Statement of Mr Marc Mangel (expert called by Australia)

Appendix 2: M Mange!, An Assessment of Japanese Whale Research Programs
Under Special PermitAntarctic (JARPA, JARPA II) as Programsfor
of Whales, April 2011cResearch in the Context of Conservation and Management

An Assessment of Japanese Whale
Research Programs Under Special

Permit in the Antarctic (JARPA,
JARPA II) as Programs for

Purposes of Scientific Research in the
Context of Conservation and

Management of Whales

MarcMangeLPhD,FAAASF , RSE

Universityof California

SantaCruz

334About the Author

Marc Mangel, PhD, is Distinguished Professor of Applied Mathematics and

Statistics and Jack Baskin Endowed Chair in Technology and Information

Management at the University of California Santa Cruz (UCSC). He also directs

the Center for Stock Assessment Research, a cooperative program between the

Fisheries Ecology Division, Southwest Fisheries Science Center, NOAA Fisheries

and UCSC to train students and post-doctoral scholars in the methods of

quantitative population biology required for sustainable fisheries. His honors

include Fellowship in the American Association for the Advancement of Science

and the Royal Society ofEdinburgh. Further biographie details are given in

AppendixA.

335 TABLEOFCONTENTS

1. Executive sutnmary ...................................................................... 337

2. Introduction ........................................................................
.......... 339

3. An overview ofwhaling in the antarctic ...................................... 340

4. Characteristics of a program for purposes of scientific research ..349

5. Description and assessment of JARPA and JARPA II
as programs for purposes of scientific research in the conext of
conservation and management ofwhales .......................................... 360

6. Conclusion ........................................................................
............ 376

7. Literature cited........................................................................
...... 379

8. Appendices ........................................................................
........... 385

3361. EXECUTIVE SUMMARY

1.1. In this Expert Opinion, I provide an assessment of the Japanese Whale

Research Program under Special Permit in the Antarctic (JARPA) and JARPA II
(the second phase) as programs for purposes of scientific research in the context
of conservation and management of whales based on generally accepted scientific
practice and criteria developed by the International Whaling Commission (IWC)

for Special Permit research.

1.2. I begin with an overview ofwhaling in the Southern Ocean, as it pertains
to the assessment of JARPA and JARPA II as programs for purposes of scientific
research in the context of conservation and management of whales. I explain how
the Revised Management Procedure (RMP) of the IWC is an advance in

management ideas that allows effective conservation and management of whales
without detailed biological knowledge and without use of lethally obtained data.

1.3. Consistent with generally accepted scientific practice and with criteria
identified by the Scientific Committee of the IWC, my opinion is that the essential
characteristics of a program for the purposes of scientific research in the context
of conservation and management of whales are that the program:

a) has defined and achievable objectives that aim to contribute knowledge

that is important to the conservation and management of whale stocks;
b) employs appropriate methods that are likely to achieve the stated

objectives, including:
(i) lethal methods only where the objectives of the research cannat be

achieved by any other means (for example, by the analysis of
existing data and/or the use of non-lethal research techniques);

(ii) setting sample sizes using accepted statistical methodology; and

(iii) linking mathematical models to data consistently;
c) includes periodic review of research proposais and results and adjustment

in response to such review; and
d) is designed to avoid adverse effects on the stocks being studied.

1.4. I then assess JARPA and JARPA II against those criteria and conclude that

they meet none of them.
1.5. First, JARPA II does not- and JARPA did not- have defined and

achievable objectives that aim to contribute knowledge that is important to the
conservation and management of whales. JARPA II has - and JARPA had - broad
and vague objectives that conflate exploration and exploitation. Their stated

337objectives could be used to justify almost any activity that Japan wished to pursue.
Their contribution to management remains undemonstrated after 24 years and the
potential of JARPA II to bring new knowledge about the conservation and

management of whales is very low, if it indeed exists at ali.
1.6. Second, JARPA II does not- and JARPA did not- employ appropriate

methods likely to achieve its stated objectives. Although a variety of empirical
methods are in princip le employed in JARPA II, a majority of effort is devoted to
lethal take despite the existence of problems with the data generated by that lethal

take and despite the existence of other, non-lethal, methods that can provide
nearly aliof the same information. The reasoning that underlies the setting of
sample sizes (the number of animais killed) and the distribution of sampling effort
is vague, unclear, and at times simply wrong. The links between the proposed

models of the ecosystem and the field worlc,particularly lethal take, are weak and
unclear.

1.7. Third, most ofthe work done in association with JARPA and JARPA II is
published outside of standard peer-reviewed literature. Only about 15% of the
published papers are peer-reviewed and potentially relevant to the stated

objectives. Workers in JARPA and JARPA II have not demonstrated an ability to
respond to criticism orto admit being wrong.

1.8. Fourth, there is no record of any attention being directed to avoiding
unintended adverse consequences in the design of JARPA or JARPA II; indeed
they proceed on the assumption that the take will have no effect on the stock.

1.9. My conclusion is that JARPA II is - and JARPA was - an activity for the

collection of data in the Southern Ocean. However, both have failed at turning
data into knowledge or in improving the conservation and management ofwhales.
JARPA II is not a program for purposes of scientific research in the context of
conservation and management of whales.

