Juan River,
Nicaragua” by Blanca Ríos Touma, Licenciado in Water Resources,
University of Costa Rica, San José, Costa Rica, 5 pages (the Guti errez
Report).
CODEFORSA 2014a. Consulting Services for the Development and
Implementation of an Environmental Plan for the Juan Rafael Mora
Porras Border Road, Comisión de Desarrollo Forestal De San Carlos ,
Ciudad Quesada, San Carlos, Costa Rica (the 2014 CODEFORSA
Report)
CODEFORSA 2014b. Restoration and rehabilitation of ecosystems
affected by the construction of the Juan Rafael Mor a Porras border
13
157 road, Ruta 1856, Comisión d e Desarrollo Forestal De San Carlos ,
Ciudad Quesada, San Carlos, Costa Rica (the CODEFORSA Quarterly
Report for November 2014).
CONAVI 2014. Works on National Road N° 856: Before and After -
Updated as of December 2014 (the 2014 CONAVI Report).
(h) I have also reviewed an independent expert opinion by Professor Ian
Cowx on the impacts of the Road on fish and macroinvertebrates in
the San Juan River, Nicaragua (the Cowx Report).
3.4. Where I rely on information and data contained in these reports and
studies, or any other reports prepared in the course of the investigations
and activities referred to in the sub -sections of paragraph 3.3 above, I
indicate that I am doing so.
14
158 4. Has Sediment from Route 1856 had any significant impact on
Sediment loads in the Río San Juan?
A. Field monitoring of erosion by landslides, gullies, rills, and sheet erosion
4.1. In September 2013 the Centre for Research in Sustainable Development
(CIEDES) at the University of Costa Rica submitted to the Ministry of
Foreign Affairs of Costa Rica a document titled ‘Report on Systematic Field
monitoring of Erosion and Sediment Yield along Route 1856’ (Annex 1 to
Costa Rica’s Counter -Memorial). This report detailed their programme of
field monitoring of erosion of the Road, Cut and Fill slopes , and listed their
results to date.
4.2. The same team continued monitoring in 2014, but gained access to better
instrumentation and introduced additional sites to improve data accuracy .
They also used the new measurements to check the reliability of dat a
reported in the 2013 R eport. The combined data were then used to make
upper bound estimates of average erosion rates occurring along Route
1856.
4.3. Two technical changes were made. First, terrestrial LiDAR (Light Distance
And Rangefinding) replaced manual measurements of slope topography at
the long-term slope erosion monitoring sites (Figure 4.1). Second, new
measurement sites were added, with aerial, photogrammetric survey s
performed for three additional sites, which are very difficult to access and
which were therefore surveyed from the air (Figure 4.2) . These Sites
(numbered 11, 12 and 13 in the UCR Report) correspond to SES Sites 8.1,
9.4 and 9.5 in the 2014 Kondolf Report.
15
159 (a)
(b) (c)
Figure 4.1 (a) Terrestrial LiDAR at monitoring Site 4 on May 27, 2014 (b) LiDAR p oint
cloud (c) contour data for use in erosion measurements (from the 2014 UCR Report).
(a)
(b)
(c)
Figure 4.2 (a) Orthophoto, (b) Digital Elevation Model and (c) cross-section through gully
at Site 12 (Dr Kondolf’s SE Site 9.4) derived from photogrammetric survey on October 28,
2014 and used to estimate gully planform area and eroded volume (from the 2014 UCR
Report).
16
1604.4. These technical changes directly address and deal with Dr Kondolf’s most
serious criticisms of the 2013 UCR Report. On page 48 of his 2014 Report,
Dr Kondolf states that, “The most fundamental weakness of the UCR study is
its failure to measure erosion downstream in the more severely eroding
sites.” As explained in the 2014 UCR Report, the problem of lack of ground
access to the Road east of the Río Infiernito was resolved in 2014, allowing
addition of erosion measurements at sites 11, 12 and 13 , which
correspond to SES sites 8.1, 9.4 and 9.5 in the 2014 Kondolf Report.
4.5. On page 51 of his 2014 Report, Dr Kondolf presents his second
fundamental criticism, stating that, “The authors of the UCR Report also
applied a flawed methodology. Rather than directly measuring all significant
erosion features within the areas experiencing significant erosion and
mapping the occurrence of smaller features such as areas of rilling (thereby
collecting real data for the sites of significant erosion), they used a
complicated system to take their measured erosion rate for a feature such as
a gully, and then reduced the rate by dividing it over the area of the entire
exposed “slope” in which it occurred. This was effectively an arbitrary
reduction to the rate because the size of the exposed area in which the
eroding feature occurred was unrelated to the eroding feature itself. The
authors of the UCR Report also did not account for other erosional processes
occurring over the rest of the slope, which artificially reduced the resulting
erosion rate. This is a principal reason that the erosion rates reported in
Table 6 of Annex 1 are unreasonably low.”
4.6. Application of terrestrial LiDAR at UCR’s long -term monitoring sites has
made it possible to estimate the rate of erosion for the area of each erosion
feature individually, instead of averaging it over the entire area of
monitored slope. Hence, erosion rates now represent the areas actually
affected by each type of erosion, rather than being averages for the
17
161 monitored slope. T he 2013 field measurements remain unaffected , only
the area over which the erosion rates are average d has changed. The
LiDAR scans have allowed UCR to reprocess their 2013 data and the
results largely validate the simpler methods used in 2013. They also
indicate that er osion rates measured in 2014 are comparable to but
generally lower than those monitored in 2013.
4.7. UCR have established characteristic erosion rates for each type of erosion
rather than for each monitored slope so that these rates can be applied
specifically to the types of erosion and areas affected identified in the
Inventory of Slopes prepared by Dr Mende . This allows accurate
estimation of an upper bound annual erosion rate applicable not just to the
monitored slopes, but to every slope along the entire length of the Road.
4.8. The 2014 UCR R eport includes erosion measurement and monitoring
results for a total of 11 sites (sites 6 and 7 were not used for technical
reasons explained in the 2014 UCR Report) over a two year period ( Table
4.1 and Figure 4.3, overleaf).
18
162Table 4.1UCR erosion monitoring and measurement sites (from the 2014 UCR Report).
Site number
UCR Mende Description Coordinates
(Dr Kondolf)
Large rotational landslide 84°21'43.571" W
1 N/A on cut slope.
Un-mitigated. 10°59'30.461" N
Large rotational landslide
84°20'45.712" W
2 T-33 on cut slope. 10°56'55.931" N
Un-mitigated.
Gully on cut slope. 84°20'27.579" W
3 T-37 Un-mitigated. 10°56'50.991" N
4 T-42 Rills on cut slope. 84°19'33.653" W
Un-mitigated. 10°55'15.459" N
84°20'07.509" W
5 T-39 Sediment trap.
10°56'27.451" N
Sediment trap (not used). 84°19'26.847" W
6 C-29 Refer to 2013 report for
details. 10°55'07.199" N
Sediment trap (not used). 84°18'18.025" W
7 T-58a Refer to 2013 report for 10°54'50.528" N
details.
Gully on fill slope. 84°21'19.775" W
8 T-8b Partially mitigated.
10°59'26.769" N
Gully on fill slope. 84°18'21.896" W
9 T-57a Un-mitigated. 10°54'52.695" N
10 T-45b Rills on fill slope. 84°19'31.562" W
Un-mitigated. 10°55'09.799" N
11 Large gully on fill slope. 84°17'22.664" W
T-65
(8.1) Un-mitigated. 10°54'24.191" N
12 Large gully on fill prism. 84°17'02.137" W
(9.4) T-68 Mitigation imminent.
10°53'39.912" N
13 Large gully on fill prism. 84°16'54.725" W
(9.5) T-70 Mitigation imminent. 10°53'35.477" N
19
163 Figure 4.3 Location map showing UCR erosion monitoring and measurement sites (from
the 2014 UCR Report).
4.9. Data analysis techniques applied to derive slope erosion data and details of
how the outputs from terrestrial LiDAR scanning and stereo
photogrammetry were processed are described in the 2014 UCR Report.
4.10. The method used to estimate sheet erosion was refined from that used in
2013. The underlying principle remains using the sediment trap at site 5 to
capture sediment eroded from a typical cut slope that was subject only to
sheet erosion, but the method used to measure the depth of accumulated
sediment was enhanced. In August 2014 an excavator was used to empty
the sediment trap and a Total Station was used to establish its dimensions
and topography . Stakes with which to record the depth of sediment
throughout the trap and around its edges were installed and surveyed -in
(Figure 4.4). Sediment surface elevations relative to the stakes were
20
164 measured on two subsequent dates in October 2014. Digital elevation
models of the sediment surface were then differenced to calculate the
volume of sediment that had accumulated in the trap between visits.
(a)
(b)
Figure 4.4 (a) Sediment trap at Site 5 on October 1, 2014. (b) Survey points and contour
map of sediment surface elevations on October 22, 2014 (from the 2014 UCR Report).
4.11. Erosion measurements and rates recorded for each of the monitoring sites
are listed and fully described in the 2014 UCR Report together with site-
specific details and observations of the erosion features monitored . UCR
summarize the outcomes of their comprehensive, multi- year, multi- site
field campaign in Table 7 of their 2014 Report, which is reproduced here
as Table 4.2.
Table 4.2Maximum annual erosion rates (from the 2014 UCR Report).
Erosion type Fill slope Cut slope
erosion rate erosion rate
(m/yr) (m/yr)
Rotational landslide 0.40 a 0.40
Gully 0.76 d 0.27
b
Rill 0.16 0.16
Sheet 0.14 c 0.07
a. As no deep-seated, rotationallandslides were measured in fill slopes, the cut slope landslide
erosion rate is recommended.
b. The 2013 report conservatively used the same erosion rate for rills in cut slopes and fill slopes and
this has been repeated in this report. The estimated erosion rate for rills in fill slopes is lower (0.07
m/yr.) therefore the higher erosion rate recorded in cut slopes (0.16 m/yr.) hasbeen conservatively
recommended for both sites.
c. Recommended sheet erosion rate is estimated by doubling rate measured for a cut slope to account
for uncompacted condition of soil in fill prisms.
21
165 The equivalent Table from the 2013 UCR Report (Table 4.3) is also shown
for comparison.
Table 4.3Summary erosion monitoring results in the 2013 UCR Report.
Type of Erosion Eroded Average Average rate of land
feature type Area/Area of erosion depth surface lowering
Feature (%) (m) (m y )
Cut Slope Landslide 13 0.38 0.19
Cut Slope Gully 2 0.01 0.005
Cut Slope* Rill 50 0.12 0.06
Road bed and Sheet 100 0.02 0.095
Cut Slope
Fill Slope Gully 9 0.10 0.20
4.12. It should be noted that, as in the 2013 Report, UCR recommend using the
highest erosion rate measured for each type of erosion and slope, so the
rates listed for each type of erosion represent upper bound mean annual
rates of erosion measured at the monitoring sites over the monitoring
period and, hence, the results obtained are conservative.
4.13. In comparing the data in Tables 4.2 and 4.3, it is important to note that the
way in which average annual rates of erosion are expressed differs
between them, for the reasons set out in paragraph s 4.6 and 4 .7 above.
The erosion rates in Table 4 .2 are higher than those in Table 4 .3 only
because the volume of sediment eroded is divided by the area of the
erosion feature, rather than the entire area of the slope affected . In fact,
UCR report that, in 2014, erosion has slowed at most monitoring sites
compared to that measured in 2013.
4.14. Dr Kondolf critiques site selection in the 2013 UCR field campaign, and
opines on page 52 that, “there is no scientific justification for applying the
depths measured in small gullies to la rge gullies”. In the 2014 UCR Report,
Costa Rica’s expert engineers explain the basis upon which their long -term
monitoring sites were selected. Selection was based on pragmatic and
22
166 practical decisions that are sensible and justified. Dr Kondolf’s criticism of
the sites selected for monitoring , underpinned by his assumption on page
48 that the large size of gullies observed at his SES sites 8 and 9 necessarily
makes them “more severely eroding” than those monitored by UCR, is used
to support his assertion that UCR’s measured data under-represent
erosion rates for slopes located along the full length of the Road.
4.15. Application of stereo photogrammetry to survey and estimate rates of
erosion in large gullies at Dr Kondolf’s sites 8.1, 9.4 and 9.5 (in the least
accessible stretch of Route 1856, between the Río Infiernito and Boca San
Carlos) now allows testing of Dr Kondolf’s assumption that these large
gullies have higher average erosion rates and, hence, his assertion that
UCR rates are unrepresentatively lo w. The results obtained using stereo
photogrammetry and terrain analysis are listed in Table 4.4.
Table 4.4Measured data for fill slope gullies at UCR Sites 8, and 11-13 (from the 2014
UCR Report).
Gully area Volume eroded Annual Erosion rate
Site (m ) (m ) (m/y )*
8 86 101.4 0.76
9 18.4 8.7 0.30
11 (8 .1) 174 134. 5 0.22
12 (9 .4) 500 659. 9 0. 38
13 (9 .5) 720 303.1 0.12
*n ote : that the annual erosion rate at Site 8 is double that at the most
rapidly eroding of the Sites mentioned in the 2014 Kondolf Report .
4.16. These measurements reveal that while the gullies at Dr Kondolf’s sites 8.1,
9.4 and 9.5 are indeed larger and have eroded greater volumes of sediment
than gullies formed in fill slopes monitored at UCR Sites 8 and 9, the mean
annual erosion rates (that is their volumes eroded divided by their
planform areas, divided by their age, i.e. how much they have lowered the
ground surface in a year) at Dr Kondolf’s sites 8.1, 9.4 and 9.5 are actually
23
167 much lower than that measured at Site 8 , and are comparable to those
measured at Site 9.
4.17. The rates of erosion measured at Dr Kondolf’s sites 8.1, 9.4 and 9.5 arealso
comparable to the erosion rate reco mmended in the 2013 UCR Report,
which was 0.2 m/y (as listed in Table 4.2). Hence, Dr Kondolf’s conclusion
that the rates UCR reported in 2013 were unrepresentatively low is not
supported by the measurements madeat his sites 8.1, 9.4 and 9.5 in 2014.
4.18. In Figure 21 of the 2014 Kondolf Report, Dr Kondolf identifies several
locations he classifies as ‘ severely eroding sites’ (SES) that were not
included in UCR’s programme of long-term monitoring and measurement.
These include his SES sites 10 to 17. The key point is that UCR selected
their long-term monitoring sites specifically to be representative of
erosion by sheet flow, rills, gullies and landslides that were accessible
during all seasons. As UCR’s measurements at Dr Kondolf’s SES sites 8 and
9 now demonstrate, the largest features do not lower the land surface the
fastest. In fact, the most rapid land surface lowering is driven by mid -sized
erosion features, where the forces driving a particular erosion process are
concentrated in relatively small area. As the area of a feature increases, the
average rate at which it lowers the land surface within that area tends to
decrease, because forces driving the erosion process become diffused and
the shear stress responsible for sediment entrainment (that is the
tangential force per unit area) decreases. T he results of precise
measurements made using photogrammetric analysis of stereo pairs of
high definition, orthogonal aerial photographs of Dr Kondolf’s sites 8.1, 9.4
and 9.5 prove that long-term average rates of land surface lowering by the
large gullies that existed at those site prior to mitigation by CONAVI in late-
2014 were lower than the that measured by UCR using terrestrial LiDAR at
UCR’s Site 9. The rates of land surface lowering measured at Dr Kondolf’s
24
168 sites 8.1, 9.4 and 9.5 are comparable to the rate recommended in the 2013
UCR Report and lower than that recommended in the 2014 UCR Report
(Table 4.4).
4.19. UCR considered making measurements at the rest of Dr Kondolf’s SES but
did not do so for rea sons explained in detail in the 2014 UCR Report. In
summary: Dr Kondolf’s Site 8.2 was exclud ed because it was not possible
to isolate different types of erosion ; Dr Kondolf’s Site 9.6 was not
measured as erosion was already being mitigated at the tim e of the
photogrammetric survey; no rates of erosion were provided in either the
Kondolf or Hagans and Weaver Reports for Sites 10 -17 (estimates were
provided only for Sites 8.1, 8.2, 9.4, 9.5, and 9.6), and these sites were also
out of range of the UAV . UCR report that, based on their observations of
sites 10-17, it appears unlikely that rates of erosion at these sites would be
any higher than th ose recommended in their 2014 Report. I concur and
conclude that erosion rates measured at the sites monitored by UCR
reasonably represent those at all 201 slopes along the Road.
4.20. Some confusion is apparent in the Kondolf Report concerning the existence
of landslides on fill slopes (see page 45 of the Kondolf report ). In his
commentaries Dr Kondolf interprets erosion on fill slopes as being driven
by the juxtaposition of gullies and landslides. If Dr Kondolf is referring to
deep-seated, rotational landslides resulting from gravitational slope
instability, this is a mis interpretation. As Dr Kondolf correctly notes, the
sediment in fill slopes is unconsolidated. The cohesive strength of the fill
material is consequently low. As a result, fill slopes are at or less than the
angle of repose. Deep seated , rotational landslides do no occur on slopes
with these geotechnical and geometric properties. What actually happens
is that fill prisms along the Road are eroded by gullies that incise deeply
into the fill, creating steep scarps and side slopes that slough into the gully
25
169 by toppling and/or shallow sliding. These gully head and bank failures are
driven by fluvial scouring at the base of the head and sides of the gully . As
such, they are part of the gully erosion process and are not conventional
landslides. This is why what Dr Kondolf identifies as ‘fill slope landslides’
are l ocated at the heads or margins of fill slope gullies . Dr Kondolf’s
misidentification of the occurrence of deep-seated, rotational landslides on
fill slopes undermines the credibility of the Kondolf R eport as a treatise on
slope mechanics compared to the Reports produced by UCR’s geotechnical
engineers.
B. Estimated annual erosion of sediment from cut and fill slopes under a
‘worst case’ rainfall scenario
4.21. Inventorying cut and fill slopes in 2013 posed significant challenges to Dr
Mende in his attempt to survey about 200 slopes and inspect well over a
hundred watercourse crossings along the 108 km between Marker II and
Delta Costa Rica. These challenges were compounded by the fact that the
work had to be performed at a time of year characterized by inclem ent
weather that made making measurem ents in the field difficult, rendered
some stretches of the Road impassable by vehicle and made others
inaccessible even on foot.
4.22. In 2014 a new field campaign was therefore planned with the aim of
upgrading the inventories of slopes and watercourse crossings that were
presented in the 2013 Mende and Astorga Report.
4.23. The 2014 field campaign was undertaken during the dry season, additional
resources were allocated to support Dr Mende in the field, and accurate
measurements of slope geometry (width, height, length , angle) were made
possible through the use of a Laser Hypsometer Nikon Forestry Pro , which
is a hand-held, electronic rangefinder. This device may be used to measure
26
170 distances between 10 and 100 metres with an accuracy of +/ -0.5 m, and to
measure elevation differences to and accuracy of +/- 0.2 m. Full details of
its mode of operation may be found at http://www.forestry-
suppliers.com/product_pages/Products.asp?mi=7852&title=Nikon%C2+F
orestry+PRO+Laser+Rangefinder%2FHypsometer.
4.24. The Mende Report uses the updated upper bound mean annual erosion
rates recommended in the 2014 UCR Report, together with new field
measurements of slope heights and areas, to derive accurate upper bound
estimates of the volumes of sediment eroded from cut and fill slopes along
the entire length of the Road between Marker II and Delta Costa Rica.
Given that the erosion rates presented in the 2 014 UCR Report are
conservative for the reasons explained above, the results obtained for each
slope in the Mende Report are also conservative.
4.25. On pages 52 to 56 of his 2014 Report, Dr Kondolf criticizes the way that
slope heights were estimated in the field in 2013 and how er oded areas
were subsequently calculated and volumes of erosion estimated by Mende
and Astorga in their 2013 report. Having taken account of these comments,
the 2014 Mende Report incorporates accurate measurements of slope
height and hence its results are more robust.
4.26. The 2014 Inventories also identify the mitigation status of each slope and
crossing, categorizing these features on the basis of whether mitigation to
prevent erosion and/or eroded sediment from reaching the River is
complete, in progress, scheduled, or unnecessary. The category of ‘other’
has been added to cover minor crossings in stretches where the Road
exists only as a trail accessible on foot or horseback. The outcomes of this
assessment of the mitigation status of slopes and watercourse crossings
are considered further in Section 7, below.
27
171 4.27. Dr Mende took advantage of the additional time and resources made
available in 2014 by inspecting every slope and watercourse crossing in
person, effectively walking the entire trace of the Road (including those
stretches where it is no more than a trail ) between Marker II and Delta
Costa Rica in the process – a distance of around 108 km. C lose scrutiny
possible only through this on-the-ground survey resulted in him adding
eight small watercourse crossings and five more slopes to the 2013
Inventories. Based on his closer inspection in 2014, Dr Mende also chose to
sub-divide some complex slopes to improve the way they are presented in
the Inventory. Consequently, slope T -83 has been subdivided into six
segments (T-83a to T-83f); T-114 into three segments (T-114a to T-114c);
and T-161 into two segments (T-161a, T-161b).
4.28. The first outcome of the 2014 field campaign led by Dr Mende was, then, to
increase the number of entries in the inventory of slopes from 188 to 201,
and the number of entries in the inventory of watercourse crossings from
121 to 129. These slopes and crossings are not distributed evenly between
Marker II and Delta Costa Rica. Over 60% of slopes and crossings are
located in the 41.6 km stretch between Marker II and Boca San Carlos,
while a third are located in the 43.6 km stretch between Boca San Carlos
and Boca Sarapiqu í. There are only ten crossings and four slopes in the
22.6 km stretch between Boca Sarapiqu í and Delta Costa Rica (see Table
7.2, below).
4.29. Walking from Marker II to Delta Costa Rica provided the opportunity for
Dr Mende to observe and map the condition of the road surface, classifying
this as being a ‘gravel road’, a ‘dirt road’ or a ‘trail’ (Figure 4.5 ). Along
stretches categorized as a ‘gravel road’, the surface of the Road is protected
from erosion because it is covered by gravel-sized rocks. Along stretches
categorized as a ‘dirt road’ its surface is unprotected. Along stretches
28
172categorized as a ‘trail’ the Road does not exist except as a narrow track
passable on foot or horseback. In Section 6 (page 39) of th e 2014 Kondolf
Report, Dr Kondolf made it clear that he did not believe that the surface of
most of the road is covered in gravel . Dr Mende’s observations and map
show that it is (Figure 4.5).
29
173 % % %
= 33
km
8.0 km = 7
64.5 3511 km
30
PrGsDirTraildlion of the Road
the Road between Marker II and Delta Costa Rica derived from on the ground inspection by Dr
Current condition of
5
(from the Mende Report).
Figur. 4.
1744.30. The Mende Report presents accurately measured , detailed information on
the relative frequencies of the different types of erosion acting on slopes
along the Road (Table 4.6).
Table 4.5Types of slope erosion observed along the entire length of the Road between
Marker II and Delta Costa Rica (from the Mende report).
Erosion type Slope area (ha) Slope area
(%)
Sheet erosion 16.5 53
Rill 6.4 21
Gully 5.9 19
Landslide 1.9 6
None 0.2 1
Totals 30.9 100
4.31. These observations demonstrate that the prevalent erosion process on the
slopes along the Road is sheet erosion. Rills and gullies each occupy about
a fifth of the overall area of the slopes. While landslides feature in most of
the slopes photographed in Dr Kondolf’s 2014 Report , they occupy only
6% of the overall area of slopes. Hence, any impression that landslides are
prevalent along the entire length of the Road is inaccurate.
4.32. The annual erosion rates recommended by UCR (listed in Table 4.2 above)
were accepted and applied without modification, as demonstrated in Table
9 on page 32 of the Mende Report and Table 4.7, below. As noted in
paragraph 4.12 above, in their 2014 Report, UCR again recommend using
the highest mean annual erosion rate for each type of erosion and slope, so
the rates listed for each type of erosion represent the highest mean annual
rate of erosion measured at the monitoring sites over the monitoring
period. In this sense they are, therefore, conservative.
4.33. In the Mende Report, the updated rates of erosion supplied in the 2014
UCR Report are combined with the new field measurements of slope
geometry made using more accurate techniques and instruments to
31
175 estimate the annual volume of sediment eroded from each of the 201
slopes in the inventory . In these calculations, at each slope, the upper
bound erosion rate recommended by UCR for each erosion process
(landslides, gullies, rills and sheet erosion) is multiplied by the area
recorded in the Inventory of Slopes as exhibiting that erosion process, and
the volumes eroded by each process are summed to estimate the total for
the slope as a whole. The annual volume of erosion calculated at each
slope is, therefore, also an upper bound, conservative value.
4.34. An example of the calculation procedure for Slope T -8a, which exhibits all
four erosion processes, is presented in Table 4.7.
Table 4.6Example calculation of annual erosion volume for Cut Slope T-8a.
Erosion type Cut slope Slope area Estimated annual
erosion rate affected volume of erosion
(m/y) (m ) (m /y)
Sheet erosion 0.07 185 13
Rills 0.16 554 89
Gullies 0.27 369 100
Land Slides 0.40 739 296
Totals -- 1,847 497
4.35. Applying this approach to all 201 slopes along the Road between Marker II
and Delta Costa Rica, Dr Mende estimated the total volume of slope erosion
to be just under 72,000 m /y.
4.36. This volume is based on a scenario in which all four erosion processes
operate at their upper bound rates, simultaneously at every slope along
the entire length of the Road. F or erosion of 72,000 m 3 actually to occur in
one year , it would require rainfall sufficiently heavy, frequent and
widespread to maximize annual erosion rates along the entire length of the
Road, which is improbable for the meteorological reasons explained in
Section 4D, below. Hence, I believe this to be a 'worst case' rainfall scenario
for slope erosion along the Road and one that is actually very unlikely to
32
176 occur, making it a highly conservative estimate. Also, no account is taken
of reductions in slope erosion resulting from the programme of erosion
mitigation performed by CONAVI and CODEFORSA, which has progressed
significantly since 2013 (see Section 7, below) . It follows that the annual
slope erosion volume of 72,000 m /y produced by Dr Mende is very much
a ‘worst case’ value, not a mean annual average value.
4.37. The slope erosion volu me estimated in 2014 is nearly double that
estimated in 2013, which was 36,590 m 3/y (or 61,100 t/y). The increase
results from Dr Mende’s underestimation of slope surface areas in 2013.
Use of improved instrumentation in the 2014 field campaign allowed him
to measure slope dimensions precis ely instead of estimating them , to
produce more accurate results, e specially for those fill slopes where the
road is situated at the top of the slope. Applying the same conversion rate
for cubic metres of sediment to metr ic tonnes of 1.67 t/m 3 used in 2013,
72,000 m /y converts to almost exactly 120,000 t/y.