3382. INTRODUCTION

2.1. I have been asked by the Government of Australia to prepare an
independent report on the Second Phase of the Japanese Whale Research Program

under Special Permit in the Antarctic (JARPA II) and related matters. The full
terms of reference provided to me are found in Appendix B. Briefly, they are:

To identify and outline the essential characteristics oaprogram
undertaken for purposes of scientific research; and

To provide a critica/ analysis of the objectives, methodologies and other
features of JARPA Il and, in doing sa, assess whether JARPA Il has the

essential characteristics oa program undertaken for purposes of
scientific research.

2.2. I was given background material briefly described in Appendix C.

2.3. In order to meet the terms of reference, it is essential to understand the
characteristics of a program for purposes of scientific research in general and in
the specifie contextof conservation and management of whales. It is also

essential to understand the natureof JARPA II (and its predecessor JARPA), so
that they can be assessed as to whether they may properly be characterized as
programs for purposes of scientific research in the context of conservation and
management of whales.

2.4. In this paper, I

a) give a brief overview ofwhaling in the Antarctic, emphasizing the key
points that are relevant for the subsequent analysis;

b) identify the essential characteristicsf a program for purposes of scientific
research in general and in the specifie context of conservation and

management of whales;

c) provide an overview of the relevant aspects of JARPA and JARPA II, and
assess them against the essential characteristicsf a program for purposes
of scientific research in the context of conservation and management of
whales; and

d) conclude with a summary of that assessment.

3393. AN OVERVIEW OF WHALING IN THE

ANTARCTIC

3.1. Modern commercial whaling began early in the 20 century, using land-

based stations (Mackintosh 1965). The first Antarctic whaling station was
established at South Georgia in 1904. Whaling at the South Shetland and Orkney
Islands almost exclusively used factory ships, which were tankers fitted with a

factory plant and moored in a harbor to function as a floating land station. Land­
based whaling ran from about 1904 to 1928, after which the great era of pelagie
(at sea) whaling followed. By 1930/31, there were 41 pelagie factories with over
200 catching vessels working in the Antarctic.

3.2. However, during the 1930/31 Antarctic whaling season, more whale oil

was produced than the world market could absorb. Because of this, the whaling
companies agreed to limit their output and devised a plan to regulate catches by
the amount of oil produced. Since the species of major commercial take in the
early 20 hentury were the blue, fin, sei, and humpback whales (Mackintosh

1965), an effort was made to put them into a common currency. One blue whale
was considered the same as 2 fin, 2.5 humpback, or 6 sei whales; giving rise the
notion of the Blue Whale Unit (BWU) (Gambell 1999, Gillespie 2005).

3.3. In the figure below (from Bannister 2002), I show the relative sizes of

sorne of the whales.
3.4. The blue, fin, humpback and sei were called the great whales. At the time

that the BWU was conceived, minke whales were not considered relevant to
commercial whaling because oftheir small size. T0nnessen and Johnsen (1982)
noted that, had minke whales been considered, one BWU would have been at !east

30 and possibly up to 60 minke whales.

340 Bawhoad

Righi

Pygmy rtght"""Q ( ~
i.Human1
: _.

Stuc

1

Humpbac.k.

The International Whaling Commission

3.5. The inter-governmental International Whaling Commission (IWC)

(Gambel11999, Donovan 2002) is charged, among other things, with regulating
whaling in the Southern Ocean. The IWC was established in 1946 through the

International Convention for the Regulation of Whaling (ICRW). The ICRW
consists of two parts: the convention itself and a schedule of regulations intended

to govern whaling operations. Contracting Parties to the ICRW subscribe to:

a) safeguarding for future generations the great natural resources represented

by whale stocks;

b) protecting ali species ofwhales from further over-fishing;

c) seeking the optimum leve! ofwhale stocks;

d) providing an interval for recovery to certain species of whales now

341 depleted in numbers; and
e) establishing a system of international regulation for the whale fisheries to

ensure proper and effective conservation and development of whale stocks
(Gillespie pg 396-397).

3.6. The IWC has a Scientific Committee that meets annually, usually for two
weeks before the annual Commission meeting, and that often also holds ad hoc
meetings between sessions.

3.7. From its inception until about 1972, the IWC regulated whaling using the
BWU. The overall catch limit was initially set to 16,000 BWUs, with no

reference to specifie species except that sorne species (e.g. right whales,
humpback whales) were designated as protected. This was essentially an open
access fishery (as defined by Clark 2006), in which nations raced to catch as many
whales as possible before the quota was reached, leading to waste during

processing, an uneconomical increase in the number of catcher boats, and poor
conservation of the whales (Donovan 2002). Furthermore, the quotas were often
exceeded. By 1952 it was recognized that there were problems with this

management procedure, and in 1963 a small group of eminent scientists appointed
by the IWC recommended elimination ofthe BWU as a method ofsetting catch
limits (Clapham and Baker 2002). By 1971/72 the catch limit had been reduced to
2,300 BWUs and both blue and humpback whales had been protected from

commercial whaling.

Fundamentals of the Dynamics of Populations

3.8. I now briefly describe how the dynamics of populations are characterized
and sorne ofthe terminology used in the management and conservation ofwhales.

3.9. In population dynamics, a key focus is the change in population size from
one year to the next. This is called net production and is measured most simply as

the number of individuals in one year minus the number of individuals in the
previous year. Such net production usually depends upon current size of the
population and is typically a peaked function of current population size as shown
in the figure below:

342 ....
rn
~Vl

~c=
·.c;
u
"0:s
e
0..

MSYL K

Current Population Size

3.10. When there are no individuals, in the absence of immigration, net
production is 0 since without any individuals no new individuals can be produced.

3.11. When there are many individuals (indicated by K on the x-axis in the
figure) net production is also zero because competition for food causes a balance

between births and deaths. (The rates of birth and death, the latter commonly
called the rate ofmortality, are called the demographie (or biological) parameters

ofthe population.) When such balance between births and deaths is achieved, the
population is in astate known as carrying capacity. In the absence of
enviromnental fluctuations, this is the size at which the population would stabilize

if it were to remain unexploited.

3.12. The population size that maximizes net production is calied the Maximum
Sustainable Yield Leve! (MSYL) and the leve! of production associated with that
population size is called Maximum Sustainable Yield (MSY). Any catch that is

higher than the MSY indicated by the peak of the curve in Para 3.8 is not
sustainable, since more is being taken from the population than is being produced

by it. The MSY Rate (MSYR) is the ratio ofMSY to MSYL. For many years,
and continuing to this day in sorne cases, managing fisheries for MSY was a
standard approach.

3.13. Ricker (1975) defined MSY as "[t]he largest average catch or yield that

can continuously be taken from a stock under existing environmental conditions."
Ricker's definition hinges on three key words: average, continuously, and

343existing. That is, in nature there is not a single curve as shown in Para 3.8 but a
family of such curves, depending upon environmental (both physical and
biological) conditions. For example, the changing biomass of krill as water
temperature changes will affect the carrying capacity for whales (Wiedenmann et

al2008). As environmental conditions vary, the shape ofthe curve, location of
carrying capacity (K) and the value of MSY may also vary. Moreover, if one does
not know the curve precisely and one does not know the current population size
precisely, then one never knows that the catch is sustainable even assuming that

environmental conditions are constant. Consequently, uncertainty and
environmental variation make MSY a fragile concept for management purposes.

3.14. Larkin (1977) argued that MSY should be put to rest because, among other
things, it led to yields that were too high and unsustainable. He wrote "[w]hatever
lies ahead in the development of new concepts for harvesting the resources of the

world's fresh waters and oceans, it is certain that the concept of maximum
sustained yield will alone not be sufficient" (Larkin 1977, 10). That is, MSY
should be considered a constraint rather than a target since harvests greater than
MSY are not sustainable (Mange! et al 2002).

The New Management Procedure

3.15. After 1972, the IWC abandoned the BWU and in 1974 adopted a
realignment of its management procedures through the development of a New
Management Procedure (NMP). The NMP was designed to calculate catch limits

for whale populations using the fundamental principles of population dynamics as
described in Paras 3.8-3.14. The goals of the NMP were to bring each of the
whale stocks to the particular population leve! at which MSY could occur and to
protect stocks whose population sizes were estimated to be below a fixed fraction

oftheir pre-industrial exploitation leve! (Gambell 1999, Donovan 2002).

3.16. The NMP aimed to separate stocks into the three categories, based on the
extent to which the size of the stock varied from MSYL (Gillespie 2005):

a) initial management stocks (those considered to be above the size
generating MSY and which could thus be harvested clownto that leve!);

b) sustained management stocks (which were close to the size generating
MSY and would be maintained there); and

c) protection stocks (those that would not be harvested).

3.17. However, over time it became clear that the NMP had serious problems
(Cooke, 1995; de la Mare 1986abc, Holt 2004). The NMP was based on MSY,
although at the time it was proposed, the data required to calculate MSY were
Iacking. Thus, two ad hoc rules were added. First, stocks that had been subject to

stable catches over considerable periods oftime would continue to be harvested as

344long as there was no evidence of a decline. Second, for stocks that had not been
subject to serious previous exploitation, catches would be limited to 5% of the

estimated size of the stock. This rule was precautionary, in the sense that before
harvest began, population estimates had to be obtained. However, the NMP did
not deal with the question of how to incorporate the uncertainty in the estimates of
population size. Indeed, one of the failings of the NMP was that it did not

stipulate how existing data were to be used to assess the state of the stock and it
could not handle uncertainty regarding the status of the stock in a robust manner
(Cooke 1995, pg 652).

3.18. Cooke (1995, pg 648) noted

The main difficulty in operating the NMP was that there were
insufficient data for its implementation. For most stocks there was
no reliable estimate of population size, let alone an estimate of the

MSY or the relation between the current population and the MSY
level. Furthermore there was no particular incentive to collect
data. Even if relatively good data had been available, there would

still have been considerable uncertainty about the state of whale
stocks with respect to the NMP criteria, but there were no
guidelines asto how to cope with these uncertainties. Finally, the
'behaviour' of the procedure was unknown. By this is meant the

expected long-term consequences of applying the procedure to
whale stocks.

Indeed, it was still possible for whale stocks to be depleted even if the NMP were
followed precisely and the population dynamics of the whales matched those
assumed in the NMP because of the uncertainty associated with estimates of

population size (Cooke 1995).

345 The 1982 Moratorium on Commercial Wha/ing

3.19. The IWC adopted the moratorium on commercial whaling in 1982, setting
catch limits for ali stocks at zero with effect from the 1986 coastal and the
1985/86 pelagie seasons. The commercial whaling moratorium remains in force
today.

3.20. One of the objectives of the decision to institute the moratorium was to
provide time for the IWC to establish its best estimate of population sizes together

with a suitable procedure to facilitate sustainable catch limits. A moratorium on
commercial whaling would also enable the IWC to develop mechanisms by which
the whaling industry could be effectively regulated to avoid the problems of the

past.