4.38. I note that the estimates of slope erosion by landslides, gullies and surface
erosion in the 2014 Kondolf Report (1 47,515 – 158,515 m /y, reported on
page 61) have also changed compared to those in the 2012 Kondolf Report
3
(218,400 – 273,000 m /y, reported on page 46 of that Report) , with the
2013 estimates being 1.5 to 1.7 times greater than those suggested in the
2014 Kondolf Report. Dr Kondolf’s 2014 estimates are based on applying
the average erosion rate for rills, gullies and landslides reported in the
2013 UCR Report (0.558 m/y) to all sites identified by Dr Kondolf as
‘severely eroding’. This erosion rate r eplaces that assumed in the 2012
Kondolf Report (1 m/y), which was much too high.
4.39. In In his Table A (on page 53 of the 2014 Kondolf Report), Dr Kondolf
notes a discrepancy between the erosion rates recommended in the 2013
33
177 UCR Report and those applied in the 2013 Mende and Astorga Report. As
explained by Dr Mende, this discrepancy has now been corrected and the
updated 2014 UCR erosion rates are applied in the 2014 Mende Report. In
any event, I note that in 2013 more conservative, i.e. higher, estimates of
erosion rates were applied by Mende and Astorga, resulting in a higher
estimate of erosion from the Road to the River. F urthermore, this
discrepancy had no impact on my overall conclusions in the 2013 Thorne
Report because, in establishing the relative contribution of Road -derived
sediment to the River and whether this was sufficient to cause harm to its
water quality, morphology, environment or ecosystems , I applied Dr
Kondolf’s 2013 estimate of erosion from the Road to the River, which was
substantially higher.
4.40. I am confident that the revis ed estimate of 72,000 m 3/y (equivalent to
120,000 t/y) proposed in the 2014 Mende Report represents a reliable,
‘worst case’ estimate of the annual erosion rate for slopes along the Road
between Marker II and Delta Costa Rica, because:
(a) it is based on two years of field monitoring and measurements using
accurate technologies;
(b) upper bound, measured erosion rates are applied to all the slopes
along the Road simultaneously; and
(c) no reduction is made for the mitigating effects of CONAVI and
CODEFORSA’s slope stabilizing work (which now reduces erosion at
over halfof the slopes requiring mitigation).
In summary, I believe this figure to be a highly conservative estimate,
representative of erosion under an unlikely, ‘worst case’ rainfall scenario.
34
178 C. The contribution of the sediment eroded from the Road to the
sediment load and average annual sediment budget of the River
4.41. In 2013, hydrologists and hydraulic engineers with t he Costa Rican
Institute of Electricity (ICE) prepared a report that inv estigated whether
the Road has , or could have , significant impacts on the hydrology or
sedimentology of the R ío San Juan, drawing on the best information
available at that time. In 2014, the work presented in the 2013 Report was
re-visited, taking advantage of new and better information gathered in the
interim, and applying more advanced analyses to refine the outcomes and
increase confidence in their accuracy.
4.42. Based on the outcomes of hydrological analyses presented in the 2013 ICE
Report, it was concluded in the 2013 Thorne R eport (page 65, paragraph
7.17) that, “there is no possibility that the Road has had, will have, or indeed
could ever have any measurable impact on the hydrology of the Río San
Juan”. This conclusion was not challenged in the 2014 Kondolf Report and
is restated here as confirmation that it remains my belief.
4.43. In light of this, the 2014 ICE report focuses on sediments, using advanced
techniques in the hydrological, river and engineering sciences to better
model catchment sediment erosion and yields, and to reevaluate measured
and calculated sediment transport in the Río San Juan, with a special
emphasis on establishing the best possible sediment budget. The 2014 ICE
Report also pays particular attention to identifying and quantifying natural
variability and uncertainty which are unavoidable and characteristically
high in sediment measurement, estimation and modelling.
4.44. ICE’s work in 2014 addresses the view that, “When Professor Thorne says
that the contribution of sediment from the Road is insignificant, he is not
comparing that contribution to a figure that accurately represents the
35
179 sediment load of the Río San Juan.”, which is made on page 66 of the 2014
Kondolf Report in a subsection titled, “ Costa Rica’s Experts Compare the
Road’s Contributions to Unreliable Total Load Figures”. Dr Kondolf does not
support this statement with technical criticisms of the way that ICE in their
2013 Report estimated the annual average suspended load of the R ío San
Juan, which makes up much the greater part of the total load.
4.45. However, Dr Kondolf does criticise the way that ICE estimated the bedload,
relying on views advanced in the Andrews Report, that ICE used an
exaggerated value for the slope of the river in calculating the bedload using
the Einstein bedload function. This criticism is conflated with a further
criticism (on page 67 of the 2014 Kondolf Report) that values of slopes for
sub-reaches of the river listed in Table 1 of the 2013 Thorne Report are,
“overstated by factors of about 55 to 58”. On the same page, Dr Kondolf then
concludes that, “The implications of this error are significant, in that channel
slope is a fundamental variable of rivers, which affects many river process,
including bedload transport, whose calculation can be distorted by use of
erroneously large slope values.”
4.46. The slopes listed in Table 1 of the 2013 Thorne Report are correct, but
they are expressed in degrees r ather than being dimensionless - as
incorrectly indicated in the column heading in Table 1. Any confusion the
error in the heading may have caused is regrettable. T his was purely a
labelling error and it had no bearing on calculations performed by ICE in
estimating the bedload component of the total load of the Río San Juan.
4.47. Having clarified this, the remaind er of this sub -section reports significant
advances in the analyses presented in the 2014 ICE Report (compared to
the 2013 ICE Report) that improve the accuracy of the suspended, bed and
total sediment loads estimated for the Río San Juan and, hence, the
36
180 reliability of the sediment budget. The 2014 ICE Report and my comments
in this sub -section also address Dr Andrews’ over- arching criticism (on
page 13 of the Andrews Report) that the 2013 Thorne and ICE Reports
include, “numerous examples of insufficient and poor quality hydrologic
information, incorrect and improper analysis, and unsupported or wrong
conclusions”. This criticism is inconsistent with the facts that the Andrews
Report; (1) itself relies heavily on the hydrologic information provided by
the 2013 ICE Report , (2) misconstrues data extr acted from the literature
on sediment yields from tropical basins, (3) constructs an implausible
‘natural’ sediment load for the Río San Juan, and (4) demonstrates a lack of
knowledge concerning sediment processes in the San Juan Basin, including
especially those in its mountainous, headwater basins and its coastal delta.
4.48. Before presenting the enhanced analyses performed by ICE in 2014, it is
necessary to note that it was necessary for ICE to use data from
hydrometric stations on tributaries to the Río San Juan draining from Costa
Rica, rather than using data from the R ío San Juan itself. This is the case
because Costa Rica is unable unilaterally to measure discharges and
sediment loads in the Río San Juan and , notably, Nicaragua’s experts
choose not to do so, or indeed to supply any measured discharges or
sediment loads to support any of their statements regarding the
significance of Road-derived sediment in the context of the sediment load
currently carried by the Río San Juan.
4.49. With respect to estimation and budgeting of the mean annual suspended
load, advances in 2014 include:
(a) greater spatial density and congruence in the meteorological and
hydrological information used as inputs to the 2014 analyses;
37
181 (b)longer records from discharge and sediment measurement stations and
adoption of power functions to generate sediment rating curves (as
recommended by Dr Andrews);
(c) enhanced uncertainty analysis throughout, including in the calibration
processes;
(d)better assessment of suspended sediment inputs to the R ío San Juan at
Boca San Carlos and Boca Sarapiquí made using the probabilistic flow
duration curve method of Krasovskaia and Gottschalk (2014) and the
sediment duration curve approach proposed by Garcia (2014);
(e) development of a spatially-distributed model of erosion in the San Juan
Basin downstream of L ake Nicaragua that improves on the 2013
CALCITE model by taking advantage of improvements in spatial
resolution, hydrological congruence , calibration (which is now
uncertainty weighted) and sensitivity to the use of alternative functions
for the delivery index; and
(f) enhanced consideration of uncertainty in the USLE model so that it can
better be accounted for in calculating the annual sediment budget and
variability therein.
4.50. With regard to estimation of the bedload, advances in 2014 include:
(a) improved bedload calculations using the Engelund-Hansen approach
(as recommended by Dr Andrews); and
(b) enhanced consideration of uncertainty in the estimated bedload.
4.51. The distributed sediment model was used to generate an upgraded
sediment budget for the Río San Juan and its main tributaries in a manner
38
182 similar to 2013 . The outputs of the 2014 UCR and Mende Reports were
then applied to determine the contribution of sediment eroded from Route
1856 to annual average sediment loads in the Río San Juan, lower Río San
Juan and Río Colorado.
4.52. Perhaps the most s triking finding of the 2014 ICE study and Report is the
very high level of inter-annual variability identified in the annual average
suspended sediment loads, which can be trac ed back to natural variability
in the hydrological and sediment records used to construct the sediment
budget for the Río San Juan and uncertainties in the spatial and regression
models. This is significant because it demonstrates that spatial and inter -
annual variability in the sediment load of the R ío San Juan is so large that
the sediment input from the Ro ad, even at its potential highest, is
inconsequential in comparison.
4.53. In constructing th e 2014 sediment budget for the Río San Juan, the first
step was to assemble the base physiographic data needed to build a
distributed model of soil erosion in the study area, which is the basin
between Lake Nicaragua and Delta Costa Rica ( Figure 4.6(a)). The study
area comprises 13 tributary basins (Fi gure 4.6(b)), six in Nicaragua and
seven in Costa Rica.
39
183 (a) (b)
(c) (d)
(e) (f)
Figure 4. 6 (a) Study area of the R ío San Juan Basin (b) tributary basins (c) digital
elevation model (d) mean annual precipitation (e) l and cover map (f) soil classification
map (based on maps presented in the 2014 ICE Report).
4.54. Topography was mapped using a Digital Elevation M odel (DEM) with a
30 m grid size, derived from official 1:50,000 maps of Costa Rica and point
data extracted from the ASTER GDEM for Nicaragua (Figure 4.6(c).
4.55. Mean annual precipitation was mapped by the National Institute of
Meteorology using rainfall data for Costa Rica and published INETER data
for Nicaragua, all collected between 1971 and 2000 (Figure 4.6(d)).
4.56. Soils were mapped using the USDA soil taxonomy classification
(Figure 4.6(e)), based on the soil map published by the Costa Rican
40
184 Association of Soil Science (2013) and digitalized information from
INETER (2008, p. 58).
4.57. Land cover was mapped using RapidEye satellite imagery for Costa Rica
2009-10 and visual interpretation of available thematic land-cover
imagery for the Nicaraguan part of the study area (Figure 4.6(f)).
4.58. As in 2013, the Universal Soil Loss Equation (USLE) was used to estimate
potential soil erosion (E) throughout the study area , based on factors
representing: crop and cover manag ement (C), soil erodibility (K), slope
length and steepness (LS), rainfall erosivity (R) and conservation practice
(P), which was taken to be unity (meaning that no allowance is made for
reductions in potential erosion due to soil conservation practices) . USLE
does not account for catchment erosion by gullies and landslides, and so it
is expected to under -estimate actual erosion in the San Juan Basin where
these processes are widely observed. The contributions of sediment from
gullies and landslides are acc ounted for when the sediment budget is
balanced using measured loads in the R ío Colorado and at the mouths of
the Río San Carlos and R ío Sarapiquí (as explained in paragraph 4.75 and
shown in Table 4.18, below).
4.59. The other four factors were calculated using the information mapped in
Figure 4.6and then mapped in a GIS.
(a) The land cover factor, C, accounts for differences in potential soil
erosion depending on vegetation and land use. There is no unique C value
for a given land cover and, in their report, ICE recount in detail how they
investigated alternatives for representing this variability before selecting an
asymmetrical probability density distribution to account for uncertainty in
the C factor. Mean values of ‘C’ are mapped in Figure 4.7(a). The probability
distribution selected has the f orm of a bell- curve (similar to that for a
41
185 normal distribution) centered on the most likely (modal) value, but which
differs in that the range of possible values is spread asymmetrically around
the mode. Accordingly, the bell curve is skewed rather than being
symmetrical, as would be the case for a normal distribution.
(b) Information on soil categories was used to derive maximum,
minimum and mean values of the soil erodibility factor, ‘ K’ together with
coefficients of variation estimated using an asymmetrical probability
density distribution to account for uncertainty . Mean K-factor values are
mapped in Figure 4.7(b).
(c) The rainfall erosivity factor, ‘ R’ was calculated using over sixty
thousand rainfall events recorded at 52 ICE meteorological stations
between 1995 and 2014, by linking the results to mean annual precipitation
using an empirical power function specific to the study area, with the 95%
confidence and prediction intervals used to represent uncertainty in this
relationship. Predicted R-factors are mapped in Figure 4.7(c).
4.60. The slope length parameter, ‘LS’ in the USLE was estimated and mapped
using the Digital Elevation Model (Figure 4.7(d)). ICE could not obtain the
information necessary to characterise uncertainty in this factor and so it is
not accounted for in the uncertainty analysis . Consequently, uncertainties
in values of potential soil erosion estimated by ICE are more likely to be
under rather than over-estimated.
4.61. Values of potential soil erosion (E) in the study area of the R ío San Juan
basin calculated using the USLE are mapped in Figure 4.8.
42
186 (a) (b)
(c) (d)
Figure 4.7 (a) land cover factor, C (b) soil erodibility factor, K (c) rainfall erosivity, R and
(d) slope length factor LS.
Figure 4.8 Distribution of potential soil erosion in the R ío San Juan basin between Lake
Nicaragua and Delta Costa Rica (based on maps presented in the 2014 ICE Report).
4.62. Uncertainty in potential soil erosion was quantified using a partial
derivatives approach suggested by Singh et al. (2007) and the probability
43
187 density distributions derived for C, K and R. The impacts of uncertainty on
potential erosion were explored using a catchment-aggregated approach
(in the manner of the CALCITE model used in 2013) compared to a fully-
distributed application of the USLE. The result s (Table 4 .8) reveal that
while uncertainty is reduced using the spatially distributed approach
adopted in 2014, it cannot be eliminated entirely.
4.63. Not all of the soil eroded according to the USLE will be transported out of
the catchment. Values of the delivery index estimated using the delivery
index developed by Bradbury (1995) are mapped in Figure 4.9.
Figure 4.9 Spatial distribution of the delivery index in the study area (from the 2014 ICE
Report).
4.64. Records of observed catchment sediment yield available from the 14
hydrologic and sediment gauging stations in the Costa Rican part of the
study area were used to calibrate the soil erosion and delivery model and
the results are mapped in Figure 4.10 and listed in Table 4.9.
44
188 Table 4.7Potential erosion based on catchment-aggregated and fully-distributed
applications of the USLE (from the 2014 ICE Report).
Aggregated
Basin Model Fully Distributed
Mean USLE factors (e.g. CALCITE) Model
C K LS R Erosion CV Erosion CV
Las Banderas 0.022 0.013 0.91 22300 5.67 3.43 4.73 1.07
Machado 0.020 0.010 1.05 18500 3.97 3.29 3.86 0.90
Barlota 0.020 0.015 1.35 16700 6.49 3.52 6.73 0.59
Santa Cruz 0.065 0.015 1.25 16500 19.92 1.40 19.73 0.16
Sábalos 0.109 0.015 1.14 14100 25.39 1.13 26.02 0.09
Melchora 0.091 0.018 0.91 10100 14.80 0.91 16.19 0.15
San Carlos 0.095 0.015 1.31 21200 40.20 0.69 30.20 0.04
Cureña 0.055 0.014 0.51 20300 7.72 1.16 7.21 0.44
Sarapiquí 0.084 0.014 1.45 26800 47.00 0.76 32.80 0.04
Chirripó 0.105 0.018 0.20 21500 8.11 1.06 6.86 0.48
Frío 0.089 0.015 0.72 15000 14.51 0.83 13.74 0.08
Pocosol 0.086 0.016 0.46 15100 9.19 0.90 9.38 0.14
Infiernito 0.070 0.014 0.80 18600 14.93 1.17 12.79 0.19
Study Area 0.083 0.015 1.06 19900 25.92 0.52 21.58 0.03
Note: Erosion = best estimate (t ha yr ), CV = coefficient of variation
Figure 4.10 Calibrated specific sediment yields in the study area (from the 2014 ICE
Report).
45
189 Table 4.8Soil erosion and sediment yields in the Study Area of the San Juan Basin
(from the 2014 ICE Report).
DA SSE SE SSY SY
Basin (km ) (t ha yr )1 (t yr ) (t ha yr )1 (t yr )
Melchora 305 16.19 494 000 4.97 152 000
Sábalos 571 26.02 1 486 000 7.99 456 000
Santa Cruz 418 19.73 825 000 6.06 253 000
Barlota 219 6.73 147 000 2.07 45 000
Machado 352 3.86 136 000 1.19 42 000
Las Banderas 198 4.73 94 000 1.45 29 000
Frío 1577 13.74 2 167 000 4.22 666 000
Pocosol 1224 9.38 1 148 000 2.88 353 000
Infiernillo 609 12.79 779 000 3.93 239 000
San Carlos 2642 30.20 7 979 000 9.28 2 451 000
Cureña 353 7.21 254 000 2.21 78 000
Sarapiquí 2770 32.80 9 087 000 10.07 2 791 000
Chirripó 236 6.86 162 000 2.11 50 000
Study area 11474 21.58 24 758 000 21.58 7 605 000
Note: DA = drainage area; SSE = specific soil erosion; SE = soil erosion; SSY = specific sediment
yield; SY = sediment yield.
4.65. In the 2014 Report, ICE re -examined the measured suspended load s
carried annually by rivers in the San Juan basin, based on measurements
made at 14 hydrometric statio ns in the study area (Figure 4 .11) plus
additional sites at the mouths of the Ríos San Carlos and Sarapiquí.
Figure 4.11 Hydrometric stations with sediment records in the Río San Juan basin (from
the 2014 ICE Report).
46
190 4.66. Suspended sediment rating curves (SSRC s) were generated using
measured discharges and measured suspended sediment concentrations
for each of the fourteen stations. These measurements are made routinely,
using internationally recognized equipment and techniques. The relation
between discharge and suspended sediment load is customarily
represented as a power function (and this is also recommended by Dr
Andrews in his 2014 Report) and this approach was adopted in 2014 (see
Figure 4.12 for an example).
Figure 4.12 Suspended sediment rating curve for the R ío Colorado at Delta Costa Rica
(Station 11-04) between 2010 and 2014 (from the 2014 ICE Report).
4.67. The SSRCs were combined with discharge records to generate daily and
hourly SSL time series at each station . The average annual suspended
sediment load was then calculated as the integral of the SSL time series
divided by the length of record. This approach is not straightforward and ,
appropriately, the stringent steps necessary to avoid bias in the resulting
mean annual suspended sediment loads were taken. The results are listed
in Table 4.10 and described in detail in the 2014 ICE Report.
47
191 Table 4.9Mean annual suspended sediment loads for the 14 hydrometric stations
(from the 2014 ICE Report)
Annual suspended sediment load (t yr )
CODE Station
Mean TLCI TUCI SSRC LCI SSRC UCI
11-04 Delta Colorado 7 599 000 2 611 000 12 586 000 4 023 000 15 148 000
12-03 Puerto Viejo 161 000 141 000 182 000 140 000 186 000
12-04 Veracruz 86 000 37 000 135 000 62 000 123 000
12-06 Toro 12 000 7 000 17 000 8 000 18 000
12-11 San Miguel 22 000 12 000 33 000 13 000 40 000
12-13 Río Segundo 2 000 1 000 3 000 1 000 6 000
14-02 Jabillos 215 000 155 000 274 000 170 000 274 000
14-04 Terrón Colorado 1 175 000 988 000 1 362 000 783 000 1 806 000
14-05 Peñas Blancas 141 000 115 000 167 000 116 000 172 000
14-20 Pocosol 130 000 85 000 175 000 98 000 174 000
16-02 Guatuso 55 000 49 000 61 000 48 000 62 000
16-05 Santa Lucía 3 000 3 000 4 000 3 000 4 000
Note:
TLCI = lower 95% confidence interval due to time series variability;
TUCI = upper 95% confidence interval due to time series variability;
SSRC LCI = lower 95% confidence interval due to uncertainty in SSRC;
SSRC UCI = upper 95% confidence interval due to uncertainty in SSRC.
Table 4.10Confidence intervals as normalized anomalies for the mean annual suspended
sediment loads at the hydrometric stations (from the 2014 ICE Report)
CODE Station UTSV (normalized anomalies) USSV (normalized anomalies)
LCII UCI LCII UCI
11-04 Delta Colorado -66% +66% -47% +99%
12-03 Puerto Viejo -13% +13% -13% +15%
12-04 Veracruz -58% +58% -28% +43%
12-06 Toro -40% +40% -32% +47%
12-11 San Miguel -48% +48% -43% +79%
12-13 Río Segundo -28% +28% -61% +192%
14-02 Jabillos -28% +28% -21% +28%
14-04 Terrón Colorado -16% +16% -33% +54%
14-05 Peñas Blancas -18% +18% -18% +22%
14-20 Pocosol -34% +34% -25% +34%
16-02 Guatuso -11% +11% -12% +14%
16-05 Santa Lucía -12% +12% -21% +27%
Note:
UTSV = uncertainty due to time series variability;
USSV = uncertainty due to sample variability in the suspended sediment rating curve;
LCI = lower 95% confidence interval;
UCI = upper 95% confidence interval.
4.68. High uncertainty is evident in the wide 95% confidence intervals in Figure
5.12. This arises both from variability in the SSL time series and
48
192 uncertainty in the SSRCs. The high degree of uncertainty generated in the
resulting mean annual suspended sediment loads is obvious from the
lower and upper 95% confidence intervals listed in Table 4 .10, above and
expressed as percentages of the mean in Table 4.11.
4.69. It is not possible to construct empirical bedload sediment rating curves
(like those described above for suspended sediment loads), because
bedload transport measurements are unavailable. Consequently, measured
discharges and sampled bed sediment sizes available for the Río Colorado
at the Delta were used to generate a theoretical sediment rating curve
(Figure 4.13). In producing this rating curve, the channel slope was
estimated from the Engelund-Hansen hydraulic resistance relation (García
2007, p. 125). The resulting slope of 1.79 x 10 -4m/m for the Río Colorado
immediately downstream of the Delta is very close to that recommended
by Dr Andrews in his 2014 Report for the R ío San Juan immediately
-4 -4
upstream of the Delta (which is 1.7 x 10 to 1.5 x 10 m/m).
4.70. This sediment rating curve was then combined with the discharge time
series for Station 11-04 to generate a bedload time series , and uncertainty
parameters for this time series were estimated based on sample and time
variability. The resulting estimates for the mean annual bedload
transported by the R ío Colorado are listed in Table 4 .12. Confidence and
prediction intervals are included to illustrate that uncertainty in these
estimates is high.
49
193 Figure 4.13 Bed load rating curve for the Delta Colorado (11 -04) station (from the 2014
ICE Report).
Table 4.11Mean annual bedload transport in the Río Colorado at Station 11-04
(from the 2014 ICE Report). -1
CODE River Annual Bedload (t yr )
Mean TLCI TUCI BLRC LCI BLRC UCI
11-04 Colorado 2 898 000 719 000 5 077 000 1 798 000 4 809 000
Note:
TLCI = lower confidence interval due to time series variability;
TUCI = upper confidence interval due to time series variability;
BLRC LCI= lower confidence interval due to uncertainty in the bedload rating curve;
BLRC UCI= upper confidence interval due to uncertainty in the bedload rating curve.
4.71. The estimate of bedload transported by the Río Colorado in 2011-12 by Dr
Andrews (on page 25 Andrews Report) is 330,000 t/ y. This is much less
than the mean annual bedload estimated by ICE (2,898,000 t/y). It would
be expected that bedload in that particular water year would be lower than
average as runoff was lower than usual. Dr Andrews estimated that
bedload in that period sh ould be 2.2 times lower than average for this
reason, leading him to propose that th e mean annual bedload carried by
the Río Colorado is 730,000 t/y. This figure falls within the confidence
band for ICE’s estimate, being close to but a little larger than the lower
confidence interval of 719,000 t/y. The fact that Dr Andrews’ estimate is
near the lower edge of the confidence band probably results from
50
194 application of a different bedload f unction by D r Andrews, who selected
the Fernandez-Luque and van Beek equation, while ICE based their
calculations on the Engelund-Hansen approach.
4.72. However, there is reason to suspect that the basis upon which Dr Andrews
estimates inter -annual variability of the sediment load is flawed. The
sediment load in a river draining a tectonically active basin with live
volcanoes does indeed vary widely from year to y ear. However, this is not
only due to varying rainfall and runoff; it is also dependent on the
quantities of sediment supplied by major sources such as landslides
triggered by earthquakes. For example, in 2009, the Cinchona Earthquake
generated 349 landslides that disturbed 21.7 km 2 of formerly vegetated
land around the epicentre, releasing 4 to 6 million tonnes of sediment,
95% of which entered streams draining to the Río San Juan (Alvarado
2010). Consequently, Dr Andrews is mistaken to estimate inter -annual
variability in sediment loads solely as a function of the inter -annual
variability recorded in rainfall and runoff. In my opinion, the analyses
performed by Costa Rica’s experts at ICE, which are based on sediment
records from long -term measurement stations , more reliably establish
variability and uncertainty in estimates of the annual load of the Río San
Juan.
4.73. Annual bedloads at the mouths of the Río Sarapiquí and Río San Carlos had
to be estimated using a different approach as no hydrological records are
available for those specific locations. To overcome this lack of flow data,
ICE selected, tested and calibrated a probabilistic method using flow and
sediment records from the twelve hydrometric stations located in other
parts of the drainage systems of Río Sarapiquí and Río San Carlos and then
applied this to the mouths of the rivers. Full details are presented in the
51
195 2014 ICE Report and, having reviewed these, in my opinion the approach
applied is viable. The results are listed in Table 4.13.
Table 4.12 Hydrological data, hourly and daily coefficients of variation and mean annual
suspended sediment loads at the mouths of the Río Sarapiquí and Río San Carlos
(from the 2014 ICE Report).
CODE Station DA ( km ) Qa (m s )-1 CVD b CVH b SSL (t yr )
BSa Boca Sarapiquí 2 643 377 0.647 0.683 2 342 000
BSC Boca San Carlos 2 771 266 0.644 0.678 2 928 000
Note: DA = drainage area; Qa = mean annual discharge; CVD = daily coefficient of variation; CVH =
hourly coefficient of variation; SSL = annual suspended sediment load.
aBased on rainfall- area methodology. Based on coefficient of variation -area functions. Based on
modeled sediment duration curves.