The Revised Management Procedure (RMP)

3.21. Because of the serious problems with the NMP, the IWC spent about a
decade developing a Revised Management Procedure (RMP). The specifie goals
of the RMP are:

a) to achieve stable catch limits, thus allowing the orderly development and

regulation of the whaling industry;
b) to manage acceptable risk and to ensure that a stock is not depleted to the

point where the risk of extinction is not negligible; and

c) to ensure the highest possible continuing yield from each whale stock.
3.22. In order to achieve the goals of the RMP, the IWC agreed that (IWC 1994,

1999):
a) commercial whaling would be permitted only for populations in areas and

seasons for which catch limits (calculated by its Scientific Committee and
approved by the Commission) were in force;

b) catches would reach a maximum leve! when a stock was at 72% of its
unexploited leve!; and

c) there would be no whaling on stocks that were below 54% oftheir
unexploited leve!.

3.23. The aim of the RMP is not to attempt to calculate MSY or any other
optimum leve!. Rather, it is intended to effectively manage whaling while dealing

with the inherent uncertainty in the Southern Ocean ecosystem. Holt (2004, pg
xii-xiii, italics added) described the RMP in this manner:

346 Although the RMP uses a population mode! for the estimation of
stock status and the calculation of catch limits, the mode! itself is
hugely simplified. ft does not attempt ta emulate the dynamics of

any real whale population, and, infact, does not even explicitly
include demographie parameters such as natural mortality rate.
Rather, the simple mode! is part of a freely invented algorithm that

has been shown, by simulations, to meet the targets efficiently and
to be robust to errors and such things as environmental changes
(Holt 2004, pg xii-xiii, emphasis added).

3.24. In much the same way as a good card player will compute the odds that an
opponent has a certain card, under the RMP statistical methods are used to

produce a probability distribution for the catch limit and the current population
size, which is measured as a fraction of the unexploited leve!. Catch limits are
computed using a Catch Limit Algorithm (CLA), which sets the catch limit to be
0 if the population abundance is estimated to be less than 54% of its unexploited

leve!. If population abundance is estimated to be more than 54% ofunexploited
leve!, then the catch limitis set at a specified fraction of the population above the
unexploited leve!.

3.25. The data used in the CLA comprise only:

a) total catch statistics based on previous whaling (past data); and

b) data obtained through sighting surveys in which ships follow a prescribed
track line and count the number of whales that are seen (current and future
data).

3.26. The RMP thus eliminates the use of data obtained from whaling-dependent
or other lethal-source data, which are often unreliable for purposes of

management because they represent non-random samples of the population.
Consistent with this, in 1995 the IWC adopted a Resolution (1995-9) that stated,
among other things, "that scientific research intended to assist the comprehensive
assessment ofwhale stocks and the implementation of the Revised Management

Procedure shall be undertaken by non-lethal means" (emphasis added).

3.27. The development of modern computational tools, particularly the capacity
to undertake extensive computer simulation, allowed thorough testing of the RMP
(Kirkwood 1992, Cooke 1995). That is, the RMP was tested using sets of pseudo­
data that had been generated by other, more complicated population models. The

point of this testing was to ask the question: "how effective is the RMP in setting
catch limits that maintain or restore populations to acceptable levels when various
demographie parameters are unknown, or when the structure of the actual
population dynamics differs from those assumed in the RMP?"

3473.28. The tests allowed assessments of the performance of the RMP with
incorrect assumptions about the dynamics of the stock, varying initial abundance,
bias in sighting surveys, different relationships between true abundance and catch

per unit effort (a common proxy for abundance), uncertain or inaccurate catch
histories, and/or rare episodic events (e.g. epidemies). The tests showed that the
RMP was robust to these variations, maintaining catch and preventing the

depletion of the population (Cooke 1995). Most importantly, the tests allowed the
IWC to conclude that the RMP functioned effectively without making specifie
assumptions about the population dynamics of whales and taking into account

possible errors in historie catch record.
3.29. In contrast, I know of no peer-reviewed published paper that demonstrates

fundamental flaws with the RMP that can only be corrected through field-based
programs that involve lethal take.

3.30. In conclusion, the RMP is an "advance in management ideas" (Holt 2004,
pg xiii) and consistent with other scientific work on the most effective level of

complexity for models used in management of living marine resources (Ludwig
and Walters 1985, Hilborn and Mangel1997). As a member ofthe Committee of
Scientific Advisors of the U.S. Marine Mammal Commission 1989-1996, I
observed but did not participate in the development of the RMP. Returning to it

now after a 15 year absence, I am able to assess it with a fresh viewpoint and
concur with Holt that it is indeed a substantial advance in management.

3.31. In summary

e The Southern Ocean ecosystem is characterized by uncertainty in many
dimensions including the dynamics of populations.

e The commercial whaling moratorium, effective since 1986, has allowed
the IWC to develop and test the Revised Management Procedure (RMP) as

an effective tool for the future management ofwhaling.

e TheRMP

o uses an intentionally simple model of population dynamics;
o is designed so that lethally obtained data are not required;

o is designed to encourage the collection of sighting information; and

o has been rigorously tested and found to be robust to variations

from its assumptions.

3484. CHARACTERISTICS OF A PROGRAM FOR

PURPOSES OF SCIENTIFIC RESEARCH

Science as a Process

4.1. The goal of science is to understand the natural world by providing a

framework to account for observations already taken and to make predictions of
new observations. This goal is achieved by putting new knowledge in the context
of existing knowledge, recognizing that even when there is progress the
conclusions are transient (thatis, subject to ongoing testing and revision) but the

methods are not. It is ongoing testing that is the basis for the self-correcting nature
of science. Without that self-correction, one cannat claim to be doing science.

4.2. Modern science is complex, and this has led to the 'cult of the expert'
(Jenkins 2004). However, much of the complexity can be understood without
extensive technical trainingif one focuses on the characteristics of science as a
process for converting data into knowledge. Jenkins (pg 6) wrote "[t]he essence of

science is not sorne nuggets of information about the natural world but rather an
ongoing process for gradually learning how the world works, with occasional
breakthroughs in the form of major discoveries. At any given time, the
understanding of a phenomenon is likely to be incomplete, with conflicting

explanations and evidence. Scientists have learned to tolerate such uncertainty and
even relish the challenges itoffers."

4.3. There are essentially two types of science: (i) textbook science (which
most people learn in school) and (ii) science as practiced by scientists, or 'frontier
science' (Pickett etal2007).

4.4. Textbook science is typically identified with the notion of 'the scientific
method', which involves:

a) devising alternative hypotheses;

b) devising an experiment (or severa! ofthem) with alternative possible
outcomes; each of which will, as nearly as possible, exclude one or more
ofthe hypotheses;

c) carrying out the experiment so asto get as clear a resultas possible; and

d) recycling the procedure, making sub-hypotheses or sequential hypotheses
to refine the possibilities that remain.

In textbook science, we repeatedly challenge a hypothesis with experiments, and
if the hypothesis stands up to repeated experiments, it is treated as if it were true.

3494.5. Textbook science is a simple, linear process; it is also a myth (Grinnell
2009, l.70). Science as practiced is more complicated than this and the path to
discovery is more convoluted. In the ecological sciences in particular, it is often

impossible to conduct experiments, but observation can substitute for experiment
(Mange! 2010). In consequence, scientists proceed by assembling many different
strands of evidence, which, if collected properly, can be woven into a strong and

intellectually sound fabric of conclusions.
4.6. Whether it is textbook or frontier, science does not consist of simply

accumulating data. Indeed, we now often face the problem of data 'poisoning' by
having too rouch data and too little understanding. Valiela (200 1, pg 11) noted
"[d]escription is not tantamount to understanding: descriptive data can not by

themselves furnish an explanation of the mechanisms behind the observations, nor
can they easily identify the processes that brought about the situation described.
Complicated descriptions can become goals in themselves and may delude us into
thinking progress has been made". Gopnik (2009, pg 71) noted "[a]ll seeing is

impregnated with thinking. If science were simply a bucket into which
descriptions fell, it would be a heap of facts. It is in the jump beyond, to a general
rule, a theory, even a vision, that science advances".

4.7. Science as practiced by scientists invariably involves weaving many
strands of data together to produce new knowledge. The way that this is done

depends upon the problem that is being studied, especially in complicated
ecological situations where experiments are difficult to impossible. Simply put,
the essence of science is to extract knowledge from data and, if one does not know
in advance how the data will be analyzed to extract such knowledge, one is not

ready to collect the data.

4.8. In accord with generally accepted principles of scientific practice (Valiela
2001, Jenkins 2004, Pickett et al 2007) a program for purposes of scientific
research:

a) Has an over-arching conceptual framework that leads to a set of focused
questions (hypotheses);

b) Employs the correct set of empirical tools to answer the questions
including setting sample sizes with sound statistical reasoning, and linking

mathematical models and data appropriately;
c) Has proper assessment through the community of scientists; and

d) Is designed to avoid unintended negative ecological consequences.

350An Over-Arching Conceptual Framework Leading to a Set of Focused
Questions (Hypotheses)

4.9. A program for purposes of scientific research requires an over-arching
conceptual framework. Without it, one simply does 'exploratory analyses' hoping
that something interesting will arise from random activity. This rarely works; the

Nobel-prize winning immunologist Peter Medawar once wrote "[n]o new
principle has emerged from a heap offacts". When people speak of"Newton's
Theory of Gravity" or "The Theory of Relativity" or "Darwin' s Theory of

Evolution by Natural Selection" they mean such overarching conceptual
frameworks.

4.1O. The conceptual framework provides a clearly specified context and
purpose and sets the ground for clearly defined and achievable objectives, but it
does not itself lead to a program of work. Rather, it inspires and frames the
investigation of particular questions and hypotheses.

4.11. To be testable, questions and hypotheses must be operationally defined­
that is, it must be possible to answer the question using existing empirical or

theoretical methods or there must be excellent prospect that new methods can be
developed to answer the question. For example, Valiela (2001, pg 6) noted "[i]t is
a waste of time, of course, to worry about the density of angels on any surface, let

alone the head of a pin, unless we have a working seraphometer available".
4.12. Any idea that cannot be operationally defined cannot be studied by

empirical science. Similarly, objectives that cannot be tested are not scientific and
thus not achievable. In many ecological settings, unlike textbook science,
hypotheses may not be mutually exclusive in that an observation clearly excludes
hypothesis "A" but not "B". However, even in this case if the hypotheses are

operationally defined it is possible to test them and assess the relative strength of
the hypotheses provided by the data (Hilborn and Mange! 1997, Wolf and Mange!
2008).