4.74. As no data are available for current mean annual sediment loads in the Río
San Juan immediately upstream of the Delta or the lower Río San Juan
immediately downstream of the Delta, these were estimated based on
measurements made in the R ío Colorado at gauging station 11 -04
immediately downstream of the Delta. It is believed that about 90% of the
discharge of the Río San Juan passes to the R ío Colorado and so the
suspended and bedload rating curves constructed for station 11 -04, which
is in the Río Colorado immediately downstream of the Delta , are
reasonable approximations of those for the Río San Juan. As no bedload
data are available for the R ío San Juan, and given the high variability and
uncertainty associated with sediment loads in these rivers anyway, using
the bedload rating curve developed for the R ío Colorado to esti mate
bedload in the R ío San Juan is a tenable approximation and so estimating
the sediment load this way is reasonable.
4.75. On page 22 of the Andrews Report, Dr Andrews points out that, given the
data currently available, the division of discharges at the Delta, “cannot be
determined with any confidence”. The division of flows at the Delta couldbe
determined with confidence if Nicaragua or its experts measure d and
made known the discharge of the lower Río San Juan.
52
196 4.76. As Nicaragua has chosen not to measure discharge in the lower Río San
Juan it was necessary for ICE to account for uncertainty in the division of
flow at the Delta when calculating the mean annual sediment loads in the
Río San Juan and lower R ío San Juan . To do so, these calculations were
performed not only for the scenario that 90% of the flow passes to the Río
Colorado, but also for scenarios of 85% and 95%. The discharge time
series for the R ío Colorado was modified to reflect these three possible
divisions of the flow and then transformed into suspended and bedload
time series using the relevant rating curves. Results for the R ío San Juan
and lower Río San Juan are listed in Tables 4.14 and 4.15, respectively.
Table 4.13Mean Annual suspended and bedloads in the Río San Juan for a range of
percentage discharges flowing to the Río Colorado (from the 2014 ICE Report).
Qa Annual sedimentload (t yr )
PRSJ 3 -1
(m s ) Mean TLCI TUCI SSRC LCI SSRC UCI
Suspended load
95 1055 8 286 000 2 847 000 13 725 000 4 300 000 16 951 000
90 1114 9 078 000 3 119 000 15 036 000 4 598 000 19 153 000
85 1180 9 997 000 3 435 000 16 559 000 4 919 000 21 873 000
Bedload
95 1055 3 221 000 799 000 5 643 000 1 967 000 5 447 000
90 1114 3 600 000 893 000 6 307 000 2 157 000 6 227 000
85 1180 4 050 000 1 005 000 7 095 000 2 373 000 7 191 000
Table 4.14Mean Annual suspended and bedloads in the lower Río San Juan for a range of
percentage discharges flowing from the Río San Juan to the Río Colorado
(from the 2014 ICE Report)
Qa Annual sedimentload (t yr )
PRSJ 3 -1
(m s ) Mean TLCI TUCI SSRC LCI SSRC UCI
Suspended load
95 1055 687 000 236 000 1 139 000 277 000 1 803 000
90 1114 1 479 000 508 000 2 450 000 575 000 4 005 000
85 1180 2 398 000 824 000 3 973 000 896 000 6 725 000
Bedload
95 1055 323 000 80 000 566 000 169 000 638 000
90 1114 702 000 174 000 1 230 000 359 000 1 418 000
85 1180 1 152 000 286 000 2 018 000 575 000 2 382 000
Note:
PSJR = Percentage of Río San Juan discharge flowing to the Río Colorado;
Qa = Mean annual discharge;
53
197 TLCI = lower 95% confidence interval due to time series variability;
TUCI = upper 95% confidence interval due to time series variability;
SSRC LCI= lower 95% confidence interval due to uncertainty in SSRC;
SSRC UCI= upper 95% confidence interval due to uncertainty in SSRC.
4.77. Best estimates of mean annual loads in the Río San Juan are: suspended
load = 9,078, 000, bedload = 3,600, 000, total load = 12,678,000 t/y. For
comparison, on page 27 of his report, Dr Andrews suggests a “mean annual
supply of sediment to the head of the delta of about 13.7 million tons of
suspended and bedload sediment .”, which is similar to but somewhat
greater than that estimated b y ICE. In the lower R ío San Juan the best
estimates of mean annual loads are : suspended load = 1,479,000, bedload
= 702,000 and total load = 2,181,000 t/y . For comparison, Dr Andrews
estimates that 1 ,370,000 t/y would pass to the lower R ío San Juan (that
being 10% of the total load of the Río San Juan just upstream of the Delta).
4.78. However, the proportions of the suspended load and bedload in the R ío
San Juan approaching the delta that pass to the lower R ío San Juan and Río
Colorado are not identical to those of the river water, due to non-linearity
in the suspended and bedload rating curves. The percentages of the
suspended and bedloads in the Río San Juan that pass to the lower R ío San
Juan and Colorado are illustrated in Figures 4.14 and 4.15, respectively.
Figure 4.14 Mean annual suspended loads in the R ío Colorado and lower San Juan as
percentages of suspended load in the Río San Juan assuming that (a) 95%, (b) 90% and (c)
85% of the discharge flows to the Río Colorado (from the 2014 ICE Report).
54
198Figure 4. 15 Mean annual b edloads in the Río Colorado and l ower Río San Juan as
percentages of bedload in the Río San Juan assuming that(a) 95%, (b) 90% and (c) 85% of
the discharge flows to the Río Colorado (from the 2014 ICE Report).
4.79. Road-bed e rosion rates recommended in the 2014 UCR Report for
stretches of dirt road with gentle or steep slopes were modified as
appropriate to apply to stretches where the surface of the Road is gravel
and stretches where it exists only as a trail (Table 4.15).
Table 4.15Erosion rates for the Roadsurface(from the 2014 ICE Report).
Road Character Erosion rates for Erosion rates for
-1 -1
gentle slopes (m yr ) steep slopes (m yr )
Gravel 0.0014 b 0.0044 b
a a
Dirt 0.0140 c 0.0440 c
Trail 0.0028 0.0088
aBased on ICE (2013).
bApproximated as 10% of dirt road erosion rate.
cApproximated as 20% of dirt road erosion rate.
4.80. These rates were applied to estimate the volume of sediment eroded from
the surface of the Road annually in each of the major Costa Rican tributary
basins between Mar ker II and Delta Costa Rica, under the ‘worst case’
rainfall scenario. The width of the Road was taken to be 10 m in stretches
where it actually exists and 5 m where it is no more than a trail, based on
observations in the field by Dr Mende. A sediment delivery r atio of 0.6 was
used, based on Gómez et al . (2013) – a figure accepted and adopt ed in the
2014 Kondolf Report. Yields of sediment from erosion of cut and fill slopes
in each tributary basin were derived by multiplying the annual erosion
volumes for cut and slopes listed as being in that tributary basin in the
Inventory of Slopes in the Mende Report by the delivery ratio (again tak en
55
199 as being 0.6). The results are listed in Table 4.17 and graphed in Figure
4.16. As these yields were estimated using upper bound erosion rates,
under a ‘worst case’ rainfall scenario, the figures listed and illustrated are
highly conservative.
Table 4.16Annual yields of Road-derived sediment from the basins of major Costa Rican
tributaries between Marker II and Delta Costa Rica under a ‘worst case’ rainfall scenario
(from the 2014 ICE Report).
‘Worst case’ sediment yields ‘Worst case’
Tributary Basin Road length (m yr )1 sediment yields
(km) -1
Road bed Cut slopes Fill slopes Total (t yr )
Major Costa Ricantributary basins draining directly to theRío San Juan
Infiernito 41.0 855 12,348 19,051 32,253 53,863
San Carlos 11.1 173 253 399 825 1,378
Cureña 29.5 387 1,738 8,966 11,091 18,521
Sarapiquí 4.5 172 49 --------- 221 369
Chirripó 22.8 192 190 107 489 817
Costa Rican areadraining directly to theRío San Juan
Total 108.8 1,778 14,578 28,523 44,880 74,949
Sedim
Figure 4.16 Annual yields of Road-derived sediment from the basins of major Costa Rican
tributaries between Marker II and Delta Costa Rica under a ‘worst case’ rainfall scenario
(from the 2014 ICE Report).
4.81. These yields of Road-derived sediment do not consider erosion from areas
disturbed during construction in 2011. This is because those areas have
subsequently revegetated, either naturally or due to vegetation planting by
56
200 CODEFORSA and CONAVI. Neither do the estimates consider erosion at
failed watercourses. This was criticised in relation to the 2013 estimates in
paragraph 2.119 of Nicaragua’s Reply. To explain why it was decided not to
attempt to estimate erosion at failed crossings in 2014, it is only necessary
to examine a typical example, as illustrated by Dr Kondolf in Figure 24, on
page 36 of his 2014 Report, which shows the point where the Road
intersects a small ditch draining an area of pasture. The width of the ditch
is not specified, but as the Road has an average width of 10 m and the ditch
is clearly much narrower than this, it is perhaps 2 m wide. In the vicinity of
the ditch, the channel of the Río San Juan is about 200 m wide. The River in
this reach conveys an average annual discharge of the order of 500 m 3/s
and an annual sediment load of several millions of tonnes . It follows that
volume of sediment that a ditch that is only 2 m wide could erode from a
failed crossing that extends along that ditch for about 10 m is insufficient
to have any impact on the Río San Juan or the lower Río San Juan that could
be either significant or long -lasting. In any case, of 127 watercourse
crossings surveyed in th e updated inventory of crossings in the 2014
Mende Report, erosion has already been mitigated or is in progress at 40%
and is unnecessary at 36% because these crossings are either stable (19%)
or the Road is just a trail (17%). Mitigation is scheduled at t he remaining
24% of crossingswhere it is needed.
4.82. As in the 2013 ICE Report, the input of suspended sediment from Lake
Nicaragua is taken to be 588,000 t /y.
4.83. Most of the fine sediment carried as suspended load moves quickly
through the channel of the R ío San Juan and on to Del ta Costa Rica .
Conservation of mass therefore dictates that the quantity input to the R ío
San Juan below Lake Nicaragua matches that supplied to the lower Río San
Juan and R ío Colorado at the Delta. The sediment budget for the R ío San
57
201 Juan was therefore balanced by adjusting tributary catchment sediment
yields estimated using the distributed model (as listed in Table 4.9) so that
their sum matches the suspended sediment load of the R ío San Juan
immediately upstream of the Delta (excluding the sediment input from
Lake Nicaragua), as must be the case to satisfy conservation of mass. In
closing the budget, the time series-based sediment yields developed from
measurements at the mouths of the Sarapiquí and San Carlos were used as
controls and differences between the USLE-modeled and time series-based
mean annual sediment loads were redistributed in proportion to sediment
yields in the remaining basins. T he results are listed in Table 4 .18(a),
followed by a summary of the sediment load estimates for the Río San Juan
and contribution from the Road, in Table 4.18(b).
Table 4.17(a)Initial and adjusted suspended sediment yieldsfor Study Area
(from the 2014 ICE Report).
Adjusted
Suspended
Drainage sediment suspended
Basin are2 Yield sediment
(km ) (t yr Yield
(t yr )
Melchora 305 152 000 207 000
Sábalos 571 456 000 622 000
Santa Cruz 418 253 000 345 000
Barlota 219 45 000 62 000
Machado 352 42 000 57 000
Las Banderas 198 29 000 39 000
Frío 1577 666 000 907 000
Pocosol 1224 353 000 481 000
Infiernillo 609 239 000 326 000
San Carlos 2642 2 451 000 2 928 000
Cureña 353 78 000 106 000
Sarapiquí 2770 2 791 000 2 342 000
Chirripó 236 50 000 68 000
Study Area 11474 7 605 000 8 490 000
Lake Nicaragua 29067 --------- 588 000
Note: Confidence is highest in values in bold; therefore, the difference between erosion-based and time
series-based yields was distributed between the remaining catchmentsso that the sum of all sources
matches the suspended load of the Río San Juan (excluding input from L. Nicaragua) = 8 490 000 (t yr
58
202Table 4.17(b) Summary of estimated sediment loads and sediment inputs from the Road
into the RíoSan Juan.
Source Suspended Load Bedload Total Load(t/y)
(t/y) (t/y)
Río San Juan 9,078,000 3,600,000 12,678,000
Río Lower San Juan 1,479,000 702,000 2,181,000
Road (ICE, 2014) 67,454 - 71,202 7,495 - 3,747 74,949
Road (Kondolf 2014) -- -- 177,020 - 250,500
4.84. ICE’s data in Tables 4.17(a) and (b) were used to create a mean annual
suspended sediment budget diagram for the study area of the San Juan
basin (Figure 4.17). The budget is based on 90% of the flow in the R ío San
Juan flowing to the Río Colorado, but changing that to 85% or 95% would
only slightly change the appearance of the diagram (see Figure 4.15).
4.85. A mean annual coarse load sediment budget d iagram is also shown in
Figure 4.17. As the bedload inputs for eleven of the fourteen tributaries are
unknown (bedload input from L ake Nicaragua is probably small) only an
estimated division of bedload at the Delta can be depicted in that diagram.
4.86. In Figure 4.18, the ‘worst case’ annual contribution of suspended sediment
derived from the Road to each reach of the Río San Juan (taken from Table
4.17) i s highlighted in red. In deriving the contributions of suspended
sediment, it was necessary to subtract the coarse fraction from the overall
contribution of Road -derived sediment. In the 2014 ICE Report, the
division of the Road -derived sediment at the Delta was found to be
insensitive to the percentage of coarse sediment that is assumed when this
is varied across a range between 5 and 30%. Hence, the appearance of the
red lines in Figure 4 .18 does not change appreciably depending on the
percentage of Road -derived sediment that is assumed to be coarse-
grained.
59
203 Figure 4.17 Suspended and bed sediment load budgets for the Río San Juan between
Marker II and the Delta (values in t yr) (from the 2014 ICE Report).
Figure 4. 18 Suspended sediment budget for the San Juan -Colorado System with the
contribution of fine-grainedsediment from Route 1856 indicated in red at the lower edge
of the diagram. Additional loads entering the lower San Juan and Colorado are based on
the assumption that 5% of sediment derived from the Road is coarse -grained (values in
parenthesis correspond to a 10% fraction of coarse material) (from the 2014 ICE Report).
60
2044.87. Nevertheless, additions of Road -derived suspended load to the lower R ío
San Juan and R ío Colorado are specified twice in Figure 4.18: first on the
assumption that 5% of sediment derived from the Road that reaches the
Delta is coarse-grained and second ( with values in parenthesis ) on the
assumption that 10% is coarse-grained.
4.88. The contributions of Road -derived sediment to the mean annual
suspended sediment load of the R ío San Juan in each of the sub -reaches
between Marker II and Delta Costa Rica listed in Table 4.17 are not just
insignificant; they are indiscernible.
4.89. Inspection of Figure 4.18 shows that the same is also true of the
contribution of Road -derived, fine- grained sediment to the suspended
sediment load entering the lower Río San Juan.
4.90. Based on the data listed in Table 4.17, above, the ‘worst case’ annual input
to the Río San Juan of Road -derived sediment is 44,880 m 3/y. In the 2013
Thorne Report (see paragraph 8.60, page 85) it was assumed that the bulk
density of Road-derived sediment is about 1.67 t/m , and that 5 to 10% of
that sediment is coarse. Applying these assumptions again here suggests
that of the upper bound mean annual input of coarse sediment to the R ío
San Juan is 2,244 to 4,488 m 3/y, which is equivalent to 3,747 to 7,495 t/y.
No data are available for the bedloads in the sub-reaches between Marker
II and Delta Costa Rica, but ICE’s lower bound estimate of the annual
bedload of the River immediately upstream of the Delta is 799,000 t/y.
ICE’s best and upper bound estimates are much larger (3,600,000 and
7,191,000 t/y) reflecting very high variability and uncertainty in the
equation-based bedload transport calculations necessary to estimate the
bedload in this river.
61
205 4.91. Based on the upper bound assumption for the percentage of Road -derived
sediment that is coarse-grained (10%) and using the lower bound estimate
of bedload in the Río San Juan, the contribution of coarse-grained sediment
from the Road would still be less than 1%. Given the high variability and
uncertainties associated with these estimates, the contrib ution of Road -
derived sediment to the mean annual bed load of the R ío San Juan is
therefore not just insignificant, it is also indiscernible.
4.92. On page 28 of the Andrews Report, Dr Andrews uses Dr Kondolf ’s
estimates that the quantity of sediment delivered to the River annually due
3
to erosion along the Road is between 106,000 and 150,000 m /y,
depending on whether access roads are included.
4.93. Converting these estimates to t/y would suggest that, according to Dr
Kondolf’s estimates (which I do not accept), the contribution of Road -
derived sediment to the Río San Juan is between 177,020 and 250,500 t/y,
which would constitute just 1 to 2% of the mean annual total load (that is
suspended load plus bedload) of the R ío San Juan estimated by ICE, which
is 12,678,000 t/y.
4.94. Using the UCR upper bound estimates of erosion and the Mende-ICE ‘worst
case’ figure for delivery of Road-derived sediment to the River , which I
believe to be more reliable, the contribution of Road -derived sediment to
the load in the Río San Juan (44,880 m 3/y or 74,949 t/y) is an indiscernible
0.6% of the total load in that river, which is estimated by ICE to be
12,678,000 t/y.
4.95. Uncertainty in estimated bedloads in the Río San Juan is very high. To test
whether uncertainties associated with the bedload are significant, the
calculation above may be repeated with the bedload excluded. According
to Dr Kondolf’s estimates (which I do not accept) the contribution of Road -
62
206 derived sediment to the Río San Juan is between 177,020 and 250,500 t/y,
which would constitute just 2 to 3% of the mean annual suspended load of
the Río San Juan (9,078,000 t/y) estimated by ICE.
4.96. On page 27 of his report, Dr Andrews suggests a “mean annual supply of
sediment to the head of the delta of about 13.7 million tons of suspended and
bedload sediment ” which is similar to but somewhat greater than that
estimated by ICE. If Dr Andrews’ estimate of the mean annual total
sediment load of the R ío San Juan is accepted, and Dr Kondolf’s estimates
(which I do not accept) that the contribution of Road -derived sediment to
the Río San Juan is between 177,020 and 250,500 t/y are applied, the
contribution of the Road based on the data reported to the Court by
Nicaragua’s own experts is still just 1 to 2%.
4.97. Dr Andrews also calculates that 1 ,270 to 2, 700 m 3/y of coarse sediment
from Route 1856 would enter the lower R ío San Juan (he assum es that
10% of Road-derived sediment is carried into the Lower Río San Juan and
that 12 to 18 % of the sediment load carried by the Río San Juan is
relatively coarse).
4.98. While I do not accept these estimates, it is instructive to carry Dr Andrews’
bedload analysis through to its logical conclusion. On page 27 of the
Andrews Report, he writes that, “ The estimated mean annual transport of
bed-material at the beginning of the Lower Río San Juan is approximately
120,000 tons/year or 75,000 m 3/year of relatively coarse sediment.”. It
follows that, even using Dr Kondolf’s estimates and Dr Andrew’s analysis ,
which I do not accept, the input of coarse sediment from the Road
constitutes only 2% to 4% of the coarse sediment load expected to enter
the lower Río San Juan in an average year.
63
207 4.99. Commenting on the uncertainties associated with bedload measurements
and calculations, on pages 23/24 of his report, Dr Andrews correctly
observes that, “ Calculated bedload transport rates are particularly sensitive
to errors or uncertainty in the fluid forces acting on the river bed at a given
discharge. Fluid forces depend on hydraulic characteristics, such as flow
depth, velocity, the presence of bedforms, and river slope. Relatively small
errors in the estimation of fluid forces, e.g. +/- ten percent, will result in
much larger errors in the calculated bedload transport rate, which varies
rapidly as a function of the fluid forces. The effective exponent of the bedload
transport rate versus fluid forces decreases from about 14 as river bed
sediment beg ins to move and approaches a value of 1.5 at very high
transport rates. Thus, a +/- 10 percent error in the calculation of fluid forces
will result in errors of a few tens up to a few hundreds of percent in the
calculated bedload transport rate.” Bearing this in mind, it is clear that a
difference of 2% to 4% in the annual bedload would not only be
insignificant but scientifically undetectable, ruling out even the possibility
of demonstrating any causal relationship between construction of the
Road and any change in the quantity of coarse bedload entering the lower
Río San Juan.
4.100. Using the UCR-Mende-ICE estimates of erosion and sediment delivery from
the Road set out in Table 4.18(b) , which I believe to be more reliable, and
ICE’s estimate that 20 % of the bedload in the R ío San Juan passes to the
lower Río San Juan (see Figure 4.5, above and paragraphs 5.23 and 5.24,
below), the ‘worst case’ estimate of the contribution of Road -derived,
coarse-grained sediment to bedload in the lower Río San Juan is 450 to 900
m 3/y or 750 to 1,500 t/y. This would constitute just 0.1 to 0.2% of the
coarse load, which is estimated by ICE to be 702,000 t/y.
64
208 D. Potential impacts of a hurricane or tropical storm
4.101. On page 51 of hi s 2014 Report, Dr Kondolf expands on expert views
advanced in his previous reports that erosion along the Road to date has
been relatively low compared to that which will occur, “ during intense
rains that will inevitably accompany tropical storms and hurricanes in the
region”. The consequences and inevitability of catastrophic erosion of the
Road during a Hurricane or Tropical Storm are referred to on no less than
six separate occasions ( pages 1, 14, 28, 35, 51, as well as at length in
Section 12, (pages 71 and 72), and they also feature in Sections 4.2.2 (page
8) and 7.2.3.2 (pages 43 and 44) of the Report by Golder Associates (Annex
6 in Nicaragua’s Reply of August 2014)(the Golder Report).
4.102. There is no doubt that heavy rainfall is associated with tropical cyclones
and that widespread flooding, landsliding, destruction of property and,
regrettably, fatalities are likely within affected areas. However, Dr Kondolf
states that it is inevitable that the Road will be catastrophically eroded
during a Hurricane or Tropical Storm and that delivery of sediment eroded
from the Road during such an event would lead to unprecedented
sediment loads and concentrations in the R ío San Juan. It is in this context
that Dr Kondolf’s account of the certainty of extreme erosion in the area
around the Road during future Hurricanes and Tropic Storm s merits
careful examination.
4.103. In this regard, the Director General of the Costa Rican National
Meteorological Institute, Professor Juan Carlos Fallas Sojo (who is also
Professor of Physics and Meteorology, at the University of Costa Rica)
reviewed the 2014 Kondolf report insofar as it pertains to Hurricanes and
Tropical Storms. He presented his comments in a report titled, ‘ Comments
on the report by Dr Kondolf (as it pertains to hurricanes and tropical
65
209 storms) in: Section 12 - Risks of Larger Contributions from Rte. 1856
[Annex 1, pages 71 -74]’ (the Fallas Report). As I am not an expert in
tropical meteorology, the explanations presented in this sub -section draw
extensively on the Fallas Report.
4.104. Dr Kondolf is apparently unaware that there are s eparate naming
conventions for H urricanes in the Atlantic and Pacific Basins . This is
evident from his statement on page 71 of the 2014 Kondo lf Report that,
“The eyes of Hurricanes Irene and Olivia in 1971 both tracked just to the
north of the Río San Juan”. As explained in the Fallas Report, Hurricanes
Irene and Olivia were the same event. The Hurricane was called Irene as it
passed through the Caribbean and entered Nicaragua. When the hurricane
arrived in the Pacific Basin, it was renamed Olivia.
4.105. On page 71 of his 2014 report, Dr Kondolf challenges my statement (and
presumably NASA’s records, which provide the basis for i t) that a
Hurricane or T ropical storm has not struck Costa Rica during the period
since records began. To suppor t this he refers to, “ the tropical storm that
occurred 6-11 May 2004”. As Professor Fallas points out in his review, “ The
weather system that generated rainfall over the territory of Costa Rica was
not a tropical storm , it was a much smaller disturbance i n its intensity and
persistence, called a tropical wave”.
4.106. This is significant in the context of the wider point Dr Kondolf seeks to
make because the fact that the event he selected to support his expert view
is correctly named by NASA as a “tropical easterly wave” in the caption to
Figure 32 on the very next page ( page 72) of the Kondolf Report suggests
that Dr Kondolf does not perceive any difference between a Tropical Storm
and a Tropical Wave. This is incorrect. These types of weather phenomena
are graded by their characteristics and intensity (from lowest to highest)
66
210 as a T ropical Wave, Tropical Depression, Tropical Storm or H urricane.
Tropical Waves occur frequently in Costa Rica and, as I stated in my 2013
Report, the rivers of the region are well adapted to assimilate the
associated rainfall intensities, durations and distributions.
4.107. A lay person would envisage that the rainfall intensities and durations
associated with Hurricanes and Tropical S torms would exceed those
associated with a Tropical W ave, and they would be correct . However, as
the Fallas Report explains, there is one big difference between a Tropical
Wave and the other weather phenomena list ed above: the circulation of
the air and resulting distribution of rainfall. A Tropical Wave is a cluster of
thunderstorms (as is evident on inspection of Figure 32 in the 2014
Kondolf Report) with very little, if any, organized wind circulation. In
contrast, Tropical Depressions, Tropical Storms and H urricanes are
cyclonic: that is , they feature a well -organised pattern of air circulation
with very strong winds that rotate counter -clockwise around a single low
pressure centre (Figure 4.19).
4.108. This strong, counter -clockwise circulation combines with Costa Rica’s
prominent mountain system to promote the orographic effect , which
concentrates rainfall on the windward side of the mountains and creates a
‘rain shadow’ effect on the leeward side. This is evident from the
distribution of rainfall generated by Hurricane Mitch (Figure 4 .20, which is
reproduced from the Fallas Report).
4.109. The fact is that the rainfall associated with a H urricane or Tropical Storm
would be much greater in catchments draining to the Pacific than in
catchments draining to the Caribbean, such as that of the Río San Juan.
67
211 Figure 4.19 Distribution of wind and rain in Costa Rica due to a tropical cyclone near the
Caribbean coast of Nicaragua (from the Fallas Report).
Figure 4.20 Rainfall distribution in Costa Rica recorded during Hurricane Mitch (from the
Fallas Report).
4.110. When Dr Kondolf refers to seven fatalities caused by Hurricane Mitch in
Costa Rica on page 72 of the 2014 Kondolf Report he is correct. But these
68
212 fatalities did not occur in or near to the area around the Road. These
deaths occurred in the Pacific drainage basin, on the other side of the
continental divide. This is explained by Mitch’s circulation (Figure 4.19)
and rainfall distribution (Figure 4 .20), which were those of a Tropical
Cyclone. The deaths did not occur in the San Juan basin or even in the
Caribbean drainage basin.
4.111. Based on Professor Fallas’ expert meteorological review , and my
understanding of rainfall -driven erosion, it is my conclusion that it is
highly unlikely that the Road will be catastrophically eroded in the event
that a future Hurricane or Tropical Storm affects Costa Rica and Kondolf is
wrong when he suggests this on pages 1, 14, 28, 35, 51, 71 and 72 of his
2014 Report.
4.112. In my opinion as a geomorphologist , the risk of rapid erosion due to
intense rainfall in the area around the Road is probably greater during the
localised thunderstorms associated with a Tropical W ave than would be
the case during a Tropical Cyclone. However, the frequency of localised
downpours is high and their impacts limited because, to restate my
position, “the hydrology, sediment dynamics, morphology and environment
of the River are fully adjusted to the effects of frequent and heavy
rainstorms” (2013 ThorneReport, paragraph 6.20).