4.13. In the ecological sciences, it is often impossible to conduct experiments.
For example, it is not possible to undertake experimental manipulation when

attempting to understand the dynamics of populations of blue whales. That is,
there is no possibility to replicate an experiment, since there are so few
individuals, those individuals may actually constitute a single population, and the
time scale oftheir population dynamics is very long. Nevertheless, we are not

prevented from asking questions about blue whales and observation provides a
viable means by which to attempt to answer those questions (see for example,
Branch et al2004 on blue whales, Mangel2010 on Steller sea lions).

351The Correct Set of Too/s

4.14. Once a set of questions has been established, a program for purposes of
scientific research should focus on the important step of identifying the best tools
that will answer those questions as clearly and unambiguously as possible. These

tools should be selected following an evaluation oftheir effectiveness in achieving
the stated objectives.

Setting Sample Sizes

4.15. Setting the size of a sample of data to be taken in order to estimate an
unknown parameter so as to test a hypothesis depends on:

a) how accurately the parameter needs to be known (how close the average
value of the estimate is to the unknown parameter);

b) how precisely it needs to be known (how much variation surrounds the
estimate of the average value); and

c) what kinds of statistical assessments will be done with the data.

Formai statistical methodology provides procedures by which the sample size
required to obtain a specified confidence that we have in a particular conclusion
can be determined.

Uses Models Appropriately

4.16. Models have become a comerstone for extracting knowledge from data. A
mode! is a stylized description used in analyzing or explaining a phenomenon. A

mode! is not a hypothesis in itself. Models are rather tools used in the evaluation
of hypotheses. Models serve a number of purposes, one of which is to assist in
determining what needs to be measured and how accurate and precise the

measurement needs to be. Models that are purported to be linked to field
programs must be consistently and appropriately connected to the data from the
field program.

Proper Assessment through the Community of Scientists

4.17. Scientists form communities and networks that link to the past and provide
connectedness in the present. Grinnell (2009, 1.158) noted "[e]ach researcher or
group of researchers initiates work in the context of prevailing experiences and

beliefs- the starting point and justification for further action". That is,
individuals in a program for scientific purposes collaborate in a self-correcting
community. Even the greatest geniuses of science (Newton, Darwin, Einstein)

had networks and communities and made numerous corrections in their research
programs.

3524.18. Scientific debate and disagreement is good if it leads to questions that can
be resolved by reliable research. Views that are not debatable because they are

based on immutable assertions are not scientific since self-correction is not
possible. A community with a free exchange of ideas allows scientists to identify
occasions when they may be wrong in their research and affords them the
opportunity to change their minds. Indeed, delight in the unexpected is the

lifeblood of science: "[a]lmost alone in belief systems, science welcomes the
disturbingly new" (Raymo 1991, pg 179). Grinnell (2009, l.385) noted "[t]hey
[scientists] are open to the possibility ofbeing wrong". Responding to critical

comments and changing research paths is an essential part of the practice of
science. An individual who is not open to the possibility of being wrong cannat be
a scientist. Furthermore, the scientific community is obliged to expose
assumptions, whether they arise from within science or from society, and to

explore the implications of those assumptions as they affect the practice of science
(Pickett et al 2007).

4.19. That is, scientists belong to a community of independent thinkers
cooperating in a relatively free spirit so that a series of independent initiatives
becomes organized into joint achievement by "mutually adjusting themselves at

every successive stage to the situation created by ali the others are who acting
likewise" (Polanyi 1969, pg 51).

4.20. Individual scientists sit at the nexus of the world to be studied (in which
discovery is the objective) and the research community (in which credibility is the
metric ). The individual scientist investigates the world and when he or she

believes that a discovery is made, the process of conversion from discovery to
credibility begins (Grinnell 2009, l.83). The community of scientists is
responsible for the proper assessment and quality control of scientific ideas, in
which discovery becomes credibility, through the process of peer review.

4.21. Peer review is a key component for the assessment of the value of ideas
(Resnik 2011) and is essential because when the value of an idea is undermined it

must be rejected. Peer review both provides quality control on the leve! of
standards of scholarship and methodology for the scientific community and it
helps authors improve their research proposais and resulting manuscripts. Peer­

review also leads to the generation and establishment of scientific opinion
(Polanyi 1969), which is held not by a single individual, but by a collection of
individual scientists each ofwhom endorses the opinions of others. Of course,
scientific opinion can be wrong, but reliable science responds to valid criticism,

which is how science advances.

4.22. Peer-review requires a multi-dimensional approach for both proposais for
research and manuscripts describing the results of research (Polyani 1969). For
problems in applied ecology, at the minimum peer-review assesses:

353 a) plausibility of an idea;

b) scientific value of an idea, consisting of accuracy, intrinsic interest, and
importance;

c) originality of an idea, (which is often assessed by the degree of surprise
brought about by the idea); and

d) applicability of an idea, assessed by how the work can inform the
motivating applied question.

Plausibility and scientific value encourage conformity whereas originality
encourages creative thinking and dissent. Applicability ensures that the ideas and
the work contribute to solving the motivating applied problem.

4.23. The criteria in Para 4.22 can be melded into questions typically asked by
referees assessing proposais for research (Grinnell 2009, 1.332):

a) Is there is a question to be answered?

b) Can the research group answer it?

c) Will getting the answer will be worth the effort?

Unless the answers to ali ofthese questions are "yes", work should not begin.

4.24. For publications after work has already been done, the questions typically
asked by referees are (Grinnell2009, 1.715):

a) Are the techniques appropriate?

b) Could any scientist potentially have done the work?

c) Are the results interpreted in an appropriate fashion?

d) Are the studies reasonable in light of ideas previously accepted by the
community?