E. The Natural Sediment Load of the San Juan River and how this may
differ from that immediately prior to construction of the Road
4.113. In the Andrews Report, Dr Andrew s states that, “ proper analysis of the
impacts that the construction of Route 1856 has had and will have in the
coming decades on the supply, transport and deposition of sediment to the
Río San Juan must involve a comparison. ” The preceding sections provide
the basis for just such a comparison: that of sediment budget of the Río San
69
213 Juan to a ‘worst case’ estimate of the annual volume of sediment input due
to construction of the Road. That comparison demonstrates that the
quantity of sediment eroded and delivered to the R ío San Juan from the
Road (estimated by ICE to be 74,949 t/y in the 2014 Report) is less than
1% of the mean annual suspended load (9,078,000 t/y,excluding bedload).
This is indiscernible given the wide confidence band on the mean annual
suspended sediment load, which varies between 8,286,000 to 9,997,000
t/y due to uncertainty in the division of flow at the Delta alone (without
considering inter-annual variability and uncertainty in the sediment rating
curves).
4.114. The Road’s contribution would still be indiscernible using the much higher
range of values for sediment delivery fr om the Road and its access roads
proposed in the 2014 Kondolf Report (177,020 - 250,500 t/y), which I do
not accept. C omparison of Dr Kondolf’s estimated range with the best
estimate of the mean annual suspended load of the River ( 9,078,000 t/y)
indicates it to constitute 2 to 3% of the mean annual suspended load of the
Río San Juan. This is still indiscernible given the inter -annual variability of
the suspended sediment load carried by the R ío San Juan and the
uncertainties inherent to its estimation.
4.115. However, Dr Andrews avoids making that comparison by suggesting on
page 6 of his Report that, “ The question is whether 61,000 to 240,000 tons
per year is a relatively small or large amount of sediment in comparison to
the natural sediment yield. ” and pointing out that, “ The answer to this
question depends largely upon the basin -wide sediment yield that is
determined to be ‘natural.’”
4.116. Before presenting my technical response, two over-arching issues arising
from Dr Andrews’ framing of the comparison should be highlighted. First,
70
214 based on my understanding of the case before the Court, the comparison
Dr Andrews suggests of, “whether 61,000 to 240,000 tons per year is a
relatively small or large amount of sediment in comparison to the natural
sediment yield” is not ‘the question’. The scientific question, as I understand
it, is whether construction of a Road in Costa Rica has caused harm to the
environment or ecology R ío San Juan de Nicaragua. This requires
consideration of whether construction of the Road has had any significant
impact on sediment processes, morphology, aquatic life, or navigation as
they existed i n and along the R ío San Juan prior to construction of the
Road. So far as I understand it (and this is obviously a matter for the
Court), it has nothing to do with whether the sediment load of the Río San
Juan is ‘natural’.
4.117. Second, Dr Andrews ’ belief that the current load carried by the R ío San
Juan is unnatural puts him at odds with other sediment experts with good
working knowledge s of the River. For example, o n page 10 of their
independent report entitled, ‘Morphological Stability of the San Juan River
Delta, Nicaragua/Costa Rica’ (submitted to the Court by Nicaragua in 2011
as Document 18 in its application for provisional measures ), Professors
van Rhee and de Vriend conclude that dredging is not likely to cause any
measurable environmental harm , “given the pre -existing high natural
turbidity of the river.”
4.118. On page 11 of a further report entitled ‘The Influence of Dredging on the
Discharge and Environment of the San Juan River ’ (submitted to the Court
in 2012 as Appendix 2 to Nicaragua’s Counter Memorial in Certain
Activities), Professors van Rhee and d e Vriend restate their expert view,
writing that, “as explained in VRDV 2011 [the earlier report referred to
above], the San Juan River is both naturally turbid and relatively stable”.
71
215 4.119. Similarly, while Dr Andrews rejects statements I make in my 2013 Report
concerning the natural load and tur bidity of the River, other experts have
explicitly accepted my opinion as it was expressed in my report submitted
by Costa Rica as part of its Memorial in Certain Activities. For example, in a
passage on page 10 of their 2012 Report that explains why dredging would
not pose environmental problems in the lower Río San Juan, Professors
van Rhee and de Vriend note that, “Thorne actually confirms this conclusion,
stating ‘Natural high sediment and nutrient concentrations in the river are
likely to limit impacts on turbidity and water quality that are customary
associated with dredging (Thorne, p. vii)’”.
4.120. Dr Andrews suggests that the environmental impacts of activities that
affect sediments in the river should be judged against the ‘natural’ load of
the Río San Juan and that the ‘natural’ load is only 170,000 to 420,000 tons
per year (see page 10 of the Andrews Report) . Incidentally, if this position
is tenable, the environmental and ecological impacts of increased turbidity
due to re-suspension of sediment associated with Nicar agua’s substantial
dredging operations ( unavoidable according to Nicar agua’s dredging
experts) would be even more unacceptable.
4.121. I take the opportunity provided by Dr Andrews’ posing of this question to
explain why I believe his estimate of 170,000 to 420,000 tons per year as
the natural sediment load of the R ío San Juan (page 10 of the Andrews
Report) to be unrealistically low and why, in paragraph 6.45, of the 2013
Thorne Report, I characterise the Río San Juan as having “naturally high
concentrations of suspended sediment”; a characterisation that is effectively
endorsed by other experts (including Professors van Rhee and de Vriend).
4.122. On page 8 of his report, Dr Andrews states that, “Without the benefit of
useful gage records from the Río San Juan Basin to analyze the Basin’s
72
216 natural sediment yields, the best recourse is to search for data from
comparable forested tropical river basins”. This is true, but it is crucial that,
in Dr Andrews’ words, the data should come from “ comparable forested
tropical river basins”: the key word here being‘comparable’.
4.123. Based on the sources that Dr Andrews selected as being “forested tropical
river basins with a wide range of precipitation, geology and topographical
relief, includi ng basins that, like the Río San Juan basin, contain areas of
volcanic soil, steep slopes, and receive significant rainfall” (page 8), he finds
sediment yields reported in the literature (listed in his Table 1, also on
page 8) to show that, “ sediment yield s from tropical river basins with
undisturbed primary forests vary from 1 to 120 tons/km 2–year” and on that
basis he concludes later on the same page that, “sediment yields in the Río
San Juan Basin prior to appreciable forest clearing and landscape
disturbance were likely to fall between 20 to 50 tons/km per year”.
4.124. However, according to Dunne (1979, page 292) - one of the sources c ited
in Dr Andrews’ Table 1 , "The range of yields from the small Kenyan sample
-2 -1
(~20 - 30 t km yr ) therefore, seems to be representative of undi sturbed,
humid catchments under tropical forest in tectonically stable areas" [my
emphasis]. As pointed out in my 2013 Report (Sections 6.3 to 6.5) and
expand on below, the Costa Rican tributary basins of the Río San Juan are
certainly not tectonically stable.
4.125. Dr Allan Astorga has studied the geology of the basin of the Río San Juan in
depth. In his 2014 Report (the Astorga Report), he explains why sediment
loads in the Río San Juan are both naturally high and highly variable, and
demonstrates that this has been the case for at least the last 10 million
years.
73
217 4.126. The Astorga Report describes the basin’s complex geological history which
is related to the tectonic evolution of the lithospheric blocks making up the
Caribbean Tectonic Plate and the basin’s location in the Central American
back-arc area – which is characterized by multiple active faults ( Figure
4.21) and live volcanoes, most of which form a chain extending along the
basin’s southern watershed in Costa Rica (Figure 4.22).
4.127. Figure 4.23 presents a synopsis of earthquakes with magnitudes greater
than 6 and volcanic eruptions known to have generated mud flows in
tributaries draining from the Costa Rican side of the basin dur ing the last
three centuries.
Figure 4.21 Tectonic map of South Central America, indicating the main tectonic and
neotectonic elements affecting the San Juan basin (indicated by the blue line). The lower
basin lies in an area of the central America back-arc(from the Astorga Report).
74
218Figure 4.22 Volcanoes with heights of between 2,000 and 3,000 metres that periodically
contribute extraordinary amounts of sediment to the San Ju an drainage system (from the
Astorga Report).
Figure 4.23 Historical record of earthquakes and volcanic eruptions inputting
extraordinary amounts of sediment to the San Juan basin from the Costa Rican part of the
basin (from the Astorga Report).
4.128. This record illustrates that there have been multiple earthquakes and
volcanic eruptions per century in the basin, each capable of generating an
extraordinary input of sediment to the drainage system.
4.129. For example, as recently as January 8, 2009, the Cinchona Earthquake
(magnitude 6.1 Mw), occurred in the mountainous headwaters of the R ío
Sarapiquí. This single natural event generated thousands of landslides,
82% of which happened in areas covered by forest (Ruiz et al., 2011) . In
75
219 the area immediately surrounding the epicentre, 349 landslides disturb ed
21.7 km of formerly vegetated land ( Alvarado 2010 and Figure 4.24) and
3
delivered 2.5 to 3.5 million m of sediment (equivalent to 4 to 6 million
tonnes) to the drainage system s affected – 95% of which enter ed Costa
Rican tributaries to the Río San Juan . For comparison, the overall area
disturbed by construction of the Road along its full 108 km length
alongside the River is just 3.5 km (an area confirmed by D r Kondolf on
page 62 of his 2014 Report) and, even by Dr Kondolf’s estimate, which I do
not accept, the quantity of sediment delivered to the River annually is a t
most a quarter of a million tonnes.
Figure 4. 24 Map of landslides triggered by the 2009 Cinchona earthquake (from
Alvarado, 2010).
4.130. Dr Andrews also sources data from a 1967 paper by Professor Ian Douglas,
but does not mention Douglas’ follow–up article titled, ‘ The impact of
76
220 land-use changes on sediment yields in humid tropical Southeast Asia ”
(Douglas, 1996) in which he builds on his 1967 paper . In t his follow-up
article Professor Douglas notes that, "Volcanic catchments in Java have high
erosion rates even under forest, as the Cilutung catchment (the highest point
in column C in Fig. 1) illustrates, at 2250 t km year (Van Dijk & Vogelzang,
1948)". As well as the volcanic soils that are mentioned by Dr Andrews, the
area of Costa Rica draining to the R ío San Juan features nearly a dozen
active volcanoes, as illustrated in Figure 4.22, above.
4.131. In summary, i t is entirely inappropriate to deduce the ‘natural’ sediment
yields of tributary basins of the R ío San Juan in Costa Rica, known to be
tectonically and volcanically active, based on rates measured in
tectonically stable, non -volcanic basins. This is especially so for the San
Carlos and Sarapiqu í sub-basins, which have mountain headwaters with
elevations exceeding 3,000 m. To explain why , consider that, as Professor
Douglas states in his 1996 paper , "The highest sediment yi elds occur in
tectonically active areas, where earth tremors trigger frequent mass
movements which supply large volumes of sediment to rivers. The lowest
sediment yields are on old land surfaces of low relief and deep weathering
profiles. The contrast quan titatively is the difference between yields of the
order of 10 000 t km -2year in mountains of New Guinea, Taiwan and the
South Island of New Zealand (Pickup et al., 1981; Shimen Reservoir
Authority, 1975; Griffiths, 1979) and yields of around 100 t km year in-1
Africa (Milliman & Meade, 1983)."
4.132. Further insights into the ‘natural’ sediment yield of the basin can be gained
from closer examination of the maps of sub -catchments, terrain, rainfall
and the USLE factors land cover C, rainfall, R, and slope length LS (Figures
4.6 and 4.7, above) and the results of ICE’s distributed soil -erosion model
(Table 4.19 (data abstracted from Table 4.9, above) and Figure 4.25).
77
221 Table 4.18Specific sediment yields in Nicaraguan and Costa Rican Basins
(from the 2014 ICE Report)
Drainage Specific Total Sediment
Basin Area Sediment Yield Yield
(km ) (t km yr )1 (t yr )
Tributary basins in Nicaragua
Las Banderas* 198 145 29 000
Machado* 352 119 42 000
Barlota* 219 207 45 000
Santa Cruz 418 606 253 000
Melchora 305 497 152 000
Sábalos 571 799 456 000
Nicaragua
Basins 2,063 473 977,000
Tributary Basins in Costa Rica
Chirripó 236 211 50 000
Cureña 353 221 78 000
Pocosol 1224 288 353 000
Infiernillo 609 393 239 000
Frío 1577 422 666 000
5 Costa Rica
Basins 3,999 346 1,386,000
San Carlos 2642 928 2 451 000
Sarapiquí 2770 1007 2 791 000
San Carlos &
Sarapiquí 5,412 969 5,242,000
7 CR Basins 9,411 704 6,628,000
Study area 11,474 663 7,605,000
*the land cover map (Figure 5.25(f) below) illustrates clearly that vegetation in
these three basins is predominantly undisturbed forest (Code =FORE).
4.133. The data reveal that despite the fact their primary forest cover is
undisturbed, specific sediment yields in the Las Banderas (145 t km yr ),1
Machado (119 t km yr ) and Barlota (207 t km yr 1) tributarybasins do
not come close to supporting Dr Andrews’ proposition that, “sediment
yields in the Río San Juan Basin prior to appreciable forest clearing and
landscape disturbance were likely to fall between 20 to 50 tons/km 2 per
year”.
78
222 (a)
(b) (c)
(d) (e)
(f) (g)
(h)
Figure 4.25 (a) tributary basins (b) digital elevation model (c) slope length factor LS (d)
mean annual precipitation (e) rainfall erosivity factor, R (f) land cover (g) land cover
factor, C and (h) calibrated specific sediment yields (E) in the study area.
79
223 4.134. With res pect to the impacts of development in the San Juan basin, the
-2 -1
average specific yield of basins in Nicaragua (473 t km yr ) is comparable
to, but somewhat higher than, that averaged for five of the seven tributary
basins in Costa Rica (346 t km yr ), while excluding the San Carlos and
Sarapiquí basins.
4.135. The muc h higher yields in the San Carlos and Sarapiquí basins are
explained by inspection of Figure 4 .25, in which maps (a) and (b) show
that high mountain terrains are found mostly in the headwaters of the San
Carlos and Sarapiquí basins. This explains why the slope length and
steepness factor, LS , shown in map (c) is much higher there than for the
remainder of the study area. Map (d) re veals that the San C arlos and
Sarapiquí basins receive much higher rainfall than any other basins in the
study area, explaining why the rainfall erosivity factor, R, (map (e)) is also
very high there. Map (f) shows that vegetation in the head waters of the
San Carlos and Sarapiquí basins is forest, trees and shrubs (code = FOSM),
such that the land cover factor, C, is below average for the study area. Map
(h) illustrates that the high values of LS and R outweigh the ameliorating
effect of land cover C in the USLE, to produce very high specific sediment
yields for the headwater basins of the San Carlos and Sarapiquí rivers.
Local sediment yields in the headwaters (which are mostly undisturbed
forest and shrubs within National Parks), are commensurate with , but
lower than those in Professor Douglas’ definitive statement quoted in
paragraph4.130, above.
4.136. In conclusion, I believe that Dr Andrews’ estimate of 170,000 to 420,000
tons per year as the ‘natural’ sediment load of the R ío San Juan is
unsupported by the relevant literature, inconsistent with the geology of
the San Juan Basin, and far too low. I reiterate here the statement I made in
80
224 paragraph 6.45 of my 2013 Report, that the Río San Juan has “naturally
high concentrations of suspended sediment.”
4.137. I also reject Dr Andrews’ conclusion that, “the present sediment load of the
Río San Juan is unnaturally elevated due primarily to deforestation and
associated land disturbance in the Costa Rican parts of the basin .” [my
emphasis]. While I agree that deforestation and agricultural development
will have increased specific sediment yields in the areas cleared ,
deforestation is certainly not confined to “ the Costa Rica n parts of the
basin.” The land cover map in Figure 4.25(f) indicates that forest has been
widely disturbed in three of the six Nicaraguan sub -basins draining to the
Río San Juan , and modelling results listed in Table 4.19 suggest that
specific sediment yields in the Santa Cruz, Melchora and Sábalos basins
may be higher than those in five of the seven sub-basins in Costa Rica.
4.138. Evidence that heavy suspended loads and high levels of turbidity occur in
Nicaraguan as well as Costa Rican tributaries is not difficult to find. For
example, Figure 4.26 shows highly turbid water entering th e Río San Juan
from the Río Santa Cruz and Río Sábalos on 23 December 2012. The plume
of highly turbid water from the Río Sábalos is especially prominent and
can be traced downstream in the Río San Juan for some distance.
81
225 (a) (b)
Figure 4.26 Turbid water draining to the Río San Juan from Nicaraguan tributaries on 23
December 2012 (a) Río Santa Cruz (b) Río Sábalos.
4.139. Sediment yields from the San Carlos and Sarapiquí basins are much higher
than those in the other eleven sub -basins in the study area, but this is not
primarily due to land use changes ( which have occurred not only in the
San Carlos and Sarapiquí basins, but in 10 of the thirteen sub -basins). It
occurs be cause these rivers have steep, mountainous , headwater
catchments that are tectonically-active and which feature active volcanoes
that periodically supply extraordinary amounts of sediment to their fluvial
systems and, hence, to the Río San Juan.
4.140. Dr Kondolf refers to sediment being regarded as a pollutant on page 63 of
his 2014 Report. I agree that when sediment loads are increased as a result
of anthropogenic activities , sediment is treated as a polluta nt by
environmental regulators. However, s ediment is only regarded as a
pollutant if its concentration or load is artificially increased above that
expected given the natural of the watercourse and its catchment context.
Hence, to be a pollutant, sediment concentrations and loads must be
elevated compared to the natural sediment concentrations and loads in the
river. Rivers draining tectonically active basins, and especially those with
82
226 live volcanoes, are known to carry very h eavy and highly variable
sediment loads that persis t over geological peri ods (i.e. millennia to
millions of years), as pointed out in Douglas (1996). Sediment
concentrations in the Río San Juan are high and highly variable because the
basin experiences extraordinary sediment yields associated with
earthquakes and volcanic eruptions that are a natural consequence of its
geology (for reasons set out in the Astorga Report). The geology of the San
Juan Basin dictates that sediment is not and cannot be regarded as a
pollutant in the Río San Juan.
4.141. In attempting to challenge my assertion that the sediment load of the R ío
San Juan is naturally high , Dr Kondolf relies heavily on the evidence
provided by Dr. Andrews. For example, in paragraph 3, on page 68 he
states that, “ Dr. Andrews presents the evidence and literature regarding the
land use that has resulted in such an unnaturally elevated load in the Río San
Juan.” While uncertainties concerning sediment yields and loads in the San
Juan Basin remain high, Dr Andrews’ evidence does not appear plausible
and this casts doubts on expert views in the 2014 Kondolf Report that rely
on that evidence.
4.142. In Section V.C of the Andrews Report, Dr Andrews contends that the
measurements of suspended load in the Río San Juan at La Trinidad made
between 1974 and 1976 are too few in number to draw any meaningful
conclusions. It is indeed unfortunate that only these limited records are
available, but the fact remains that in 1974 -76, 12 measurements were
made using the best field methods then available, as part of a programme
performed by joi ntly by Costa Rica and Nicaragua. It is , therefore,
undeniable that these measurements provide at least indications of
suspended sediment concentrations and load s transported by the Río San
Juan during that period.
83
227 4.143. The annual suspended sediment load in th e San Juan, at least during the
measurement period in 1974- 76, can be estimated by combining the
measurements of suspended sediment concentrations with the volumes of
water flowing through the River during the relevant period (referred to as
the average annua l hydrograph). On this basis, the average annual
suspended sediment load for 1974-1976 was on the order of 8 million t/y.
This is falls well within the confidence band for the annual suspended
loads established by ICE in their 2014 Report (see Table 4.1 4, above),
which accounts for uncertainty due to inter -annual variability and the
scatter of measured data around the suspended sediment rating curve.
4.144. The measurements of suspended sediment concentration made in the
1974-76 joint programme are few in number, but they provide the only
available indication of suspended concentrations and annual load
transported prior to construction of the Road.
4.145. In the 2013 ICE Report, these measurements were compared graphically
to those made by ICE at Station 11 -04 on the Río Colorado immediately
downstream of the Delta, in 2010 -13. In Figure 3 , on page 7 of the 2013
ICE Report, Costa Rica’s experts used simple, linear regression to compare
suspended sediment concentration records for 1974-76 and 2010-13, with
the regression curves constrained to pass through the origin (that is point
(0, 0) on the graph . On page 30 of the Andrews Report, Dr Andrews
suggests removing this constraint because, as he states o n page 31, this is
the “statistically proper” thing to do . In Figure 3 on the same page, Dr
Andrews presents alternative linear regression lines that are not
constrained to pass through the origin. However, fitting regression lines
that do not pass through the origin to data that express the suspended
sediment concentratio n as a function of the discharge is physically
nonsensical. This is the case because (as Dr Andrews’ Figure 3 shows) the
84
228 regression relationships that result indicate that in 1974-76 the suspended
sediment concentration in the Río San Juan would be zero a t low
discharges, while in 2010 -13 the river would carry a small but finite
suspended sediment load even if it there were no flow whatsoever. These
conditions are both physically implausible, which over-rides Dr Andrews’
claim that not constraining the regression lines to pass through the origin
is the “statistically proper” thing to do.
4.146. On page 16 of his Report, Dr Andrews states that, “a couple of years of river
flow and a few tens of suspended sediment samples are insufficient and
cannot be relied upon”. Yet on page 34 he applies logarithmic
transformations to the 12 measurements at La Trinidad made in 1974 -76
and the 31 made at Delta Costa Rica in 2010-13, and then fits power curves
to these records which, in his Figure 4, he extrapo lates over two log cycles
(that is suspended sediment concentrations ranging from 10 to 1000 mg/l
3
and discharges ranging from 10 to 1000 m /s). This is inappropriate and
Figure 4 in the Andrews Report does not represent an improvement over
Figure 3 in the 2013 ICE Report. Logarithmic transformation of suspended
sediment concentration and discharge data is appropriate when
constructing sediment rating curves, but it is not helpful when comparing a
few measurements made in 1974-76 to a few moremade in 2010-13.
4.147. In my opinion, the graph shown in Figure 3 of the 2013 ICE Report
(reproduced as Figure 26 on page 63 of the 2013 Thorne Report) provides
a suitably simple platform for comparison of the few available pre- and
post-Road measurements of suspended sediment concentrations in the Río
San Juan – Colorado system. As I stated in paragraph 8.5 the 2013 Thorne
Report, I conclude that, “If additional sediment from the Road had caused an
increase in the rate of sediment transport in the Río San Juan, this would
reflect in Figure 26 through increases in the SSCs measured since 2010 and a
85
229 corresponding upward shift in the 2010 -2013 suspended sediment rating
curve compared to that for 1974 -1976. It is clear from Figure 26 that this is
not the case.”.
86
230 5. Has Construction of Route 1856 had any significant impacts on
channel morphology in the Río San Juan?
A. Reach scale impacts
5.1. In my 2013 Report I applied the Montgomery -Buffington (1997)
classification system to the R ío San Juan- Colorado fluvial system
downstream of Lake Nicaragua (Figure 5.1).
Locations of bedrock outcrops with rapids
Transport Response
Figure 5.1 Designation of reaches of the R ío San Juan according to the Montgomery -
Buffington classification (from the 2013 Thorne Report).
5.2. On the basis of this assessment, I concluded that the addition of sediment
from the Road would have no significant impacts on the morphology of the
Río San Juan with in the first (Lake Nicaragua to Río Pocosol) and second
(Río Pocosol to Boca San Carlos) geomorphic reaches of the river because
these are bedrock-controlled, ‘t ransport’ reaches with ample sediment
transport capacity to carry any additional sediment input to them . This
was not challenged by Dr Kondolf in his 2014 Report.
5.3. In my 2013 report, I further concluded that the addition of sediment from
the Road would have no significant impacts on the morphology of the R ío
San Juan within the third and fourth reaches (Boca San Carlos to Boca
Sarapiquí, and Boca Sarapiquí to the Delta) because, although these are
alluvial, ‘response’ reaches, they receive volumes of sediment input from
the San Carlos and Sarapiquí basins that dwarf any additional supply from
the Road . Although the Andrews Report proposed that the very high
sediment yields of the San Carlos and Sarapiquí basins are unnatural (a
87
231 proposal that is unfounded, for reasons set out in s ection 4E, above), this
was the only part of my explanation of sediment processes in the reaches
between Boca San Carlos and the Delta that was challenged in the technical
annexes that accompanied Nicaragua’s Reply in the ‘ Construction of a
Road’ case.
5.4. As the classification of the geomorphic status of the R ío San Juan between
Lake Nicaragua and the Delta, together with my explanation of why that
status negates the possibility of reach -scale morphological impacts due to
construction of the Road presented in the 2013 Thorne Report, has not
been challenged, I refer readers to that report.
5.5. While he makes no reference to reach -scale impacts upstream of the Delta,
Dr Kondolf continues to contend that construction of the Road has had
local morphological impacts by building or adding to sediment deltas at
the mouths of at least eight small tributary streams draining from the
Costa Rica between Marker II and Boca San Carlos (Appendix F in the 2014
Kondolf Report). In Section 5B, below, I focus on the degree to which these
deltas and their morphological impacts can be attributed solely to
construction of the Road. Aquatic organisms living on and between gravel
particles forming these deltas were sampled by Nicaragua’s experts . The
results were presented in the Rios Report and discussed in the 2014
Kondolf Report. I address the ecology ofthe deltas in Section 6A, below.
5.6. The fifth reach is the lower Río San Juan below the Delta. This is a
Response reach (Figure 5.1 ). I explained in my 2013 report that the
regional, neotectonic uplift of the Chortis Block (which lies to the north of
the Santa Elena - Hess Fault) dictates that the discharge of water and the
transport capacity of the lower Río San Juan naturally decrease gradually
through time, to drive a long -term, depositional trend ( as explained in
88
232 Section II.2.1 of the 2011 Thorne Report that accompanies Costa Rica’s
memorial in the Certain Activities case). This explanation has been
accepted by other experts. For example, Professors van Rhee and de
Vriend state in the 2012 Report (Appendix 2 to Nicaragua’s Counter
Memorial in Certain Activities) that, “As Thorne correctly notes, river
discharges to the Lower San Juan River will gradually decrease without
dredging due to the geological trends in the area (Thorne, p. II -10)”.
Professors van Rhee and de Vriend go as far as to use my explanation to
justify continued dredging along the entire length of the lower Río San
Juan. T he R ío Colorado also constitutes a Response reach, although
regional tectonic subsidence south of the Santa Elena - Hess Fault means
that flows and the sediment transport capacity of the R ío Colorado
naturally increases gradually through time. None of this was challenged by
Dr Kondolf in his 2013 Report.