Unless the answers to ali of these questions are "yes", the article should not be
published.

4.25. Articles that are not peer-reviewed are considered to be 'grey literature'
and are given Jessweight than those that have survived the peer-review process.
As retractions in high profile journals show, peer review is not perfect, but it is

nevertheless an essential characteristic of the practice of science.
4.26. In summary, it is essential to a program for purposes of scientific research

that there be peer review from the outset of the research program (since a program
should not begin until it has been assessed as feasible through a matching of
methods and objectives); that there be peer review throughout the operation of the

program (since throughout its duration a program should respond to deviations
from objectives by adjusting methods or even abandoning the program in the face

354of inadequate progress); and that the program end with publication of results in
peer-reviewed literature (since it is through peer-reviewed publication that claims

of discovery are given scientific credibility).

ls Designed ta Avoid Unintended Negative Consequences
4.27. The history ofhuman interaction with the natural world is replete with

examples in which human interventions have led to unexpected and surprising
consequences. Sorne of the best examples include those involving the resistance
of bacteria to antibiotics and of insects and weeds to pesticides or herbicides

respectively.
4.28. Scientific research may have unintended consequences that increase the

chance that the population being studied will decline or possibly become extinct.
For example, Harrison et al (1991) concluded that their very study of a population
of butterflies in California may have lead to its extinction.

4.29. Thus a program for purposes of scientific research will be designed to
achieve a clearly identified outcome while avoiding unintended negative

consequences that will put the population or stock being studied at risk. This
should include the identification of potential problems before fieldwork begins
and monitoring of the risk of unintended negative consequences during empirical

work.

355/WC Criteria for Special Permit Whaling

4.30. The Scientific Committee of the IWC bas spent many years considering
how the broad concepts in the previous paragraphs apply to scientific research in
the context of conservation and management of whales. Their most recent
thinking is summarized in IWC (2009). According to IWC (2009), proposais for

Special Permit research are to be structured according to:
a) Objectives of the study (Paras 4.9-4.13; 4.39a);

b) Methods to address the objectives (Paras 4.14-4.16; 4.39b)

c) Assessment of potential effects of catches on the stocks invoIved (Paras
4.27-4.29; 4.39d);

d) A note on the provisions for co-operative research for both field and

analytical studies(Paras 4.17-4.26; 4.39c); and
e) A list of scientists proposed to be sent to intersessional review workshops

(Paras 4.17-4.26; 4.39c).
4.31. According to IWC (2009), the objectives of the study should:

a) Be quantified to the extent possible;

b) Be arranged in two or three categories (primary, secondary, ancillary);

c) Include a statement for each primary category regarding whether it
involves lethal sampling, non-lethal sampling, or both;

d) Include at least a brief statement of the value of each primary objective
assessed according to the ability toi) improve the conservation and

management of whales stocks; ii) improve the conservation and
management of other living marine resources in the ecosystem; and/or iii)
test hypotheses not directly related to the management of living marine
resources; and

e) Refer, particularly for d(i) and d(ii), to past recommendations of the

Scientific Committee, carrying out implementations or reviews of the
RMP, improved understanding of other high priority issues, or
recommendations of other inter-governmental agencies.

4.32. According to IWC (2009), the methods should include:

a) Field methods that describe the species studied, the number, time frame
and area; the sampling protocol for lethal aspects; and an explanation of
why non-lethal methods or analyses of past data are insufficient;

b) Laboratory methods;

356 c) Analytical methods, including when appropriate estimates of whether the
proposed sample sizes will be sufficient to provide accurate answers to the

questions being studied; and
d) A time frame with intermediary targets.

4.33. According to IWC (2009), the assessment ofpotential effects of the
proposed take on the stock should include:

a) A summary ofwhat is known concerning stock structure in the area

concerned;
b) An estimate of abundance of the species to be studied, including an

assessment of the level ofuncertainty of the estimates of abundance;

c) Submission of a simulation study on the effects of permitted takes on the
catch, taking into account uncertainty and projecting forward for the life of
the proposed permit, and into the future.

4.34. IWC scientists, like marine mammal biologists in general, understand that
sometimes lethal take can provide information that other means of study cannat
(Paras 4.31, 4.32). For example, although progress is being made, there are still

no effective non-lethal means of aging whales, so if age information is absolutely
required, then lethal take is also required.

4.35. Lethal take destroys the abject of study and thus eliminates the possibility
of future information gained from the animal that is killed. Thus, scientists must
ask how much information is gained using a lethal method relative to the

information gained using a non-lethal method. Consequently, before using lethal
take, one must carefully weigh the balance between the immediate information
produced by killing the individual animal and the loss of future information that

could be obtained were a non-lethal method used. In my opinion, only when the
balance is strongly in favor of the former should the lethal take be used. That is to
say, the information gained must be proportional to the impact resulting from the
Jossof the individual.

4.36. The Society for Marine Mammalogy, the only international professional
society of marine mammalogists, recently published guidelines for treatment of

marine mammals in field research in its official journal Marine Mammal Science.
These guidelines recognize that lethal take may sometimes be appropriate and
state that (Gales et al 2009, pg 736):

a) researchers should use alternative non-lethal procedures when they are
available and satisfy the objectivesof the research;

b) animais should be killed in the most humane and rapid method available;

c) any population or stock-scale impacts should be minimized through
prudent selection of animais (e.g., avoidance of reproductive females if

357 possible) and sample size; and

d) where possible on-going activities outside the research community (e.g.,
hunts, by-catch events, strandings) should be utilized as a source of
material for scientific studies of marine mammals.