5.7. However, both the 2014 Kondolf Report and the Andrews Report continue
to argue that additional coa rse sediment supplied to the lower R ío San
Juan from the Road presents a serious hazard to navigation that
necessitates continued dredging , at least in the first 3 kilometres
downstream of the Delta. In Section 5 C, below, I focus on the degree to
which aggradation in the lower R ío San Juan can be attributed to
construction of the Road. N avigational aspects of aggradation in the lower
Río San Juan are addressed in Section 6B.
B. The Tributary Deltas
5.8. On page 69 of the 2014 Kondolf Report in the section on ‘Morph ological
Impacts of Rte 1856’, Dr Kondolf describes how and why tributary streams
build sediment deltas at the point where they confluence with the R ío San
Juan. I agree with Dr Kondolf’s general description, but not the conclusions
89
233 he bases on that description that are specific to tributary deltas along the
south bank of the Río San Juan.
5.9. In my 2013 Report, I pointed out that deltas similar to, but sometimes
larger than, those along the south bank of the Río San Juan also exist on the
Nicaraguan bank of the River. I substantiated this observation with
photographs of thirteen north bank deltas . No GPS was available when
those photographs were taken, and so I was unable to record coordinates
for the deltas observ ed in May 2013. Hence, on receipt of a request from
Nicaragua for the coordinates in 2014 ( note of 25 March 2014, reference
HOL-EMB-046) another flight, with GPS available, was arranged to meet
that request and a new set of photographs w as supplied to Nicaragua on
21 May 2014 (referenc e ECRPB -071-14). Despite these photographs
having been supplied to Nicaragua, Dr Kondolf does not address them in
his 2014 Report. On the second flight, two additional north bank deltas
were observedand all 15 deltas are shown in Figure 5.2.
5.10. From the air, the Nicaraguan deltas appear very similar to, though perhaps
larger than, those on the south bank, even allowing for seasonal changes in
river stage that influence the proportion of each delta that is exposed . My
general conclusion is that, in terms of th eir morphological impacts on the
river, there is no significant difference between tributary deltas found
along the north and south banks.
5.11. I have been unable to establish whether the deltas recorded by Dr Kondolf
in Annex F of his 2014 Report existed prior to construction of the Road
because the cloud free, high -resolution satellite images necessary to
ascertain whether sediment deltas were present prior to 2011 are not
available for many of the locations . That said, cloud -free images for two
locations (Figure 5.3) establish that at least two deltas definitely pre- date
90
234the Road and I cannot rule out the possibility that this is actually the case
for most, if not all, of them.
91
235 1. 3.
92
2.
2365. 7.
93
4. 6.
237 9.
11.
94
8. 10.
238 .he
13. 15.
, which were also
the 2014 Kondolf Report
Appendix F of
shown in
se 95
d to tho
12. 14.
.
o San Juan
í
some are considerably larger than any of those photographed by Dr Kondolf along the south bank
Fifteen north bank deltas photographed from Costa Rican airspace in April 2014. These deltas are formed in sediment
2
Figure 5.from Nicaragua and of low flow in the Rbe compare
239 Figure 5.3 Pre- and Post-Road satellite images establishing that at least two of the eight
south bank deltas identified as being formed from sediment derived from the Road were
present prior to construction of the Road.
96
2405.12. Notwithstanding the overall similarity of deltas on the north and south
banks when viewed at a distance, Dr Kondolf’s closer, on the ground
inspection led him to differentiate between stream and Road -derived
sediments within the deltas, based on three characteristics of the
individual grains. In the third paragraph on page 70 of his 2014 Report, he
observes that sediment from the Road is, “ reddish in color and is easily -
crumbled (what we have previously referred to as ‘angular, friable clasts’ ),
reflecting the deeply -weathered hillslope from which the s ediment recently
came. These clasts are distinct from the more rounded, competent gravels
that one typically encounters in a natural stream, and which dominate the
deltas on the northern bank of the River.” This may sound like
inconsequential detail, but it is actually revealing for three reasons:
1. the fact that Road -derived clasts are friable indicates that they will
quickly weather down to rounded, gravel-sized particles, similar to those
that would have formed deltas along the south bank of the Río San Juan
prior to construction of the Road;
2. the fact that deltas along the north bank are formed in, “more rounded,
competent [i.e. less crumbly] gravels” indicates that these deltas are
formed from stream bed material that has been transported considerable
distances from its eroding source, that these grains do not crumble, and
that they will remain too large for the R ío San Juan to transport
downstream for yearsor decades; and
3. the fact that the Road-derived gravel particles (clasts) are angular despite
being friable ( easily-crumbled) indicates that they have only recently
entered the fluvial system.
5.13. With respect to south bank deltas, Dr Kondolf points out in paragraph 3 on
page 69 of the 2014 Kondolf Report that, “Some are pre- existing deltas of
97
241 natural streams on which road -derived sediment has deposited, while some
are completely new features built of sediment eroded from the road”. In the
last paragraph on page 70, he states that the new deltas are formed “ by a
‘lag deposit’ of coarser sedimen ts”: which suggests that these deltas only
exist due to the relatively large size of the Road -derived sediment that
forms them. It is likely therefore that at least some of them will decrease
in size as the friable clasts break-up.
5.14. Based on Dr Kondolf’s own observations concerning the friability o f Road-
gravel clasts it is likely that the half-life of their residence on deltas will be
measured in months rather than years. This is the case because these
clasts will quickly crumble to sizes easily entrained by rainy season
discharges in the R ío San Juan. Once entrained in the flow, any Road-
derived clasts will wear down (through attrition and corrosion) to highly
mobile fine sand, silt and clay sizes that are easily absorbed within the
existing load of the Río San Juan.
5.15. Where sediment derived from the Road has accumulated on a pre- existing
delta at the south bank, any local, small -scale impacts will be transitory
and short-lived. In contrast, the ‘competent gravels’ making up the north
bank deltas (and, presumably, the pre-existing south bank ones too) will
not crumble, so that these stable and persistent morphological features
will continue to provide morphological diversity and environmental
benefits to the River.
5.16. If Road-derived sediment has formed any entirely new deltas, these will be
removed by the Río San Juan as the mitigation works choke off the supply
of new clasts, those currently forming the delta disintegrate, and the River
entrains and transports the crumbling clasts away, quickly wearing them
down to sand, silt and clay-sized particles in the process.
98
242 5.17. In short, the friable nature of the “lag deposits” formed by clasts derived
from the Road means that any contribution they make to morphological
features in the River is insignificant due to their spatially restricted extent
and because their existence in the channel will be short lived.
C. Has coarse sediment eroded from the Road affected bed elevations in
the lower Río San Juan?
5.18. On the very first page of the 2014 Kondolf Report, Dr Kondolf considers
the possibility that Road-derived sediment is causing the bed of the lower
Río San Juan to aggrade faster than would otherwise be the case. His
concern is that sediment contributions f rom the Road are causing
morphological changes because, “the Lower San Juan is already overloaded
with sediment from Costa Rica’s other high contributions, such that
additional inputs are likely to aggrade and accrete” . Having stated this, he
does not further explore the issue, though it is addressed by Dr Andrews ,
in Annex 3 to Nicaragua’s Reply.
5.19. As I explained in Section II.2.1 of the 2011 Thorne Report and reiterated in
the 2013 Report, long -term aggradation of the lower R ío San Juan is
inevitable due to neotectonic uplift of the Chortis Block (which lies to the
north of the Santa Elena - Hess Fault), which dictates that the discharge of
water and transport capacity of this distributary channel will naturally
decrease gradually through time, to drive a long-term, depositional trend.
This explanation has been accepted by other experts. For example,
Professors van Rhee and de Vriend stated in their 2012 Report (Appendix
2 to Nicaragua’s Counter Memorial in Certain Activities) that, “As Thorne
correctly notes, river discharges to the Lower San Juan River will gradually
decrease without dredging due to the geological trends in the area (Thorne,
p. II-10)”.
99
243 5.20. While there is no question that the majority of sediment supplied to the
lower Río San Juan is sourced from Costa Rica; this is to be expected given
that the Costa Rican tributary basins are much larger than those draining
from Nicaragua and because they have mountainous headwaters, receive
heavier rainfall , and experience seismic events and volcanic erupt ions.
Thematic mapping (Figure 4.25 (f)) establishes that deforestation and
agricultural development have taken place in Nicaragua as well as Costa
Rica, and specific sediment yields estimated using distributed soil erosion
modelling for the Santa Cruz, Melchora and Sábalos basins are actually
higher than those in five of the seven sub-basins in Costa Rica (Table 4.19).
5.21. My conclusion is that Dr Andrews is wrong to state in paragraph 3 on page
27 of his Report that, “Poor land -use practices in Costa Rica ov er recent
decades have greatly increased the supply of sediment to the Río San Juan
Delta area.”. As explained in paragraphs 4.124 and 4.130 above, and in the
Astorga Report, sediment yields from some Costa Rican tributaries have
been high, and highly variable, for millennia because their basins have
steep, mountainous, rainy, headwater catchments that are tectonically -
active and feature live volcanoes that periodically supply huge amounts of
sediment to the Río San Juan between Lake Nicaragua and the Delta.
5.22. Dr Andrews criticizes my 2013 Report for ignoring the bedload transport
capacity of the lower R ío San Juan when estimating the increase in bed
elevation that could occur should sand from the Road be deposited rather
than passing to the Bay of S an Juan del Norte. I agree that when
determining the net rate of river bed aggradation it would be preferable to
consider both the supply of sediment and the rate of sediment transport
through the reach. I n my 2013 Report, I considered only the supply of
sediment in order to remove the uncertainty associated with attempting to
calculate the bedload trans port capacity of the lower R ío S an Juan, by
100
244 assuming it to be filled by coarse sediment from sources other than the
Road – making my estimate of the thickness of the layer that might be
deposited by sediment derived from the Road conservative. I prefer to
stick to that assumption, simply because the uncertainty associated with
estimates of bed material transport made using un -calibrated sediment
transport equations is notoriously high (see for example, Gomez and
Church, 1989), a point stressed by Dr And rews on pages 23 and 24 of his
Report.
5.23. As set out in Table 4.17 , under a ‘worst case’ rainfall scenario, the mean
annual input to the R ío San Juan of Road -derived sediment is 44, 880 m 3/y.
As in the 2013 Thorne Report (see paragraph 8.60, page 85) , again
assuming that 5 to 10% of that sediment is coarse (i.e. sand moving as bed
material load) suggests that the ‘worst case’ mean annual input coarse
sediment to the Río San Juan is 2,244 to 4,488 m /y. Based on 90% of the
discharge of the Río San Juan flowing to the Colorado and 10% to the lower
Río San Juan, ICE’s bedload computations suggest that, on average, 20%
3
(450 to 900 m /y or 750 to 1,500 t/y) passes to the lower Río San Juan (see
Figure 4.15, above).
5.24. Supposing that , as Dr Andrews suggests, all of the additional coarse
sediment (450 to 900 m 3/y) were to be deposited within 3 km of the Delta,
which I believe to be unrealistic (for reasons set out in paragraphs 5. 26 to
5.29, below). Spread across a channel 90 m wide and 3,000 m in length,
this would cause the bed to rise by an average of 1 to 3 mm, annually. Such
a rise would simply be unmeasurable even using the mo st sophisticated
hydrographic survey equipment. I agree that the actual distribution of
sedimentation would be spatial ly and temporally complex – making it in
practice impossible to separate the contribution to changes in bed level
attributable to the coarse fraction of Road -derived sediment from
101
245 continuous bed level changes associated with the migration of ripples
(with amplitudes of 10 to 30 mm), dunes (with amplitudes of 10 to 50 cm )
and bars (with amplitudes greater than 1 to 2 m ). In this context the
hypothetical addition of 1-3 mm is trivial. Also, this annual rate would not
be sustained because erosion from the Road will decrease as mitigation
takes eff ect, disturbed areas revegetate and slopes relax
geomorphologically, as expected according to the ‘rate law’ (Graf 1977).
5.25. In fact, sand deposition is not restricted to the first three kilometres of the
lower Río San Juan, but is distributed along its entire length. This is evident
for three reasons.
5.26. First, the lower Río San Juan has a mobile sand bed throughout its length.
This was established through bed material sampling performed as part of
the environmental impact study for Nicaragua’s dredging programme. In
the EIA report submitted to the Court in 2011 as Annex 7 in Volume II of
Nicaragua’s Counter Memorial in the Certain Activities case, the bed of the
lower Río San Juan is defined as being made up of:
(a) Delta – San Juanillo: gross to fine sand, with diameters 0.58 mm to
0.90 mm.
(b) San Juanillo – Mouth: fine to large sand, with diameters 0 .45 to 0.68
mm.
5.27. Second, continued growth of the micro-delta 30 kilometres downstream of
the Delta indicates that the lower Río San Juan has the capacity to
transport sand throughout its length.
5.28. Third, more than 20 sites where the bed has been dredged to r emove
accumulating sand during Nicaragua’s dredging programme are located
downstream of the first three kilometres of the lower Río San Juan (see
102
246 sketch map 5.1 on page 229 of Costa Rica’s Memorial in the Certain
Activities case).
5.29. Further reasons that it would be difficult, if not impossible, to demonstrate
a causal relationship between the addition of relatively small amounts of
coarse sediment derived from the Road and bed elevation change in the
lower R ío San Juan include : (1) complexity in the sediment transfer
system, (2) the time lag between coarse sediment input and downstream
response caused by the relatively slow transfer of bedload through the
fluvial system and (3) temporal variability in coarse load s supplied from
sources other than the Road.
5.30. Uncertainty in estimates of the quantity of bedload that enters the lower
Río San Juan at the Delta is extremely high for reasons set out in the ICE
Report that are summarized in Section 4 C, above. To understand why this
is the case consider that uncertainty stems from three different sources:
(a) time series variability due to inter -annual variability in sediment
loads;
(b) scatter in the data used to develop the sediment rating curves;
(c) the division of flows between the lower Río San Juan and Colorado at
the Delta, which is likely to vary seasonally.
5.31. To understand the impacts of these uncertainties on estimates of the mean
annual bedload in the lower Río San Juan, it is only necessary to consider
the confidence intervals listed in Table 4 .14 (which is reproduced in part
below, for ease of reference).
103
247 Table 4.14Mean Annual bedloads in the lower RíoSan Juan for a range of percentage
discharges flowingfrom the Río San Juanto the Río Colorado (from 2014 ICE Report).
Qa Annual sedimentload (t yr )
PRSJ 3 -1
(m s ) Mean TLCI TUCI SSRC LCI SSRC UCI
Bedload
95 1055 323 000 80 000 566 000 169 000 638 000
90 1114 702 000 174 000 1 230 000 359 000 1 418 000
85 1180 1 152 000 286 000 2 018 000 575 000 2 382 000
Note:
PSJR = Percentage of Río San Juan discharge flowing to the Río Colorado;
Qa = Mean annual discharge;
TLCI = lower 95% confidence interval due to time series variability;
TUCI = upper 95% confidence interval due to time series variability;
SSRC LCI= lower 95% confidence interval due to uncertainty in SRC;
SSRC UCI= upper 95% confidence interval due to uncertainty in SRC.
5.32. As Dr Andrews points out in the last paragraph on page 22 of the Andrews
Report, “the relative portions of annual flow in the delta distributary
channels cannot be determined with any confidence”. For this reason, ICE
tested the sensitivity of estimates of the bedload entering the lower R ío
San Juan based on bedload measurements in the R ío Colorado for
scenarios in which 85, 90 or 95% of the flow in the R ío San Juan
approaching the Delta passes to the Río Colorado. The results indicate that
depending on the choice of scenario alone (and without even considering
uncertainties associated with time series variability and uncertainty in the
sediment rating cur ve for Station 11 -04), the mean annual bedload
estimated to be entering the lower R ío San Juan ranges between 323,000
and 1,152,000 t/y (Table 4.15), which represents a range of -54% to +64%
around the best estimate for a 90:10% scenario, which is 702,000 t/y. The
additional sand load from the Road ( 591 to 1,181 t/y) constitutes 0.1 to
0.2% of the estimated mean annual bedload for a 90:10% scenario , which
is inconsequential in the context of the range in estimated bedload s
associated with uncertainty in the division of flow at the Delta. When inter-
annual variability in the time series of measured loads and uncertainty in
the sediment rating curve are also considered , the range of possible loads
104
248 expands to from eighty thousand to over two million tonnes per year and
the impossibility of proving that the Road has had any appreciable impact
on bed loads and changes in bed elevation in the lower R ío San Juan
becomes clear.
5.33. Were I to accept Dr Kondolf ’s 2014 estimate that the quantity of sediment
derived from the Road plus all the access roads delivered to the River
annually is between 116,000 and 150,000 m , which I do not, and applying
Dr Andrews’ assumptions that 10 % of that sediment is carried into the
Lower Río San Juan and that 12 to 18% of it is relatively coarse, then 1,390
3
to 2,700 m of sand from Route 1856 plus its access roads would be added
to the non- Road related coarse load . If all of this were to be deposited
within 3 km of the Delta, which I believe to be unrealistic, this would st ill
only cause the bed to rise by an average of 5 to 10 mm. Hence, by Dr
Andrews’ calculation, any change in average bed elevation would still be
unmeasurable. Also, sedimentation would be time- limited because any
supply of sand from the Road would decreas e as mitigation takes effect,
disturbed areas revegetate, and slopes relax towards equilibrium
according to the geomorphological ‘rate law’ (Graf 1977).
5.34. In any case, as Dr Andrews points out, division of discharges (and hence
division of coarse sediment load) at the Delta, “cannot be determined with
any confidence” , precluding the possibility of establishing a causal link
between construction of the Road and changes in bed level in the lower Río
San Juan until such time as the data necessary to determine the divisions of
discharges and sediment loads at the Delta have been collected.
5.35. On this basis, while it is almost certain that coarse sediment derived fro m
erosion of the Road cannot have had any discernable impact on either
sediment loads or bed elevations in the lower Río San Juan immediately
105
249 downstream of the Delta, it would be impossible to prove that it has had
any such impact.
5.36. Turning to the fine load, on page 29 of his Report Dr Andrews opines that,
“The finer sediment particles – fine silt and clay, which comprise a majority
of the river’s sediment – will be transported downstream along the delta
channels until the fresh river water begins to mix wi th tidal surges of ocean
water” and that consequently, “The vast majority of the relatively fine
sediment will be deposited within the delta and not carried into the ocean as
Thorne states”.
5.37. Dr Andrews’ opinion is contradicted by the available field data and satellite
imagery. In the ‘Environmental Impact Study for Improving Navigation on
the San Juan de Nicaragua River (September 2006)’ that was submitted as
Annex 7 in Volume II of Nicaragua’s Counter Memorial in the Certain
Activities case, the bed material of the lower R ío San Juan at its outlet to
the Caribbean Sea is defined on page 10 as consis ting of, “clean, fine sand
grains, with diameters from 0.31 to 0.58 mm”. No mention is made of either
silt or clay being present in the bed – on the contrary, the description of
the sand as ‘clean’ indicates that fine sediment is not deposited within the
delta channel.
5.38. Dr Andrews’ opinion might be correct for a delta building into a marine
water body that experiences frequent “tidal surges”. But the Caribbean has
a micro-tidal regime, with a diurnal tidal amplitude averaging only about
20 cm (Kjerfve, 1981). This explains why most of the fine sediment carried
by the lower R ío San Juan is not deposited within the delta but is carried
into the Caribbean Sea, as I indicated in my 2013 Report and as illustrated
in typical, rainy season satellite images (Figure 5.4), that show plumes of
106
250 turbid river water extending into the Bay or San Juan del Norte and the
littoral zone of the Caribbean Sea.
(a)
(b)
Figure 5.4 Satellite images showing that flow from the lower Río San Juan carries turbid
water with ahighconcentration of fine sediment into both the Bay of San Juan del Norte
and the littoral sediment system of the Caribbean Sea. Image dates (a) 13 December 1997
(b) 26 November 2013.
107
251 6. Has Route 1856 had any significant impacts on ecology or fishery
of the Río San Juan, or any impact on navigation?
A. Comments on Fish and other Aquatic Life in the Río San Juan by
Nicaragua’s Experts
6.1. On pages 63 to 65 of the 2014 Kondolf Report, Dr Kondolf sets out the
negative impacts that increased delivery of coarse and/or fine grained
sediment can have on aquatic species and habitats. Taking particular issue
with the assertion on page 50 of the 2013 Thorne Report that, “Fish and
other aquatic organisms in the Río San Juan do not find high turbidity
problematic because they are fully adapted to it” , he notes that I presented
no citations to scientific literature to support my assertion.
6.2. To counter my assertion, Dr Kondolf cites papers drawn from the
literature concluding that, “What the lit erature actually demonstrates is
that some of the most prevalent fish known to exist in the Río San Juan (as
reported in Procuenca 2004 and the EDA, Annex 10), such as Cichlids,
members of the family Mugiliidae, and Poecilids, are vulnerable to increases
in turbidity and suspended sediment”.
6.3. With respect to other aquatic life in the R ío San Juan, Dr Kondolf concludes
that “The heavy loads of suspended sediment have a negative effect on algal
and macroinvertebrate communities in the Río San Juan, as evidence d by
differences in ecological communities established on deltas on the north
bank, at the mouths of streams draining forest preserve in Nicaragua, which
are not affected by Rte 1856, contrasted with those established on the south -
bank deltas, which are affected by sediment eroded from the road”. He relies
for this conclusion on a field investigation performed by Dr Blanca Rios
entitled, ‘Ecological Impacts of Rte 1856 on the San Juan River (2014) ’ (the
Rios Report), which is Annex 4 in Nicaragua’s Reply.
108
252 6.4. As a river scientist, I take it to be self-evident that species in the R ío San
Juan that are not adapted to high turbidity would have either:
(a) been extirpated, given that according Dr Andrews’ explanation
(which I reject) turbidity has been high due to defo restation for half
a century; or
(b) had time to adapt to high turbidity ( accepting my explanation that
the sediment load of the River is naturally high and has been so for
millennia).
6.5. Also, I note that on page 66 of the ‘Environmental Impact Study for
Improving Navigation on the San Juan de Nicaragua River (September
2006)’ that was submitted to the Court in 2011 as Annex 7 in Volume II of
Nicaragua’s Counter Memorial in the Certain Activities case, in the section
describing planktonic and benthonic organisms in the R ío San Juan,
Nicaragua’s ecologists reported that, “A low density of organism was noted
both in the water and in the sediment, with the predominance of species that
are tolerant and adapted to adverse cond itions” noting that, “the locations
where the samples were taken is located at the outlet of the San Juan River
where the majority of the contaminants have been carried, as well as gross
sediment, which translates into the presence of tolerant species.”
6.6. However, I am not a specialist in aquatic biology and, while I have a good
working knowledge of river ecology, neither am I an expert in fish or
macroinvertebrates. For these reasons I draw here on an independent
report by Professor Ian Cowx (the Cowx Report).
B. Fish
6.7. The independent expert report by Professor Ian Cowx, an internationally
recognized leader in the m anagement of inland fish eries and aquatic
109
253 resource, deals primarily with fish and the fishery of the R ío San Juan. It
directly addresses Dr Kondolf’s overall conclusion (on page 66 of the 2014
Kondolf Report) that, “ Professor Thorne’s unsupported assertion that “Fish
and other aquatic organisms in the Río San Juan do not find high turbidity
problematic because they are fully adapted to it” is not only inconsistent with
the literature on the species of fish and macroinvertebrates known to exist in
the San Juan River, but also inconsistent with recent aquatic ecology
sampling in the San Juan River itself.”
6.8. Professor Cowx finds statements presented in the 2014 Kondolf Report to
be either over-generalised, fundamentally flawed or to misinterpret the
peer reviewed literature. He finds expert opinions stated in the 2014
Kondolf Report that pertain to the possible impacts of sediment derived
from the Road on fish to be unsupported by empirical evidence fr om the
San Juan itself, while those based on the published literature are taken out
of context.
6.9. Professor Cowx notes that the annual hydrograph of the R ío San Ju an
exhibits a wet season flood pulse typical of tropical rivers and that its
sediment load is consistent with this, being naturally high and variable
(Bussing 2002). He finds that fishes of the San Juan are well adapted to
high and variable sediment loads, being accustomed to high and variable
turbidity (Bussing 2002).
6.10. With respect to the species of fish living in the river adjacent to the Road ,
Professor Cowx points out that these have not been explicitly defined by
Nicaragua’s experts, who describe their characteristics at family level
rather than providing information that is specific to species actually
present in the Río San Juan. He notes that it would require intense research
110
254 using specialist equipment over a protracted period to identify the species-
specific adaptations of fish living in the River.
6.11. With respect to commercially valuable species (the fishery) , Professor
Cowx finds it likely that the coastal fishery is more productive than inland
fishery, mostly due to exploitation of snook an d sport fisheries for tarpon.
In his opinion, th e coastal fisheries are unlikely to be adversely impacted
by any additional sediment loading from the Road.
6.12. Professor Cowx closes the first part of his report (dealing with fish and the
fishery) by stating that, “My literature review, together with close inspection
of literature cited in the 2014 Kondolf Report, provides the basis to evaluate
Dr Kondolf’s general statement on page 64 that, What the literature actually
demonstrates is that some of the most prevalent fish known to exist in the Río
San Juan (as reported in Procuenca 2004 and the EDA, Annex 10), such as
Cichlids, members of the family Mugiliidae, and Poeciliids, are vulnerable to
increases in turbidity and suspended sediment”.
6.13. Professor Cowx goes on to conclude that, “ What the literature actually
demonstrates is that Dr Kondolf’s statement is a gross over -generalisation.
While some members of the families of fishes he names are vulnerable to
increases in turbidity and suspended sediment, others members of those
families are adapted to high sediment loading and this is illustrated through
the species specific review summarised herein and reported in detail in the
references cited.”
6.14. He further notes that, “ Empirical data on the species impacted with
particular reference to the San Juan River are required to justify and
substantiate claims of any long -term impact of construction of Route 1856
on the fish and fisheries of the river. No such data have been provided by
111
255 Nicaragua’s experts. The examples used as ev idence are general and
unspecific to the San Juan River and the species that inhabit it. ”
6.15. Professor Cowx therefore concludes that, “there is no evidence that the
fish and fisheries of the San Juan have or will be impacted by
construction of Route 1856.” [his emphasis through use of bold text].
6.16. An expert literature review on fish in the R ío San Juan was performed as
part of the 2014 ecological a ssessment by t he Tropical Science Center
(CCT). This report, titled ‘Fish Fauna in the San Juan River’, was authored
by the Costa Rican fish expert, Arturo Angulo Sibaja, (the Angulo Report).
6.17. The Angulo Report is a technical treatise that draws on examples from
multiple rivers in Costa Rica. Although no data are available for the R ío San
Juan in the reach adjacent to the Road, Angulo report s data for the Río
Colorado (which is the downstream extension of the R ío San Juan,
receiving about 90% of its flow). Despite its high and variable sediment
load, the R ío Colorado has one of the most diverse freshwater fish
assemblages found in Central America (Bussing 1998, Angulo et al. 2013),
with about 115 species, which is 46% of those found in Costa Rica (Angulo
et al. 2013).
6.18. Information on fish species living in tributaries to the R ío San Juan that
carry suspended sediments at concentrations even greater than those in
the main river is also germane. For example, at the Terrón hydrometric
station located on the Río San Carlos, Angulo cites a specific sediment yield
of 817 t /km /y and ICE report an annual suspended sediment load of
1,175,000 t/y (see Table 4.10). Yet the R ío San Carlos is home to no less
than 54 fish species (including Cichlidae (n = 15), Poeciliidae (n = 10) and
Characidae (n = 8) ) that apparently do not find problematic the high
turbidity associated with such heavy loads of suspended sediment.
112
256 6.19. In the Reventazón basin, which like the R ío San Juan, drains to the
Caribbean Sea, specific sediment yields of up to 1,159 t/km /y have been
measured (Jimenez et al. 2005). Even so , the river supports a diverse fish
population with 65 species including Cichlidae (n = 15), Poeciliidae (n = 6)
and Characidae (n = 5) (Molina 2011).
6.20. In the Angulo Report, it is concluded that, “ The presence of these taxa in
rivers with high sediment yield s might suggest high levels of tolerance, as
various authors have proposed (Bussing 1998 , Tiffer-Sotomayor 2005, Rojas
and Rodriguez 2008, Saenz et al. 2009), and is supported by the presented
revision.”
6.21. In the Cowx Report, Professor Cowx endorses the find ings of the Angulo
Report, writing that his report, “acknowledges and builds on the insightful
comments made by Arturo Angulo Sabaja”.
C. Invertebrates
6.22. In section 3 of the Cowx Report, Professor Cowx re -examines evidence
presented in the Ríos Report related to the impact of the Road on
macroinvertebrates in the R ío San Juan and reviews statements made in
the 2014 Kondolf report that rely on the outcomes of the Ríos Report.
6.23. Professor Cowx finds that, while the Ríos Report appears to show
differences between mac roinvertebrate communities (and associate d
parameters) on deltas at the north and s outh banks of the R ío San Juan
River, these findings are compromised by the fact that the drainage areas
of the northern bank deltas are systematically l arger than those of the
south bank deltas. Also Dr Ríos does not consider or account for the effects
of natural vegetation and catchment land use on stream water quality and
delta habitat. He contends that these confounding factors may account, at
113
257 least in part, for the differences that both Dr Ríos and Dr Kondolf attribute
to sedimentation from the Road. This point is also stressed in a review of
the Ríos Report conducted by Bernald Pacheco Chaves, an aquatic ecologist
at the Tropical Science Center (CCT) (the Pacheco Report).
6.24. Professor Cowx draws attention to low abundance, low richness and high
within site variability in macroinvertebrate assemblages, which sug gest
that conditions at all of the sample sites are dynamic and variable. He
notes that while multiple patterns and differences are reported by Dr Ríos,
these may be attributed to differences in the areas , natural vegetation and
land use in the catchments draining to the deltas, which are not controlled
for in the statistical analyses. Professor Cowx notes that the sam e
conclusion is arrived at in the statistical review by Gutiérrez (2014) (the
Gutiérrez Report).
6.25. I have reviewed the Gutiérrez Report: though I am a user of statistics, and
am familiar with the appropriateness, strengths and weaknesses of
alternative parametric and non-parametric tests, I am not an expert in that
field. Therefore I am not well placed to judge the degree to which the
errors highlighted therein undermine Dr Rios’ conclusions. That said,
setting Gutiérrez’s comments alongside those of Professor Cowx inevitably
reduces confidence in the value of the statistical support used to justif y Dr
Rios’ conclusions, and the statements by Dr Kondolf that rely on them.
6.26. Professor Cowx summarises his re -examination of the Ríos Report thus, “I
consider that the Ríos (2014) Report does not provide the evidence necessary
to prove that construction of the Road has adversely impacted the benthic
macroinvertebrates living in sediment deltas along the southern bank of the
San Juan River.”
114
258 6.27. The Pacheco Report comes to the same conclusion, viz., “It is considered
that the study of Rios Touma (2014) does not provide valid evidence to
demonstrate significant degradation of aquatic communities in the San Juan
river due to sediment discharge by works in Route 1856”. I note that the
Pacheco Report reaches that conclusion on the basis of a deeper technical
treatment than that performed byProfessor Cowx.
6.28. Pacheco drills into issues associated with contrasts in land cover, land use
and riparian vegetation in sub-catchments draining to the north and south
banks. He points out that 14 of the 16 sites sampled by Dr R íos are located
in a single, short reach between the mouths of the Infiernito and San Carlos
tributaries (Figure 6.1).
Figure 6.1 Sampling Points along the San Juan River between El Castillo and Boca San
Carlos. Each point corresponds to a delta formed by a creek draining to Río San Juan ( This
is Figure 1 in the Ríos Report).
6.29. Pacheco notes that this short reach features what is probably the highest
local concentration of cut slopes and large fill prisms anywhere along the
Road, making it highly atypical of conditions more generally. Yet, the
115
259 results of the Ríos study are presented as though they represent the entire
length of the Road, or at least that between Marker II and Boca San Carlos.
6.30. Professor Cowx criticizes Dr Kondolf for citing examples of sediment
intolerance in fish that were made at family level, and drawn from
literature for species found in rivers other than the R ío San Juan, rather
than using examples specific to species known to live in the R ío San Juan.
Similarly, Pacheco criticizes Dr R íos for mentioning taxa found to be
sensitive to sediment based on studies in the United States (Zweig &
Rabeni 2001; Carlisle et al. 2007).
6.31. Pacheco points out two major flaws in this regard. First, the studies that Dr
Ríos uses as reference do not correspond to the R ío San Juan, Costa Rica ,
Central America or even the T ropics. Environmental conditions found in
temperate areas are obviously different to those in the T ropics, and
macroinvertebrates are known to respond differently to environmental
stimuli even in different regions of the same country (Heino 2014). Second,
the level of taxonomic resolution used by Dr R íos reaches only to the
family and gender level, which is normal because taxonomic identification
to species level in macroinvertebrates is often not possible with the
scientific literature published to date and requires a high degree of
taxonomic expertise.
6.32. For this reason, it is very difficult to know whether or not the species
studied in the literature that Dr Ríos cites in her references correspond to
species present in R o San Juan. Although some species may occur both in
the rivers cited in the literature and in the R ío San Juan, this is unlikely.
The point here is that the tolerances to sediments of macroinvertebrates
may vary depending on the taxonomic resolution used (Bailey et al. 2001),
and may vary even between different species within the s ame genus
116
260 (Flowers 2009), making it unreliable to use macroinvertebrates as bio -
indicators of environmental and water quality deterioration on the basis of
family and gender-level taxonomic resolution – which is precisely what Dr
Ríos does in her Report.
6.33. Pacheco points out that Dr Ríos concludes that construction of Route 1856
has had no impact on deltas along the north bank of the R ío San Juan. This
indicates that, even if construction of the Road were to ha ve had any
significant impact on aquatic life in the R iver (which I do not accept) , that
impact would be restricted to a few deltas along the Costa Rican bank ,
contained within a short reach of the River.
6.34. If erosion and delivery of sediments from the Road had caused a significant
increase the concentration and l oad of sediment in the River then, as
Pacheco points out, this would be expected to have altered conditions
throughout the aquatic environment, impact ing not only isolated spots
along the near bank, but also the bed and the opposite bank - especially
given that the channel in Dr R íos’ study reach is less than 200 m wide.
6.35. In Pacheco’s opinion, Dr R íos’ finding that the north bank of the river has
been unaffected by construction Route 1856 contradicts statements by
Nicaragua and Nicaragua’s experts to the effect that the Road has done
significant harm to aquatic life in the Río San Juan.
6.36. Turning to the 2014 Kondolf Report, P rofessor Cowx writes that, in his
expert opinion, in using m acroinvertebrate fauna as indicator s of
environmental degradation (see page 65), Dr Kondolf misses the point that
macroinvertebrates are better bio -indicators of adverse impacts of water
quality than they are of deterioration of hydromorphology (Bonada et al.
2006, Resh 2008). Consequently, Professor Cowx questions the reliability
117
261 of using macroinvertebrate studies in the R o San Juan River to infer that
the Road has had significant hydromorphic impacts on the River.
6.37. In this context, Professor Cowx points out that, on page 65 of his 2014
Report Dr Kondolf concludes that, “The heavy loads of suspended sediment
have a negative effect on algal and macroinvertebrate communities in the
Río San Juan, as evidenced by differences in ecological communities
established on deltas on the north bank, at the mouths of streams draining
forest preserve in Nicaragua, which are not affected by Rte 1856, contrasted
with those established on the south -bank deltas, which are affected by
sediment eroded from the road”. To support his conclusion, Dr Kondolf cites
evidence from the Ríos Report, which found “much higher EPT abundance
and richness, on deltas on the north side of the Río San Juan, than on the
south-bank deltas impacted by sediment from the road.” Dr Kondolf explains
the signif icance of the EPT results thus; “EPT refers to the orders
Ephemeroptera (mayflies), Plecoptera (stoneflies), and Trichoptera
(caddisflies), which are known to be sensitive to sediment and other
pollutants, and thus are important indicators of water quality.”
6.38. EPT fauna are indeed important indicators of water quality, but they are
less reliable in indicating hydromorphological impacts due to the dynamic
nature of hydromorphological features such as sediment deltas. Professor
Cowx’s point is that not only are the EPT richness and abundance scores
very low for all the deltas studied ( to the point that th ey lack statistical
robustness, as discussed in the Gutiérrez Report), but also differences
between northern and southern bank deltas can be attributed to
differences in water quality in streams draining to the deltas that are
attributable to contrasts in basin areas, vegetation and land use. The
failure of Dr Ríos’ study to control for confounding factors such as the
effects of agricultural development thereby becomes doubly significant.
118
2626.39. Professor Cowx concludes his indep endent review by stating that;
“Evidence provided in the Ríos Report that compares environmental bio-
indicators for deltas on the northern and southern banks is largely
inconclusive and fails to provide the robust empirical data necessary to prove
that sediment eroded from the Road has adversely impacted the aquatic
ecology of the San Juan River. ” He adds; “It is therefore unsound for Dr
Kondolf to conclude that Road -derived sediment has had negative effects
[on] invertebrate communities in the San Juan River.”
6.40. In paragraphs 5.13 to 5.18, above, I set out m y interpretation of the
significance of sedime ntological differences between R oad-derived and
stream-bed sediments identified by Dr Kondolf during his field inspections
of deltas on the north and south banks of the Río San Juan. I explain below
why the specific properties of the Road-derived sediment observed by Dr
Kondolf mean that any influence it may have on periphyton and
macroinvertebrate assemblages is localised and temporary.
6.41. In the 2014 Kondolf Report, Dr Kondolf accepts my proposition that fresh
sediments provide new substrate for periphyton (algae and other
organisms growing on the surfaces of gravel and rock) and ma cro-
invertebrates to colonise.
6.42. Recognizing this, t he important point is that Road-derived sediment
deposits do not provide habitat that is intrinsically inferior or of low er
quality, but that Road -derived clasts are larger, fresher and cleaner than
older stream gravels, having only recently been sourced from newly-
exposed bodies of sedimentary rock. The issue is that such clasts are too
clean to support abundant and diverse microbial ecosystems and they
have not been in the fluvial system long enough to develop the rich
periphyton that is the base of the food chain. But this is a transitory
119
263 condition. Given time, energy and a supply of nutrients, the larger, fresher
clasts lose their angularity, grade from cobble to gravel sizes (due to their
friable composition) and grow a slimy coating – the periphyton that
provides the food source necessary to support succession towards rich er
and more diverse macroinvertebrate communities . That said, and as
Professor Cowx points out, sampled macroinvertebrate richness, diversity
and EPT scores are low for north as well as south bank deltas. This reflects
the morphodynamic nature of the deltas, which accumulate sediment
derived from the streams that feed them during localized rainstorms , but
which are periodically disturbed by floods in the main river that re-entrain
and distribute the stream -deposited tributary delta material along the
bank and across the bed of the Río San Juan. All of this is a consequence of
the naturally high sediment load of the River, as explained in paragraphs
4.109 to 4.128, above.
6.43. Professor Cowx criticizes the 2014 Kondolf Report for placing too much
emphasis on the use of macroinvertebrate fauna as indicators of
environmental degradation. Dr Kondolf also interprets periphyton and
macroinvertebrates as bio -indicators of water quality, stating on page 71
that the new deltas, “are also subject to unnaturally high and deleterious
suspended sediment loads, which result in communities of algae and
macroinvertebrates that reflect deteriorated water quality conditions ”. But
this statement is not supported by any evidence.
6.44. On page 11 (paragraph 2) of his Third Report, Dr Kondolf reports
suspended sediment concentrations in three samples of muddy -water in
plumes in the River , which had entered the River following a 15 -minute
downpour. The samples had SSCs of 364, 459 and 483 grams per cubic
metre. Dr Kondolf describes these SSCs as ‘ high’. He also took two samples
of River water, both of which had SSCs of 8 grams per cubic metre.
120
264 6.45. The 2013 Thorne Report includes a graph (Figure 18 in the 2013 Thorne
Report, r eproduced here as Figure 6 .2) showing 2,409 individual
measurements of suspended sediment concentration in tributaries to the
R o San Juan and the R ío Colorado at the Delt a – which carries about 90%
of the flow in the Ro San Juan immediately upstream.
6.46. SSCs measured in this large data set vary from less than 10 parts per
million to mor e than 10 000 parts per million, clearly illustrating the
extreme natural variability in sediment concentrations, and associated
levels of turbidity, characteristic of rivers in the Río San Juan – Colorado
system.
Figure 6. 2 Suspended sediment concentration as a function of discharge f or 2, 409
samples taken from the Río Colorado, Río San Juan and its Costa Rican tributaries. Note:
Station 11-04 is the Delta Colorado (Station) which receives about 90% of the flow in the
Río San Juan immediately upstream (from the 2013 Thorne Report).
6.47. The background SSC in the Río San Juan at the time it was measured by Dr
Kondolf was indeed low – as I have pointed out, SSCs and suspended
sediment loads are extremely variable and May is the low water season.
However, sediment concentrations in the plume of muddy -water he
121
265 sampled are not ‘high’ in the context of SSCs routinely observed in runoff
draining to the Río San Juan, or even in the River itself (as represented by
Station 11-04 in Figure 6.2).
6.48. It is unsurprising that a 15-minute rainstorm in May produced a striking
contrast between SSCs in local runoff and the receiving water because
under these circumstances the source of sediment is localised to the area
of the rainstorm, while discharge and background SSCs in the River (which
in May is at its lowest (base flow) discharge) ar e at their lowest. However,
judging from the existence of deltas at their tributary confluences, plumes
with similarly high SSCs will be associated with local storm runoff from
streams confluencing at the north as well as the south bank.
6.49. In his Fourth Rep ort, Dr Kondolf accepted that the sediment
concentrations he measured in the muddy plume, “were not very high
compared to concentrations measured in the river and its large tributaries
1
during high flows.” In doing so, he effectively retracted the statement in
his Third Report that these measurements showed, “that the runoff from
the road carried high suspended sediment contributions”. 2Dr Kondolf went
on to state in his Four th Report that the measurements “demonstrate the
essential fact that sediment from the road is entering the R ío San Juan.” I 3
agree, but the central point remains this: in order to assess whether the
concentrations of suspended sediment measured in runoff from the Road
have harmed or may in future cause harm to life in the River, it is
necessary to consider them within the context of sediment concentrations
that aquatic plants and animals in the river system experience routinely
1 Fourth Kondolf Report, p. 11.
2
Third Kondolf Report, p. 11.
3 Fourth Kondolf Report, p. 11.
122
266 and to which they are well adapted. The measured data in Figure 6.2 show
that concentrations in tributaries to the R ío San Juan often exceed 500
grams per cubic metre and so those measured in May 2013 (364, 459 and
483 grams per cubic metre) have not damaged, and will not damage life in
the River.
6.50. For these and all the other reasons set out in the Cowx and Angulo Reports,
the wide range of SSCs and seasonal and local variability therein are not a
water quality problem and suspended sediment cannot be considered as a
pollutant. High SSCs during the annual flood pulse and associated
individual rainstorms (which occur throughout the year) are, as I wrote in
my 2013 Report, “a long -standing fact of life to which the River’s aquatic
and riparian ecosystem is fully adapted”.
D. Navigation
6.51. In its Memorial in the Construction of a Road case, Nicaragua claimed that
constructing the Road had involved “Dumping of trees and soil along the
route of the road into the river flow, making more difficult and riski ng the
navigation in its waters”. This claim of adverse impacts on the navigability
of the Río San Juan due to construction of the Road clearly applies to the
reach along the route of the Road between Marker II and Delta Costa Rica.
6.52. Nicaragua’s experts have made no mention of trees or sediment from the
Road causing risks to navigation in the R ío San Juan between Marker II and
Delta Costa Rica and neither have I seen any evidence that this has , or is
likely to , happen during any of my fieldtrips and overflights. On the
contrary, I have observed Nicaraguan vessels navigating the R o San Juan
between M arker II and Delta Costa Rica without difficulty on multiple
occasions, during the low water as well as the high water season.
123
267 6.53. In any case, in- stream deposition of trees and soil is part of the natural
functioning of the Río San Juan . This is not only my opinion but is also
documented in the ‘Environmental Impact Study for Improving Navigation
on the San Juan de Nicaragua River ( September 2006)’ that was submitted
to the Court in 2011 as Annex 7 in Volume II of Nicaragua’s Counter
Memorial in the Certain Activities case. On page 1 0 of this document, in
reference to a bar in the Río San Juan at the point where the channel
widens as it approaches the Delta, it is stated that, “During probes of this
section, the remains of organic material were found, including trees 0.6
meters in diameter buried under the riverbed up to the depth of 3.00 meters. ”
The authors go on to explain that the origin and functioning of t hese
deposited trees writing that they, “come downstream during periods when
the river is very high go downstream, bumping aga inst the islands, which
retain them, serving as energy dissipaters, retaining some sediment which,
upon accumulation during the year increases the size of the islands or causes
new, small islands to be created...”. This passage confirms that large woody
debris was present in the Río San Juan long before construction of the
Road and that it plays a natural role in the geomorphic and environmental
functioning of the River.
6.54. In his 2012 R eport, Dr Kondolf referred to “significant damage to aquatic
and wetland environments and navigability of the lower Río San Juan
through causing excessive sedimentation” . This claim refers not to
navigation in the reach alongside the Road, but to navigation in the lower
R o San Juan, downstream of Delta Costa Rica.
6.55. Based on the reasoning explained in paragraph 5.29 above, and supposing
that, as Dr Andrews suggests, all of the Road-derived sand that might reach
Delta Costa Rica and enter the lower R ío San Juan were to be deposited
within the first 3 km downstream of the delta, wh ich I believe to be
124
268 unrealistic for the reasons set out above, in paragraphs 5.26 to 5.29, this
would cause the bed to rise by an average of 5 to 10 mm.
6.56. Even allowing for deposition being concentrated on bars, deposition of all
the Road-derived sand in the first three kilometres of the lower R ío San
Juan would be insufficient to significantly affect navigation. In any case,
variability in the sediment loads entering the lower R ío San Juan would
make it, in practice, impossible to discern the contribution to increasing
bed or bar heights attributable to Road -derived sand from long-term
trends of aggradation, bar building during floods and the continuous
changes in bar elevations resulting from seasonal and event- driven
variability in flows and sediment loads, and other morphological
adjustments in the fluvial system.
6.57. If sedimentation does pose a problem for navigation during the dry season,
this is attributable to an aggradational trend that affects the entire river,
being driven by natural tectonic and fluvial processes that, for the reasons
explained in the Astorga Report, have operated in the lower R ío San Juan
for millennia. This, geological, explanation for aggradation in the lower Río
San Juan is endorsed by other experts. For example, Professors van Rhee
and de Vriend wrote in their 2012 report (Appendix 2 to Nicaragua’s
Counter Memorial in Certain Activities) that, “As Thorne correctly notes,
river discharges to the Lower San Juan River will gradually decrease without
dredging due to the geological trends in the area (Thorne, p. II-10).”
6.58. I conclude that any navigation problems in the lower Río San Juan are
associated with long -term aggradation driven by natural processes. T he
addition of Road -derived sand to the sediment load of the lower Río San
Juan cannot in itself have impeded navigation, nor can it be proven to have
caused the bed elevation in the river to rise by any measurable amount.
125
269 7. What effect have Costa Rica’s Mitigation Works had, and how
have they progressed since late-2013?
A. CODEFORSA’s Reforestation and Slope Recovery programmes
7.1. In April 2012, as part of implementation of the Environmental
Management Plan for the Road, the Commission for Forestry Development
in San Carlos ( Comisión de Des arrollo Forestal De San Carlos o r
CODEFORSA) was contracted to plant and maintain 25,000 trees at 12 sites
along Route 1856 between Marker II and Boca San Carlos . In 2013, the
contract was extended to add a further 19 sites, with provision to plant
and maintain 24,000 more trees. To date, a total of 50,709 trees have
actually been planted, covering an area of around 46 hectares, 98% of
which lies between the Road and the south bank of the Río San Juan. These
trees are currently aged between 2 and 28 months and rang e between 50
cm and 7 m in height.
7.2. Full details of this programme of reforestation and revegetation may be
found in reports authored by CODEFORSA and titled, ‘ Consulting Services
for the Development and Implementation of an Environmental Plan for the
Juan Rafael Mora Porras Border Road’ (the 2014 CODEFORSA Report) and
‘Restoration and rehabilitation of ecosystems affected by the construction
of the Juan Rafael Mora Porras border road, Ruta 1856’ (the C ODEFORSA
Quarterly Report for November 2014).
7.3. Here, I focus on responding to Dr Kondolf’s dismissal of the reforestation
programme on page 42 of the 2014 Kondolf Report , where he wrote,
“Annex 2 includes photographs of a tree- planting program, but does not
provide essential information such as whether the plantings will actually
address slope s tability issues (the answer in most cases will be no, because
the failure planes of landslides would be deeper than the rooting depth of
126
270 plants), and whether the plants have survived since planting (in our
observations from the river, it appeared that most have died)”.
7.4. With regard to provision of essential information concerning the
reforestation programme, the 2014 CODEFORSA Report provid es
thorough accounts of both phases , plus a two-page information sheet for
each of the 31 sites that documents the location, planting date,
maintenance activities and current status of the trees and which includes a
time sequence of site photographs. These sites are not restricted to the
first 15 km of the Road downstream of Marker II.
7.5. Dr Kondolf is right to surmise that, “the failure planes of landslides would be
deeper than the rooting depth of plants ” and inspection of the 2014
CODEFORSA Report confirms that trees have not been planted on slopes
prone to instability due to deep -seated landsliding. In fact, other steps are
being taken to improve the stability of cut and fill slopes that are high and
steep, as is described in detail in the 2014 CONAVI Report, CODEFORSA’s
Quarterly Report for November 2014, and in summary below.
7.6. Areas designated for reforestation (including gentle but not steep slopes)
were selected based on their being locations where trees can effectively:
(a) reduce the erosivity of rainfall by intercepting precipitation;
(b) reduce the erodibility of the soil by decreasing soil moistur e levels
through evapotranspiration and by providing root reinforcement;
(c) reduce the generation of overland flow by increasing infiltration;
(d) intercept surface runoff along concentrated flow paths by increasing
surface roughness and ground permeability, to protect the soil and
downslope areas from sheet, rill or gully erosion;
127
271 (e) intercept surface runoff that might otherwise reach the Río San Juan;
and,
(f) create valuable wildlife habitat.
7.7. When I first read the 2014 Kondolf Report in August 2014, I was surprised
by Dr Kondolf’s questioning of, “ whether the plants have survived since
planting” and that, “(in our observations from the river, i t appeared that
most have died)”. In fact, my own observations in April 2014 indicated that
the vast majority of th e trees and grasses plant ed by CODEFORSA up to
that time had survived and were, indeed flourishing.
7.8. Having now visited the reforestation sites three times in 2014, having
spoken at length to CODEDFORSA’s forestry experts and having reviewed
the CODEFORSA Reports, I am convinced that Dr Kondolf’s statements are
groundless.
7.9. Both t he phase 1 and 2 contracts awarded to CODEFORSA included
provision for two years of post -planting maintenance, with activities
specified to include:
(a) Monitoring of planted areas (including the health of the trees);
(b) Mowing;
(c) Spot herbicide treatment around trees;
(d) Fertiliser application;
(e) Removal of suckers;
(f) Maintenance of the fences; and
(g) Follow-up visits to the planted areas.
7.10. In CODEFORSA’s contract, the number of trees growing at the end of the
contract period i s specified rather than the area to be covered.
Consequently, throughout phase 1 of the project, CODEFORSA identified
128
272 and replaced lost trees within the maintenance programme. This is also
happening during phase 2 (which is on-going).
7.11. Table 7 .1 lists data from mortality reports for maintenance visits
performed between 2012 and2014.
Table 7.2 Mortality report for the CODEFORSA reforestation programme
(from the 2014 CODEFORSA Report).
Phase 1 (26,575 TREES PLANTED)
MORTALITY IN MORTALITY IN
NAME OF THE PARTY TREES 2012 2013 MORTALITY IN 2014
PLANTED
N % N % N %
Escuela Delta Costa Rica 325 28 8,6% 25 7,7% 0 0,0%
Escuela y Policía 500 0 0,0% 25 5,0% 0 0,0%
Tito Hernández Ferreto 366 14 3,8% 5 1,4% 0 0,0%
María Hilaria Miranda Rivas 500 30 6,0% 20 4,0% 0 0,0%
Felix Hernández Jarquín
260 17 6,5% 15 5,8% 0 0,0%
Fabio Vargas 407 52 12,8% 10 2,5% 0 0,0%
Escuela Boca La Ceiba 117 10 8,5% 5 4,3% 0 0,0%
Melis Góngora Moraga
252 25 9,9% 5 2,0% 0 0,0%
Iglesia Boca La Ceiba 225 30 13,3% 7 3,1% 0 0,0%
Fredy Ulate Castro 3180 150 4,7% 0 0,0% 0 0,0%
Fabio Cedeño G. (F. Ochoa)
5345 329 6,2% 345 6,5% 0 0,0%
Fabio Cedeño G. (San Antonio ) 1600 57 3,6% 50 3,1% 0 0,0%
Marcelo Méndez Morales 1.870 75 4,0% 25 1,3% 0 0,0%
Daniel Jiménez Berrocal (El Guabo) 1.907 125 6,6% 30 1,6% 0 0,0%
Daniel Jiménez Berrocal (Alonso) 200 20 10,0% 0 0,0% 0 0,0%
Daniel Jiménez Berrocal (Slopes) 1.000 40 4,0% 20 2,0% 0 0,0%
Olman Quesada Campos 650 26 4,0% 5 0,8% 0 0,0%
Daniel Jiménez Berrocal (Lote Pilo) 950 95 10,0% 4 0,4% 0 0,0%
Daniel Jiménez Berrocal ( Bismark) 1.280 78 6,1% 8 0,6% 0 0,0%
William Cortés Madrigal 1.460 150 10,3% 25 1,7% 0 0,0%
German Díaz Ruiz 4.095 274 6,7% 100 2,4% 0 0,0%
Edgar Salazar Ramírez 86 0 0,0% 5 5,8% 0 0,0%
TOTAL 26.575 1.625 6,6% 734 2,8% 0 0,0%
129
273 Phase 2 ( 24,134 TREES PLANTED )
NAME OFTHE PARTY TO THE N TREES MORTALITY IN MORTALITY IN
2012 2013 MORTALITY IN 2014
AGREEMENT PLANTED N % N % N %
Fabio Cedeño González (Ochoa) 3.100 150 4,8%
Melis Góngora Moraga 220 5 2,3%
Tito Hernández Ferreto 570 55 9,6%
Edwin Segura Retana 2.610 270 10,3%
Marcelo Méndez Morales 1.345 255 19,0%
Eylin Cruz Campos 3.550 40 1,1%
Frits Perera Jiménez (Palo Seco) 2.500 50 2,0%
Porfirio Rodríguez Campos 920 150 16,3%
Fabio Vargas Vargas (Chachalaca) 4.050 250 6,2%
Daniel Jiménez Berrocal (El Almendro) 2.463 0 0,0%
256 0 0,0%
Daniel Jiménez Berrocal (La Laguna)
Frits Perera Jiménez (Pindongo) 2.550 0 0,0%
TOTAL 24,134 1.225 6.0%
7.12. The main causes of mortality in the first year were: dry conditions (several
rainless days immediately after planting) , the volunteers ’ initial lack of
experience in planting techniques, local water-logging around some of the
newly planted trees, grazing by ruminants and equines gaining access to
the planted area, and application of inappropriate herbicides for weed
control at some sites . These issues were addressed as the project
proceeded and mortality was much lower in the second year .
Consequently, in the survey at the end of the phase 1 , no dead trees were
found in the planted areas. The outcome is expected to be similar for the
trees planted in phase 2, which has yet to be completed.
7.13. CODEFORSA are also engaged in a coordinated programme of slope
stabilization performed in conjunction with CONAVI. In this programme,
CONAVI address potential slope instability with respect to deep -seated
130
274 landsliding (which is the mechanism highlighted by Dr Kondolf in his 2014
report). Measures to prevent deep -seated landslides include slope
terracing, re-profiling and drainage improvements. These are described in
detail in the 2014 CONAVI Reportand, in summary, below in section 7B.
7.14. CODEFORSA undertake a range of related measures to protect slopes from
erosion by raindrop impact, manage drainage to prevent erosion by sheet
erosion and gullying, and improve stability with respec t to shallow
landslides. Together, the efforts of CONAVI and CODEFORSA constitute an
integrated programme of erosion mitigation.
7.15. CODEFORSA have so far completed mitigation work at multiple slopes in
the area around Tiricias, and more near Boca San Carlos, deploying
pocketed geofabrics and sowing vetiver grass (Chrysopogon zizanioides)
using the "adobe" approach, in which the survivability of the grass clumps
is vastly improved by providing them with a body of fertile soil around the
roots (Figure 7.1).
Figure 7.1 Examples of slope preparation and planting to provide surface protection by
CODEFORSA along the Road between Marker II and Boca San Carlos
7.16. As with the reforestation programme, slopes planted b y CODEFORSA are
monitored and maintained as necessary to ensure satisfac tory pl ant
131
275 survival. Figure (7.2 ) is an example taken from the 2014 CODEFORSA
Report, showing a sequence of photographs taken at slope 9, near Tiricias.
Figure 7.2 Slope 9 (near Tiricias) in February 2014, following treatment by CODEFORSA
for surface protection (example taken from 2014 CODEFORSA Report)
7.17. CODEFORSA’s current efforts centre on cut and fill slopes between the Río
Infiernito and the community of Chorreras, the area around Caño Cureñita,
where erosion had not been mitigated at the time that Dr Kondolf and his
team inspected the Road in preparation for writing their 2014 Reports.
Measures currently being implemented in that area by CODEFORSA
include:
(a) surface protection using geofabrics and revegetation using native
species of grasses and trees;
(b) digging small drainage channels to intercept runoff above, on and
below treated slopes;
132
276 (c) installation of cross-drains to manage runoff on the Road; and,
(d) construction of sediment traps to catch and retain eroded sediment
on site.
Implementation of these measures is scheduled for completion early in
2015.
7.18. CODEFORSA’s efforts do not end when implementation is completed,
however. CODEFORSA are contracted up to the end of September 2016 to
deliver a programme of monitoring and maintenance during which the
performance of all measures is appraised, any elements that fail are
replaced and adaptive management is practiced to ensure that erosion and
sedimentation are effectively mitigated at all treated sites. The
CODEFORSA Quarterly Report for November 2014 report describes and
illustrates these measures in detail, locating all slopes treated and
scheduled for treatment using their geographical coordinates and the
slope referencing system developed by Dr Mende.
7.19. Based on my discussions with CODEFORSA personnel and review of the
2014 CODEFORSA Report s, I conclude that s ubstantial efforts led by
experienced foresters and expert bioengineers have been made to reforest
over forty hectares of land and protect and revegetate the cut and fill
slopes along the Road between Marker II and Boca San Carlos.
CODEFORSA pay special attention to maintaining as well as implement ing
erosion mitigation measures and planting trees and grasses and I am
convinced that they are committed to ensuring the success of their efforts
in mitigating erosion and soil loss.
133
277 B. CONAVI’s Mitigation work at slopes and watercourse crossings
7.20. In December 2014, CONAVI delivered a repor t updating works performed
to mitigate erosion along the Road between Marker II and the community
of Chorreras, including the stretch of Road around Caño Cureñita (the
2014 CONAVI Report ). This report presents photographs of over forty
mitigation sites, ‘before’ and ‘after’ treatment that extend well beyond the
first 15 km downstream of Marker II. Specific measures undertaken
include:
(a) Surfacing the Road with gravel to stabilise and protect it from
surface erosion;
(b) Reprofiling and terracing steep slopes to stabilise those susceptible
to deep-seated landslides;
(c) Compaction and terracing and of loose fill slopes;
(d) Protection of cut anf fill slopes from surface erosion using coconut
fibre and hydroseeding;
(e) Clearing and safe disposal of slumped soil accumulated at the base of
slopes;
(f) Management of concentrated runoff using berms and concrete -lined
ditches, with energy dissipaters where necessary;
(g) Placement of silt fences and traps to intercept and retain eroded
sediment;
(h) Installation of culverts with concrete head and tail structures to
stabilise small watercourse crossings;
134
278 (i) Replacement of log bridges with modular bridges at larger
watercourse crossings.
Work is currently on-going, but is scheduled for completion early in 2015.
C. My Personal Observations in 2014
7.21. I visited the Road on three occasions in 2014. In this section I update the
over-arching impressions of the Road as it was in 2013 (which are
reported in the 2013 Thorne Report) . I also paid close attention to
mitigation works, ( some completed, others in progress), along the Road
between Marker II and Boca San Carlos.
7.22. While inspecting the works, I spoke to senior engineers from CONAVI
(http://www.conavi.go.cr/ ) who are in charge of mitigation and Meco
(http://constructorameco.com) who lead installation of the structures ,
and I took the opportunity to discuss lessons learned from implementation
of mitigation works performed in 2013 and plans for completion of
remaining mitigation works. I continue to be impressed by the way the
engineers appreciate the challenges posed by erosion control, their broad
experience, and their determination to complete the work necessary to
stabilise slopes, manage runoff and make good th e watercourse crossings
along the Road.
7.23. In his 2014 Report, the examples Dr Kondolf uses and the majority of the
photographs he includes in the body of the text are drawn from the 41 .6
km stretch of the Road between Marker II and Boca San Carlos. He pays
much less attention to the stretch between the mouths of the San Carlos
and Sarapiqu í (including just a handful of ‘Severely Eroding Sites’ and
photographs thereof) and ignores the stretch between Boca Sarapiqu í and
the Delta.
135
279 7.24. To explain why this is the case, it is only necessary to consider the
distribution of slopes and watercourse crossings provided in the 2014
Mende Report (Table 7.2).
Table 7.3 Distribution of slopes and watercourse crossings (from the Mende Report).
Stretch of Road Length Slopes Crossings
(km)
Nunber % Number %
Marker II - Boca San Carlos 41.4 126 6 77 6
3 0
Boca San Carlos - Boca 43.6 66 3 42 3
Sarapiqui 3 2
Boca Sarapiqui - Delta Costa 22.6 9 4 10 8
Rica
Totals 107.6 201 1 129 1
0 0
0 0
7.25. As I stated in my 2013 Report, I understand why Dr Kondolf focuses on
short, selected stretches of the Road and emphases erosion at sites he
describes as ‘severely eroding’ within those stretches . The fact remains
that the sites and stretches Dr Kondolf selects are atypical and Dr
Kondolf’s coverage gives an unrepresentative impression of both the
propensity for, and extent of , road-related erosion at cut and fill slopes ,
and likewise the potential for road-derived sediment to be delivered to the
Río San Juan via streams and ditches crossed by the Road.
7.26. My impression of the Road gained in 2014 is not that erosion has “visibly
worsened” (as Dr Kondolf states on page 11 of his 2014 Report) but, on the
contrary, that it has slowed. This is partly due to the natural recovery of
stability that follows disturbance of a landscape: the geomorphic ‘rate law’
which predicts that rates of change decrease exponentially with time since
disturbance (Graf, 19 77), but is also thanks to the c oncerted efforts of
CONAVI and CODEFORSA in mitigating erosion at multiple sites, including
those between the Río Infiernito and Boca San Carlos and especially those
east of the Río Infiernito (which have also been mitigated since Dr Kondolf
136
280 wrote his 2014 Report). For these reasons, Dr Kondolf’s equating of trends
in erosion along the Road with those reported by Ramos -Scharron and
McDonald (2005) on the island of St John (on page 44 of his 2014 Report),
is inapt.
7.27. In 2014 I visited several of the sites reforested and slopes revegetated by
CODEFORSA and I inspected many of the crossings and slopes where
erosion has been mitigated by CONAVI working in cooperation with
CODEFORSA. Summary statistics provided in the 2014 Mende Report
indicate that considerable progress has been made in mitigating erosion at
both slopes and crossings.
7.28. Work ongoing at Las Crucitas, La Chorera and El Jardín is now making real
progress and the situation is far better than when Dr Kondolf and his team
took the photographs included in the 2014 expert reports. D ata listed in
Table 7.3 indicate that this progress is not restricted to those locations:
mitigation is already complete, underway or unnecessary at over 70% of
slopes and is scheduled at the remaining 30% of sites.
Table 7.4 Mitigation status of slopes (from the 2014 Mende Report).
Slopes
Mitigation Status
(number) (%)
Mitigated 25 12
Mitigation in progress 107 53
Mitigation scheduled 58 29
No mitigation necessary 11 6
Totals 201 100
Data listed in Table 7 .4 indicate that mitigation is complete, underway or
unnecessary at over two thirds of watercourse crossings and that work is
137
281 scheduled for the remaining quarter of crossings where mitigation is
required.
Table 7.5 Mitigation status of watercourse crossings (from the 2014 Mende Report).
Watercourses
Mitigation Status
(number) (%)
Mitigated 28 22
Mitigation in progress 23 18
Mitigation scheduled 31 24
No mitigation necessary 24 19
Other 21 17
Totals 127 100
7.29. To support these statements and statistics I include below:
(a) photographs of the Road between the Delta and Boca San Carlos
taken during an overflight of on November 17, 2014 (Figures 7.3 and
7.4);
(b) photographs at sites being reforested by CODEFORSA taken by the
author in 2014 (Figure 7.5);
(c) photographs of slopes being revegetated by CODEFORSA taken by
the author in 2014 (Figure 7.6); and
photographs taken by the author in November 2014 of erosion
mitigation works on slopes and crossings around Caño Cureñita
being undertaken by CONAVI with assistance from CODEFORSA
(Figure 7.7).
138
282Figure 7. 3 Photographs representative of conditions along the Road between Boca
Sarapiquí and the Delta observed from the air by the author on 17 November, 2014.
139
283 Figure 7.4 Photographs representative of conditions along the Road between Boca San
Carlos and Boca Sarapiquí observed from the air by the author on 17 November 2014
140
284Figure 7.5 Condition of some of the CODEFORSA reforestation sites visited in 2014.
Photographs by author.
141
285 Figure 7.6 Condition of some of the slopes revegetated by CODEFORSA that were visited
in 2014. Photographs by the author.
142
286Figure 7.7 Condition of additional CONAVI erosion mitigation works inspected by the
author in 2014: left watercourse crossings, and right slopes. Photographs by author.
143
287 Figure 7.8 Large scale mitigation works by CONAVI on -going at sites around and Caño
Cureñita, inspected in November 2014. Photographs by author.
7.30. During my overflight on 17 November 2014, I observed sediment entering
the Río San Juan in appreciable quantities at five separate landslides along
144
288 the north (Nicaragua) bank (for example, see Figur e 7.9). The existence of
these natural features demonstrates how sediment delivered to the River
generates its high sediment load and turbidity, especially during the rainy
season.
Figure 7.9 Natural landslide at the north (Nicaragua) bank of the R ío San Juan observed
from Costa Rican airspace at coordinates W 084o 03’ 58.5’’ N 10 o 45’ 31.5’’ on 17
November, 2014. Note the temporary delta formed by sediment and fallen trees delivered
directly to the river by the landslide. Sediment and trees enter the river due to natural
processes to give Río San Juan naturally high sediment and debris loads and high turbidity,
especially during the rainy season. Four other similar landslides were also observed
during a single overflight that day. Photograph by author.
7.31. In August 2014, I took the opportunity to drive along some of the access
roads linking Route 1856 to the wider road network to the south. Most of
these roads pre- existed construction of Route 1856 and were practically
unchanged, other stretches were new or had been improved (Figure 7.10).
7.32. Bearing in mind the stable condition of the access roads, their remoteness
from the River and the scarcity of streams linking them to the River, in my
opinion it is highly unlikely that s ediment from these access roads reaches
the Río San Juan in any appreciable quantities.
145
289 Figure 7.10 Typical views of a ccess roads traversed during the field visit on 29 August
2014. Photographs by author.
7.33. In 2014, I revisited five mitigation sites that featured in the 2013 Thorne
Report. The first site was near Marker II, where the Road approaches the
Río San Juan from the west. In February 2013 the Road corridor featured
extensive areas of bare soil and a developing gully along the inboard edge
(Figure 7.11a ). In May the area had been transformed by recently
completed erosion mitigation measures including a concrete- lined, in-
board ditch to convey water draining off a relatively steeply sloping stretch
of t he road while preventing concentrated flow erosion, and coconut
matting to protect the bare soil areas from raindrop, sheet and rill erosion,
while allowing it to re-vegetate naturally (Figure 7.11b). On 23 April 2014,
all mitigation measures had survived the rainy season intact and by 29
August 2014 vegetation was well established along both margins of the
road, colonising the areas protected using coconut matting (Figures 7.11c
and d). I detected no visible erosion of the gravel road surface at this
location.
146
290 (a) (b)
(c) (d)
Figure 7.11 The Road near Marker II (a) prior to mitigation work on 15 February 2013 (b) on 7
May 2013 with mitigation measures in place: note in-board drainage channel and extensive
biodegradable, erosion control matting (c) on 23 April showing that all mitigation
measures survived the rainy season and (d) on 29 August showing that vegetation has
stabilized both margins of the road bed and was spreading across the areas protected by
coconut matting. Photographs by author.
7.34. On 15 February 2013 I observed a gully eroding into a fill prism located to
the west of Marker II (Figure 7.12a ). When I revisited the same site on 7
May 2013 a culvert had been installed to convey runoff from the micro -
basin beneath the road and a concrete- lined channel had been constructed
to carry it down the fill slope. The surrounding fill slope surface had been
protected from raindrop impact, sheet and rill erosion by extensive
deployment of coconut matting (Figure 7.12b). In 2014, visits in April and
August revealed that the culvert and drainage channel were performing as
intended and that vegetation had re- colonised much of the area protected
147
291 using coconut matting (Figure 7.12c and d). However, there was a sm all
gully beneath the matting at one point that needed attention and the final
section of the concrete channel had cracked due to uneven settlement and
required repair. The need for follow -up maintenance was brought to the
attention of CONAVI and maintenance was scheduled.
(a) (b)
(c) (d)
Figure 7.12 View down a large gully in a fill prism created by concentrated runoff from
the Road draining to Costa Rican territory to the west of Marker II (a) in February when it
was actively eroding and (b) in May when the gully had been back -filled and stabilized
using a culvert and concrete drainage channel, with coconut matting used to protect the
surrounding fill slope from sheet and rill erosion . Subsequent visits i n (c) April and (d)
August showed the culverted crossing to be intact after the rainy season and vegetation to
be recolonizing the surrounding area. Photographs by the author.
7.35. The third location where erosion had been noted during the February field
visit was about 6.4 km east of Marker II where runoff from a relatively
steep stretch of road had created two gullies on the out- board slope and
initial attempts at erosion control using geof abric had been unsuccessful.
148
292 Also, runoff was eroding the unlined in- board ditch. If left untreated, there
was a risk that scour in the inboard ditch might undercut the toe of a cut
slope at the top of hill and might trigger a landslide (Figure 7.13a ). During
the May visit I observed that extensive concrete drainage channels had
been constructed to convey both out- board and inboard runoff down the
steeply sl oping stretch of road (Figure 7.13b ). The channels were
functioning as intended and there had been no further toe erosion of the
cut slope, which appeared to be stable and unchanged from February. As
the photographs taken in April and August show, erosion at this site was
successfully mitigated in 2014, and vegetation is recolonizing both the fill
and cut slopes (Figure 7.13a and d).
(a)
(b)
(c) (d)
Figure 7.13 Road at East 497867, North 325463 about 6.4 km east of Marker II (a) on 15
February when failure of geotextile slope protection had allowed concentrated out-board
runoff from the Road to create two gullies and in -board runoff was undercutting a cut
slope (b) on 7 May 2013 after construction of concrete -lined out-board and in board
ditches (c) and (d) in 2014 erosion has been effectively mitigated, the gullies have healed
and both cut and fill slopes are revegetating. Photographs by author.
149
293 7.36. In February 2013, I observed a network of rills and gullies on an outboard
slope about 6.6 km east of Marker 2 (Figure 7.14a ). Soil eroded from the
gullies had accumulated on the terrace surface but on 15 February 2013 no
Road-related sediment appeared to have reached the Río San Juan. By 7
May 2013, there had been extensive mitigation work (Figure 7.14b). The
gullies had been replaced b y an engineered drainage system, silt fences
had been installed to prevent overland flow and a sediment trap had been
constructed to prevent sediment from reaching the river bank. On 23 April
2014, the area had largely revegetated and local erosion appeared to have
ceased (Figure 7.14c). On 29 August 2014, I observed that the gully, rill
and sheet erosion measures were still in place and functioning as intended.
The watercourse crossing had also been rep laced, with a larger culvert
installed lower in the channel and the loose fill prism replaced by
compacted soil-cement mixture to stabilize the crossing (Figure 7.14d).
150
294 (a) (b)
(c) (d)
Figure 7.14 Road at East 498072, North 325345, about 6.6 km east of Marker 2 (a) on 15
February 2013 showing a network of gullies on an outboard slope and sediment
accumulated as a run -out deposit on the flat terrace surface (b) on 7 May 2013 showing
mitigation works (concrete channels, drop structures, silt fences and sediment trap to
prevent sediment reaching the River (c) by April 2014 local erosion had ceased and the
slopes had revegetated and (d) in August the undersized culvert and fill prism beneath the
road at the watercourse crossing had been replaced by a larger culvert with head and exit
works, covered by a compacted soil-cement mixture. Photographs by author.
7.37. In February 2013, I observed evidence of sheet and rill erosion adjacent to
the road bed on a relatively steeply sloping stretch of the Road close to the
Río Infiernito (Figure 7.15a). Although the lower part of the rilled area was
re-vegetating naturally, I was concerned that this may not happen quickly
enough to stabilize the slope during the 2013 wet season. In the event, the
risk of serious future erosion at this site was reduced through a multi-
element, engineering sol ution designed to manage surface water runoff
from the road bed and adjacent disturbed slopes in an integrated manner
(Figure 7.15b ). In April 2014, vegetation was recolonizing previously
eroding slopes though silt fences were showing signs of distress (Figure
151
295 7.15c). In August 2014, revegetation was effectively complete and the
reforested area between the road and the river (not visible in the previous
photographs) was flourishing. No visible erosion of the gravel road could
be detected (Figure 7.15d).
(a) (b)
(c) (d)
Figure 7.15 Road at East 502480, North 321561, close to the Río Infiernito (a) on 15
February when surface unmanaged runoff from the road bed and surrounding slopes
disturbed during construction had caused sheet and rill erosion of bare soil surfaces. (b)
The same stretch of road on 7 May 2013 after protection of the road surface using crushed
rock, installation of silt fences to prevent sheet and rill erosion while directing down-slope
surface runoff into concrete -lined outboard and inboard ditches (c) in April silt fences
were showing wear and tear, but vegetation was spreading fast and by August (d) erosion
mitigation had been successful at this site. Photographs by author.
7.38. The final site between Marker II and Boca San Carlos visited on the ground
on 15 February was at Crucitas just east of the crossing on the Río
Infiernito. That was then as far as Route 1856 was accessible by
conventional 4-wheel drive vehicle. It was possible to observe the path
cleared in preparation for construction of the Road to the east. The
152
296exposed soil surface was subject to sheet and rill erosion (Figure 7.16a).
On 7 May 2013, erosion had been controlled using integrated runoff
management measures (Figure 7.16b ). Nothing had changed significantly
at this site when it was visited in April 2014 (Figure 7.16c ). In August the
site was photographed from th e air (Figure 7.16d ) as access by road was
not possible due to failure of the log bridge over the Río Infiernito . It was
observed that erosion mitigation continued to be successful and no
sediment from this site was entering the R ío San Juan. The Río Infi ernito
crossing site is scheduled for an engineered bridge in 2015.
153
297 (a) (b)
(c) (d)
Figure 7.16 Path cleared for the Road near Crucitas, just east of the R ío Infiernito (a) on
15 February when unmanaged runoff from the path cleared in preparation for
construction of the road bed had caused sheet and rill erosion (b) The same area on 7 May
2013 after installation integrated measures to manage runoff involving regrading, s ilt
fences, and concrete -lined outboard ditch (c) in April 2014, nothing had changed
significantly and in August 2014 it was apparent that erosion mitigation continued to be
successful and that no sediment from the site was reaching the Rí o San Juan. Photographs
by author.
154
298 8. Conclusions
8.1. Based on the scientific and technical studies reported in this document and
those to which I refer within it, I conclude that the Road has had no
significant impact on the hydrology of the Río San Juan.
8.2. The Road has had no significant impact on sediment transport in the R ío
San J uan because the quantity of additional sediment derived from the
Road is tiny compared to the heavy sediment load that was already being
carried by the R ío San Juan prior to construction of th e Road. Also, the
additional load from the Road is indiscernible due to high seasonal and
inter-annual variability in sediment supplies from other sources and
complexity in sediment transport processes.
8.3. In the reach upstream of Boca San Carlos the morphology of the Rí o San
Juan is insensitive to changes in the sediment load because its morphology
is controlled by bed rock rapids that fix the channel form, bed elevation,
long profile and slope. The morphological impacts of Road -derived
sediment are r estricted to deposition of coarse clasts on small sediment
deltas along the south bank that are concentrated in a short reach
upstream of Boca San Carlos . Morphologically similar, but generally larger
deltas also exist along the Nicaraguan side of the River. Downstream of
Boca San Carlos, the R ío San Juan is responsive to changes in sediment
supply, but the sediment regime is dominated by naturally high inputs ,
particularly from the mountainous, tectonically and volcanically-active San
Carlos and Sarapiquí basins.
8.4. There is absolutely nothing to suggest that the Road has adversely
impacted the water quality, fishes or fisheries in the River or the coastal
area around its outfalls . This is to be expected as expert opinion and
evidence from other rivers in the region suggest that fish species in the R ío
155
299 San Juan are well adapted to high and seasonally variable sediment loads
and concentrations.
8.5. Differences in communities of periphyton and macroinvertebrates
sampled in deltas along the banks of the R ío San Juan may be attributed to
contrasts in drainage area, vegetation and land use in micro -basins
draining to deltas along the north and south banks that were not
controlled for in the study by Dr Rios. Differences between the periphyton
and macroinvertebrate communities associated with large, friable clasts of
Road-derived sediment and those associated with older, rounded stream
gravels probably result from the short time that Road -derived sediment
has been in the fluvial system and cannot be taken as evidence of pollution
or habitat degradation.
8.6. There is no scientific justification for ‘ active efforts, including dredging, to
maintain the capacity and quantity of the river’s waters ’ in the lower Río
San Juan on the pretext of having to remove Road -derived sediment.
Coarse load and deposition calculations using an upper bound estimate of
the amount of Road -derived coarse sediment entering the lower Río San
Juan suggest that this is indiscernible compared to pre- existing coarse
load, especially when allowance is made for uncertainty concerning
estimation of the bedload carried by the River and the proportions in
which flow and sediment are divided when flow bifurcates at the Delta.
8.7. Sediment continuity d ictates that even if all of the coarse Road -derived
sediment supplied to the lower R ío San Juan in one year according to Dr
Andrew’s estimate (which I do not accept) were to be deposited on the bed
of the channel within the first three kilometres downstream of the Delta it
would, on average, raise the bed of the river by less than 5 to 10 mm.
156
3008.8. In fact, sand deposition is not restricted to the first three kilometres of the
lower Río San Juan, but is distributed along its entire length. This is evident
because:
(a) the lower Río San Juan has a mobile sand bed throughout its length;
(b) the micro-delta at the end of the lower R ío San Juan (30 kilometres
downstream of the Delta) continues to grow; and
(c) Nicaragua’s dredging programme includes more than 20 sites
downstream of the first three kilometres of the lower Río San Juan
where the bed has been dredged to remove accumulating sand (see
sketch map 5.1 on page 229 of Costa Rica’s M emorial in the Certain
Activities case).
8.9. Even according to Dr Kondolf’s over -estimate (which I do not accept) the
contribution of sediment f rom the Road is tiny (less than 3% of the mean
annual sediment load in Río San Juan) . Using the more reliable upper
bound estimates reported herein the contribution is probably less than 1%
of the mean annual load of the River. In either case, this contribution
would in practice be indiscernible, due to uncertainty and naturally
variability in quantity of sediment carried by the Río San Juan.
8.10. The Road has not caused harm to the hydrology, sediments, morphology,
environment, or ecology of the River, all of which are well -adapted to the
heavy load and highly variable sediment regime this River has experienced
for millennia due to its geology and climate.
8.11. Based on Professor Fallas’ expert meteorological review, and my
understanding of rainfall driven erosion, it is highly unlikely that the Road
will be catastrophically eroded in the event that a future Hurricane or
Tropical Storm affects Costa Rica. This is the case because the cyclonic
157
301 nature of the wind pattern associated with a Hurricane or Tropical Storm,
coupled with the existence of high mountains to the west and south of the
San Juan Basin mean that rainfall amounts and intensities in the area of the
Road would probably not be exceptionally high even should a Hurricane or
Tropical Storm affect Costa Rica. Dr Kondolf is incorrect to suggest that the
occurrence of a Hurricane or Tropical Storm would result in
unprecedented sediment loads and concentrations of sediment being input
from the Road to the Río San Juan.
8.12. The area of the basin disturbed construction of the Road is much smaller
than those disturbed by earthquakes and volcanic eruptions that are
known to have delivered exceptional quantities of sediment to the R ío San
Juan drainage system frequently during the last three centuries . For
example, the area of landslides documented in the immediate vicinity of
the Cinchona earthquake of December 2009 was nearly 22 km 2. While the
overall area disturbed during construction of the Road in 2011 was only
3.5 km .
8.13. Dr Andrews’ estimate of the natural load of the river is based on data from
tectonically-stable basins and it is therefore entirely inapplicable to the
San Juan Basin.
8.14. The exceptionally high sediment loads associated with seismic and
volcanic events do not cause harm to the hydrology, sediments,
morphology, environment, or eco logy of the River and the far smaller
contribution from the Road has no potential to do so whatsoever . For
example, according to Alvarado (2010), the Cinchona earthquake alone
supplied 2.5 to 3.5 million m 3of sediment (equivalent to 4 to 6 million
tonnes) to the Río San Juan drainage system. Even if Dr Kondolf’s over -
estimate of sediment delivery from the Road and its feeder roads
158
302 (116,000-150,000 m - equivalent to 194,000 to 250,000 t) were accepted
(and it is not), by his reckoning the contribution from the Road would still
be 3 to 6% of that from a single seismic event.
8.15. The concerted effort being made by CONAVI and CODEFORSA to mitigate
erosion along the Road is making progress and will continue. At the time of
writing the 2013 Thorne Report, mitigation had taken place around
Marker II and in the vicinity of Tiricias. In 2014, dozens more sites have
been mitigated, at locations along the length of the Road alongside the
River between Marker II and Delta Costa Rica that include several of those
that were classified earlier in 2014 by Dr Kondolf as ‘severely eroding’.
Planning for completion of Route 1856 is now well advanced and the Road
should be constructed to the highest standards and as quickly as possible
to provide a permanent solution to erosion issues along the Road.
159
303 9. References
Alvarado, G. 2000. Volcanes de Costa Rica: Editorial UNED, 269 p.
Alvarado G. 2010. Aspectos geohidrológicos y sedimentológicos de los flujos de lodo
asociados al terremoto de Chinchona (Mw 6,2) del 8 de enero del 2009, Costa Rica.
Revista Geológica de América Central, 43: 67 – 95.
Angulo, A., C.A. Garita-Alvarado, W.A. Bussing & M.I. López. 2013. Annotated checklist
of the freshwater fishes of continental and insular Costa Rica: additions and
nomenclatural revisions. Check list 9 (5): 987-1019.
(https://www.researchgate.net/publication/260752820_Annotated_checklist_…
freshwater_fishes_of_continental_and_insular_Costa_Rica_additions_and_no menclat
ural_revisions).
Bailey, R. C., R. H. Norris and T. B. Reynoldson. 2001. Taxonomic resolution of benthic
macroinvertebrate communities in bioassessment. J. N. Am. Benthol. Soc. 20: 280-286.
Barquero, R. and Rojas, W. 1994. Sismicidad inducida por el terremoto de Limón.
Revista Geológica de América Central. Volumen Especial. Terremoto de Limón, 22 de
abril de 1991, 111 – 120.
Bochini, I. and Montero, W. 1994. Sismicidad histórica e instrumental del Caribe de
Costa Rica. Revista Geológica de América Central. Volumen Especial. Terremoto de
Limón, 22 de abril de 1991, 65 – 72.
Bonada N., Prat N., Resh V.H. and Statzner B. 2006. Developments in aquatic insect
biomonitoring: a comparative analysis of recent approaches. Annual Review of
Entomology. doi:10.1146/annurev.ento.51.110104.151124
Bradbury, P.A. 1995. CALSITE version 3.1: User manual. United Kingdom: HR
Wallingford.
Bussing, W.A. 1998. Peces de las aguas continentales de Costa Rica. Universidad de
Costa Rica, San José, Costa Rica. 478 p.
Bussing, W.A., 2002. Peces de las aguas continentales de Costa Rica [Freshwater fishes
of Costa Rica]. 2nd Ed. San José Costa Rica: Editorial de la Universidad de Costa Rica.
468 p.
160
304Carlise, D. M., M. R. Meador, S. R. Moulton II and P. M. Ruhl. 2007. Estimation and
application of indicator values common macroinvertebrate genera and families of the
United States. Ecological Indicators.7 (1): 22-33.
Denya, P., Montero, W. and Alvarado, G. 2003. Atlas tectónico de Costa Rica. Editorial
de la Universidad de Costa Rica. Serie Reportes técnicos, 81 p.
Douglas, I. 1996. The impact of land-use changes, especially logging, shifting
cultivation, mining and urbanization on sediment yields in humid tropical Southeast
Asia: a review with special reference to Borneo. IAHS Publications-Series of
Proceedings and Reports-Intern Assoc Hydrological Sciences, 236, 463-472.
Dunne, T. 1979. Sediment yield and land use in tropical catchments. Journal of
Hydrology, 42(3), 281-300.
Flowers, R. W. 2009. A new species of Thraulodes (Ephemeroptera: Leptophlebiidae,
Atalophlebiinae) from a highly altered river in western Ecuador. Zootaxa 2052: 55-61.
García, M.H. 2007. Sediment transport and morphodynamics. In M.H. García (Ed.),
Sedimentation engineering: processes, measurements, modeling, and practice (ASCE
Manuals and Reports on Engineering Practice No. 110), (pp. 104 -105). American
Society of Civil Engineers, USA.
García, M.H. and Fernández, R. 2014. Actualización e implementación de metodologías
para estimar sedimento de fondo afluente a embalses de futuras plantas
hidroeléctricas. San José, Costa Rica: Instituto Costarricense de Electricidad ICE.
Gomez, B. and Church, M. 1989. An assessment of bed load sediment transport
formulae for gravel bed rivers. Water Resources Research, 25(6), 1161-1186.
Gómez-Delgado, F., Leitón-Montero, J.J. and Aguilar-Cabrera, C.A. 2013. Report on
hydrology and sediments for the Costa Rican river basins draining to the San Juan
River. San José: Costa Rica: Instituto Costarricense de Electricidad ICE.
Graf, W. L. 1977. The rate law in fluvial geomorphology. American Journal of Science,
277(2), 178-191.
Heino, J. 2014. Taxonomic surrogacy, numerical resolution and responses of stream
macroinvertebrate communities to ecological gradients: Are the inferences
transferable among regions? Ecological Indicators.36: 186-194.
161
305 Instituto Nicaragüense de Estudios Territoriales INETER. (2008). Estudio del suelo del
departamento de Río San Juan. Nicaragua. Retrieved from:
http://www.ineter.gob.ni/Ordenamiento/files/suelos_rio_san_juan.pdf
Jiménez, O., H.D. Farias and C. Rodríguez. 2005. Procesos de sedimentacion en
embalses en ambientes tropicales. Estudios de casos en Costa Rica y Republica
Dominicana. Revista Ingeniería del Agua 12 (3): 1-16.
(https://upcommons.upc.edu/revistes/bitstream/2099/2499/1/123article2.pdf).
Kjerfve, B. (1981). Tides of the Caribbean Sea. Journal of Geophysical Research: Oceans
(1978–2012), 86(C5), 4243-4247.
Krasovskaia, I. and Gottschalk, L. 2014. Compensatory flow phase 2:Methodology for
the determination of adaptive flow -Final report. San José, Costa Rica: Instituto
Costarricense de Electricidad ICE.
Montgomery, D. R. and Buffington, J. M. 1997. Channel -reach morphology in mountain
drainage basins. Geological Society of America Bulletin, 109(5), 596-611.
Peraldo, G. and Mora, M. 2008. Enseñanzas de la actividad histórica de los volcanes
Irazú y Turrialba, Costa Rica, América Central. En: García, V. (Coordinadora): Historia y
desastres en América Latina, III. Publicaciones de la Casa Chata: 115 – 164.
PROCUENCA-San Juan. 2004. Problemas relacionados con la degradación de los suelos
y la sedimentación. Informe tecnico. Disponible en internet.
(http://www.oas.org/sanjuan/spanish/documentos/adt/informacion/suelo.html ).
Resh, V. H. 2008. Which group is best? Attributes of different biological assemblages
used in freshwater biomonitoring programs. Environmental Monitoring and
Assessment, 138(1-3), 131-138.
Ruiz, P., Carr, M., Alvarado, G., Soto, G., Sáenz, L., and Feigenson, M. (2011) Coseismic
landslide inventory for the cinchona earthquake (01-08-2009, mw 6.2), Costa Rica. A
case of high landslide density at Poás volcano. Paper presented at the 2011 GSA
Annual Meeting, Minneapolis, Geological Society of America Abstracts with Programs,
Vol. 43, No. 5, p. 214.
Singh, V.P, Jain, S.K. and Tyagy, A.K. 2007. Risk and reliability analysis: a handbook for
civil and environmental engineers. United States of America: American Society of Civil
Engineers.
162
306Wischmeier, W. H. and D. D. Smith, 1960. A universal soil-loss equation to guide
conservation farm planning. Trans. 7th Int. Congress Soil Science (Belgium), pp. 418-
425.
Zweig, L. D. & C. F. Rabeni. 2001. Biomonitoring for deposited sediment using benthic
invertabrates: A test on 4 Missouri streams. Journal of the North American
Bentholoical Society. 20 (4): 643-657.
163
307 10. Statement of Independence and Truth
10.1. I affirm the statement of independent and truth given in my Report of
December 2014, which is Appendix A to Costa Rica’s Rejoinder.
10.2. The opinions I have expressed in this Report represent my true and
complete professional opinion. Where I have relied on the outputs of field
and analytical work performed under my supervision by the technical
team or facts supplied to me by those instructing me, I have noted this in
my Report.
10.3. I understand that my overriding duty is to the Court, both in preparing this
Report and in giving oral evidence, if required to give such evidence. I have
complied and will continue to comply with that duty.
10.4. I have set out in my Report what I understand from those instructing me to
be the questions in respect of which my opinion as an expert is required. I
have done my best, in preparing this Report, to be accurate and complete. I
have mentioned all matters that I regard as relevant to the opinions that I
have expressed. I consider that all the matters on which I have expressed
an opinion are within my field of expertise. I have drawn the attention of
the Court to all matters, of which I am aware, which might adversely affect
my opinion.
10.5. In preparing this Report, I am not aware of any conflict of interest actual or
potential which might impact upon my ability to provide an independent
expert opinion.
10.6. I confirm that I have not entered into any arrangement where the amount
or payment of my fees is in any way dependent on the outcome of this
proceeding.
164
308 10.7. In respect of matters referred to which are not within my personal
knowledge, I have indicated the source of such information.
10.8. I have not, without forming an independent view, included anything which
has been suggested to me by others, including the technical team and those
instructing me.
10.9. At the time of signing this Report I consider it to be complete and accurate
subject to any qualifications noted herein. I will notify those instructing me
if, for any reason, I subsequently consider that the Report requires any
material correction or qualification.
10.10. I understand that this Report will be the evidence that I will give, if
required, under oath, subject to any correction or qualification I may make
before swearing to its veracity.
10.11. The substance of all facts and instructions given to me which are material
to the opinions expressed in this Report or upon which those opinions are
based are reflected in my Report.
10.12. I confirm that I have made clear which facts and matters referred to in this
Report are within my own knowledge and which are not. Those that are
within my own knowledge I confirm to be true. The opinions I have
expressed represent my true and complete professional opinion.
………………………………………
Professor Colin Thorne
2 Parker Gardens,
Nottingham,NG9 8QG, UK February 2015
165
309310 Certification
I have the honour to certify that the documents annexed to this Rejoinder
are true copies and conform to the original documents and that the
translations into English made by Costa Rica are accurate translations.
Ambassador Sergio Ugalde
Co-Agent of Costa Rica
311312 List of annexes
Vol II: Technical and Environmental Reports
Vol III: Technical and Environmental Reports
Vol IV: National Legislation, Diplomatic Correspondence, Minutes of meetings,
Affidavits, Media Reports, Other Documents and Photographs
VOLUME II
Annex Document Page
No . No .
Technical and Environmental Reports
1. Professor Neil Craik, The Requirement to Perform a Prior 1
Environmental ImpactAssessment, January 2015
2. Professor Ian Cowx, Ecological Impacts of Route 1856 on the San 29
Juan River, Nicaragua, December 2014
3. Andreas Mende, Inventory of Slopes and Water Courses related to 71
o
the Border Road N 1856 between Mojón II and Delta Costa Rica:
Second Report, December 2014
VOLUME III
Annex Document Page
No . No .
4. University of Costa Rica, Centre for Research in Sustainable 1
Development, Department of Civil Engineering, Second Report on
Systematic Field monitoring of Erosion and Sediment Yield along
Route 1856, November 2014
313 Annex Document Page
No . No .
5. Instituto Costarricense de Electricidad (ICE), SBU Projects 45
and Associated Services, Centre for Basic Engineering Studies,
Department of Hydrology, Second Report on Hydrology and
Sediments for the Costa Rican River Basins draining to the San
Juan River, December 2014
6. Bernald Pacheco Chaves, Response to and Analysis of “Ecological 127
Impacts of the Route 1856 on the San Juan River, Nicaragua”,July
2014 (Ríos Touma 2014), October 2014
7. Arturo Angulo Sibaja,Environmental Diagnostic Assessment . Fish 141
Fauna in the San Juan River. Literature Review Report, November
2014
8. Pablo E. Gutiérrez Fonseca, Critical statistical analysis of the 161
report “Ecological Impacts of the Route 1856 on the San Juan
River, Nicaragua” by Blanca Ríos Touma, November 2014
9. Juan Carlos Fallas Sojo,Comments on the Report by Dr Kondolf as 175
it pertains to Hurricanes and Tropical Storms, 2014
10. Professor Allan Astorga Gättgens, Extraordinary sediment inputs 183
due to exceptional events on the San Juan River, December 2014
11. Consejo Nacional de Vialidad (CONAVI),Works on National Road 207
856: Before and After, December 2014
12. Comisión de Desarrollo Forestal de San Carlos (CODEFORSA), 253
Restoration and rehabilitation of ecosystems affected by the
construction of the Juan Rafael Mora Porras border road, Route
1856. Quaterly Report, November 2014
13. Comisión de Desarrollo Forestal de San Carlos (CODEFORSA), 323
Consulting Services for the Development and Implementation of an
Environmental Plan for the Juan Rafael Mora Porras Border Road ,
Report of Contract SINAC-CDE-004-2012, November 2014
14. Centro Científico Tropical (CCT) Follow-up and Monitoring Study 439
Route 1856 Project- EDA Ecological Component, January 2015
314 VOLUME IV
Annex Document Page
No . No .
National Legislation
15. Costa Rica, Executive Decree No. 24715-MOPT-MEIC-S, 1
6 October 1995, published in the Official Gazette number 207,
1 November 1995
Diplomatic Correspondence
16. Note from the Minister of Foreign Affairs and Worship of Costa 5
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-543-09, 27 July 2009
17. Note from theActing Minister of ForeignAffairs and Worship of 11
Costa Rica, to theActing Minister of ForeignAffairs of Nicaragua,
Reference DVM-176-09, 21August 2009
18. Note from the Minister of Foreign Affairs and Worship of Costa 15
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-674-09, 7 September 2009
19. Note from the Acting Minister of Foreign Affairs and Worship of 19
Costa Rica to theActing Minister of ForeignAffairs of Nicaragua,
Reference DM-264-11, 27April 2011
20. Note from the Minister of Foreign Affairs and Worship of Costa 25
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-161-13, 20 March 2013
21. Note from the Minister of Foreign Affairs and Worship of Costa 31
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-269-13, 21 May 2013
22. Note from the Permanent Representative of Costa Rica to the 35
United Nations-Geneva to the Secretary General of the Ramsar
Convention, Reference MPCR-ONUG/2014-324, 17 July 2013
23. Note from the Acting Minister of Foreign Affairs and Worship 39
of Costa Rica to the Minister of Foreign Affairs of Nicaragua,
Reference DM-D VM-550-2013, 24 September 2013
315 24. Note from theAgent of Nicaragua to tReegistrar of theInternational 43
Court of Justice, Request for Provisional Measures, HOL-EMB-196,
11 October 2013
25. Note from the Permanent Representative of Costa Rica to the 51
United Nations-Geneva to the Secretary General of the Ramsar
Convention, Reference MPCR-ONUG/2013/534, 25 November
2013
26. Note from the Secretary General of the Ramsar Convention to the 57
Permanent Representative of Costa Rica to the United Nations-
Geneva, 29 November 2013
27. Note from the Minister of Foreign Affairs of Nicaragua to the 63
Minister of ForeignAffairs and Worship of Costa Rica, Reference
DM-AM-685-13, 10 December 2013
28. Note from the Permanent Representative of Costa Rica to the 69
United Nations-Geneva to the Secretary General of the Ramsar
Convention, Reference MPCR-ONUG-2014-190, 26 March 2014
29. Note from the Secretary General of the Ramsar Convention to the 73
Permanent Representative of Costa Rica to the United Nations-
Geneva, Reference SG2014-103/CHB/MAR, 7 May 2014
30. Note from the Permanent Representative of Costa Rica to the 81
United Nations-Geneva to the Secretary General of the Ramsar
Convention, Reference MPCR-ONUG/2014/407, 18 June 2014
31. Note from the Minister of Foreign Affairs and Worship of Costa 93
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-0334-14, 11 July 2014
32. Note from the Acting Minister of Foreign Affairs and Worship 99
of Costa Rica to the Minister of Foreign Affairs of Nicaragua,
Reference DM-AM-348-14, 17 July 2014
33. Note from the Minister of Foreign and Worship of Costa Rica to the 103
Minister of ForeignAffairs of Nicaragua, Reference DM-0373-14,
24 July 2014
34. Note from the Minister of Foreign Affairs of Nicaragua to the 109
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM/336/8/14, 4August 2014
31635. Note from the Secretary General of the Ramsar Convention 115
to the Deputy Permanent Representative of Costa Rica to the
United Nations-Geneva, Reference SG2014-229-CHB-MAR,
18 August2014
36. Note from the Minister of Foreign Affairs of Nicaragua to the 119
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM-AJ/414/09/19, 19 September 2014
37. Note from the Minister of Foreign Affairs and Worship of Costa 127
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-574-14, 22 September 2014
38. Note from theAgent of Nicaragua to the Registrar of the International 135
Court of Justice, Reference HOL-EMB-124, 23 September 2014
39. Note from the Co-Agent of Costa Rica to the Registrar of 141
the International Court of Justice, reference ECRPB-103-14,
25 September 2014
40. Note from the Minister of Foreign Affairs and Worship of Costa 153
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-0639-10-14, 21 October 2014
41. Note from the Minister of Foreign Affairs of Nicaragua to the 157
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM/AJ/439/10/14, 27 October 2014
42. Note from the Acting Minister of Foreign Affairs and Worship 161
of Costa Rica to the Minister of Foreign Affairs of Nicaragua,
Reference DM-AM-0672-14, 28 October 2014
43. Note from the Minister of Foreign Affairs of Nicaragua to the 165
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM-AJ/448/11/14, 3 November 2014
44. Note from the Minister of Foreign Affairs of Nicaragua to the 173
Secretary General of the Ramsar Convention, Reference MRE/
DM-AJ/449/11/14, 3 November 2014
45. Note from the Minister of Foreign Affairs and Worship of Costa 177
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-0697-14, 5 November 2014
317 46. Note from the Minister of Foreign Affairs and Worship of Costa 183
Rica to the Secretary General of the Ramsar Convention, Reference
DM-AM-0706-14, 6 November 2014
47. Note from the Minister of Foreign Affairs and Worship of Costa 199
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-0707-14, 7 November 2014
48. Note from the Co-Agent of Costa Rica to the Registrar of the 205
International Court of Justice, Reference ECRPB-112-14,
10 November 2014
49. Note from the Minister of Foreign Affairs of Nicaragua to the 231
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM/DGAJST/456/11/14, 11 November 2014
50. Note from the Minister of Foreign Affairs and Worship of Costa 235
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-718-11-14, 14 November 2014
51. Note from the Minister of Foreign Affairs of Nicaragua to the 241
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM/677/12/14, 2 December 2014
52. Note from the Minister of Foreign Affairs and Worship of Costa 247
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-774-14, 2 December 2014
53. Note from the Acting Minister of Foreign Affairs and Worship 253
of Costa Rica to the Minister of Foreign Affairs of Nicaragua,
Reference DM-AM-789, 4 December 2014
54. Note from the Minister of Foreign Affairs of Nicaragua to the 259
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM-AJ/478/12/14, 5 December 2014
55. Note from the Minister of Foreign Affairs and Worship of Costa 265
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-0818-14, 12 December 2014
56. Note from the Minister of Foreign Affairs of Nicaragua to the 273
Minister of ForeignAffairs and Worship of Costa Rica, Reference
MRE/DM-AJ/482/12/14, 15 December 2014
31857. Note from the Acting Minister of Foreign Affairs and Worship 281
of Costa Rica to the Minister of Foreign Affairs of Nicaragua,
Reference DM-AM-0826-14, 16 December 2014
58. Note from the Minister of Foreign Affairs and Worship of Costa 287
Rica to the Minister of Foreign Affairs of Nicaragua, Reference
DM-AM-0832-14, 18 December 2014
Minutes
59. Press Release of 26 October 1976 and Minutes of the Meeting of 293
Liberia of 25 January 1977, in: Ministry of Foreign Affairs and
Worship of Costa Rica referring to the initiation of discussions of a
maritime boundary in the Pacific Ocean,Annual Report 1976-1977,
Vol. I, pp. 156-160
60. Minutes of the First Meeting of the Sub-Commission on Limits and 299
Cartography of 7 November 2002
61. National System of ConservationAreas, Tortuguero Conservation 307
Area, Log of the meeting held on the premises of the Nicaraguan
army post in the Delta to notify the entry by the San Juan River in
order to navigate to the disputed area declared by the International
Court of Justice, 17 December 2014
Affidavits
62. Affidavit of Mr.Victor JulioVargas Hernandez, recorded by Notary 315
Public, Mr. Gustavo Arguello Hidalgo, Deed no. 177-9, 17 July
2014
63. Affidavit of Mr. William Vargas Jimenez, recorded by Notary 321
Public, Mr. Gustavo Arguello Hidalgo, Deed no. 178-9, 21 July
2014
64. Affidavit of Ms. Mayela Vargas Arce, recorded by Notary Public, 327
Mr. GustavoArguello Hidalgo, Deed no. 179-9, 21 July 2014
65. Affidavit of Ms. Gabriela Vanessa Lopez Gomez, recorded by 333
Notary Public, Mr. Gustavo Arguello Hidalgo, Deed no. 189-9,
21 July 2014
319 66. Affidavit of Mr. Claudio Arce Rojas, recorded by Notary Public, 339
Mr. GustavoArguello Hidalgo, Deed no. 181-9, 21 July 2014
67. Affidavit of Mr. Ruben Francisco Valerio Arroyo, recorded by 345
Notary Public, Mr. Gustavo Arguello Hidalgo, Deed no. 194-9,
9 October 2014
Media Reports
68. La Na ción (Costa Rica), ‘Costa Ricans denounce mistreatment 351
and detentions in the northern border’, 3August 2014, available at
http://www.nacion.com/nacional/gobierno/Caos-frontera-provoca-
detenciones-costarricenses_0_1430656995.html
69. La Nacion (Costa Rica), ‘He demanded that I pull down my pants’, 359
3 August 2014, available at http://www.nacion.com/nacional/
gobierno/exigio-bajara-pantalones_0_1430657010.html
Other Documents
70. Department of Transit Engineering, Ministry of Public Works 365
and Transportation, Costa Rica, Authorization of Routes for the
Transport of Hazardous Materials, 1995
71. Note from the Chief of Post, Police Delegation of Sarapiquí, Costa 403
Rica, to the Regional Director of the Fourth Region-Heredia,
Reference 1571-2010-DPS, 27 September 2010
72. Manuel Coronel Kautz, Vice-Minister of Foreign Affairs of 411
Nicaragua and Designated Chairman of the Canal Authority of
Nicaragua. Grand Canal of Nicaragua Project, June 2012.
73. Ministry of ForeignAffairs and Worship of Costa Rica, New works 421
in the Northeastern Caribbean Wetland . Report to the Executive
Secretariat of the Ramsar Convention on Wetlands, July 2013.
74. Report by the Director General of the Organization for the 447
Prohibition of Chemical Weapons on the Status of Implementation
ofArticle VII of the Chemical Weapons Convention as at 31 July
2014; Additional Measures for States Parties that possess industrial
Facilities which are declarable under the Convention. Reference
EC-77.7, C-19/DG.8, 13 May 2014
32075. Report by the Director General of the Organization for the 465
Prohibition of Chemical Weapons on the Status of Implementation
ofArticle VII of the Chemical Weapons Convention as at 31 July
2014:Article VII- Initial Measures, Reference EC-77/DG.6, C-19/
DG.7, 13 May 2014.
76. Note from the Chief Engineer of the Department of Studies and 485
Designs of the Consejo Nacional de Vialidad (CONAVI) to the
Chief of the Department of Weights and Dimensions and to the
Director General of the Transit Police of Costa Rica, Reference
DGIT-ED-4697-2014, 11 June 2014
77. Internal Communication of the Costa Rican General Department 491
of Transit Engineering of the Ministry of Public Works and
Transportation, regarding the Authorization of Routes for the
Transport of Hazardous Materials, June 2014
78. Secretariat of the Ramsar Convention, Ramsar Advisory Mission 501
No. 77 Report, Wetland of International Importance Caribe Noreste,
Costa Rica,August 2014.
79. Instituto Costarricense de Electricidad, Colorado River, Gauging 525
Station 1104,Average daily flow table, 2010-2014
Photographs
80. Photographs of sediment deposit sites in Nicaraguan territory 531
321
Rejoinder of Costa Rica