4.37. The IWC criteria also recognize that when a scientific study is motivated
by an important applied problem such as the conservation and management of

whales, another crucial dimension is whether the knowledge extracted from the
data can be used to answer the motivating applied problem. If the work cannot
provide an answer to the motivating problem, it has failed in the key aspect of

scientific inquiry, even if it produces other data. That is, a program that is
motivated by an applied problem such as the conservation and management of
whales must contribute to knowledge that informs the motivating problem. It is

the responsibility of the proposers to demonstrate the objectives are both
achievable with the methods proposes and that the work will contribute to the
motivating applied problem.

358Assessment Criteria Used in This Report

4.38. Consistent with the criteria for generally accepted scientific research and
the IWC criteria described above, I now describe what I consider to be the

essential characteristics of a program for purposes of scientific research in the
context of the conservation and management of whales.

4.39. A program for the purposes of scientific research in the context of
conservation and management of whales:

a) has defined and achievable objectives that aim to contribute knowledge
that is important to the conservation and management of whale stocks;

b) uses appropriate methods that are likely to achieve the stated objectives,
including:

(i) lethal methods only where the objectives of the research cannot be
achieved by any other means (for example, by the analysis of

existing data and/or the use of non-lethal research techniques);
(ii) setting sample sizes using accepted statistical methodology; and

(iii) linking mathematical and statistical models to data consistently;

c) includes periodic review of research proposais and results and adjustment
in response to such review; and

d) is designed to avoid adverse effects on the stocks being studied.

3595. DESCRIPTION AND ASSESSMENT OF JARPA
AND JARPA Il AS PROGRAMS FOR PURPOSES OF

SCIENTIFIC RESEARCH IN THE CONEXT OF

CONSERVATION AND MANAGEMENT OF WHALES

5.1. It is now possible to provide an overview ofthe relevant aspects of
JARPA II together with an assessment ofthose aspects against the essential

characteristicsof a program for purposes of scientific research in the context of
conservation and management ofwhales. Although my primary focus is JARPA
II, several aspectsf the assessment are retrospective (e.g. peer review) and others
are prospective (the feasibility the research plan to achieve the goals).

Consequently, I consider both JARPA and JARPA II.
5.2. In brief, the analysisn this section leads to the following conclusions:

a) The objectives of JARPA II are broad and poorly defined, often based on
science by assertion in which statements are made as if they have been

demonstrated but they in fact have not, and are formulated in a way that
conflates exploration and exploitation.

b) Although a variety of empirical methods are used, the majority of effort in
JARPA II is directed toward lethal take, with sighting surveys
compromised because they are conducted in conjunction with lethal take.

The connection between JARPA II as a field activity and management
models such as the RMP has not been demonstrated, and the process for
setting sample sizesin JARPA II is not based on solid statistical reasoning
or analyses of the accuracy required to meet the stated objectives.

c) Individuals participating in JARPA II are disconnected from the self­
correcting community of scientists and have not demonstrated the ability

to revise or correct their work or methodologies,n particular by changing
their minds concerning lethal take. The majority of the worlcconducted in
association with JARPA and JARPA II is published outside the standard
peer-review process and much of the worlcthat is published in standard
peer-reviewed literature refers only to the physiology and biochemistryf

reproduction in whales, topics that are irrelevant to the stated objectives of
JARPA and JARPA II.

d) There is no record of any attention being directed to avoiding unintended
negative consequences in the design of JARPA II.

5.3. By reference to the conclusions in Para 5.2, the general practice of science,
and the IWC criteria for Special Permit Research, I conclude that JARPA II is not

360a program for purposes of scientific research in the context of conservation and
management ofwhales.

A program for purposes of scientific research in the context of
conservation and management of whales has defined and achievable
objectives that aim to contribute knowledge that is important to the
conservation and management of whales

Vague & general objectives

5.4. Neither the goals of JARPA, nor those of JARPA II, have been clearly

stated as defined and achievable objectives, nor as scientific questions or
hypotheses that will contribute knowledge important to the conservation and
management of whales.

5.5. In 1987 the objectivesof JARPA were summarized as:

e Objective 1: Estimation of biological [demographie] parameters to
improve the stock management of the Southern Hemisphere minke whale.

e Objective 2: Elucidation of the role ofwhales in the Antarctic marine

ecosystem.
5.6. In 1995-97 two additional objectives were added:

e Objective 3: Elucidation of the effect of environmental change on

cetaceans [whales and dolphins].

e Objective 4: Elucidation ofthe stock structure of the Southern Hemisphere
minke whales to improve stock management.

5.7. Objective 1 was relevant to theNMP but is not relevant to the RMP; in
addition, it was not achieved. Objectives 2, 3, and 4 are so broad that they can be
used to justify almost any activity. Objectives 3 and 4 were added with little or no

justification or connection to results that had previously been obtained under the
program at the date of their addition. A program for purposes of science research
will adjust its goals and objectives as informations obtained and analyzed, but

this needs to be done with clear justification and reference to results obtained to
date.

5.8. JARPA II continues the pattern established by JARPA ofhaving broad
objectives (IWC 2007a, pg 6):

e Objective 1: Monitoring ofthe Antarctic ecosystem.

e Objective 2: Modeling competition among whale species and developing
future management objectives.

361 0 Objective 3: Elucidation of temporal and spatial changes in stock structure.

0 Objective 4: Improving the management procedure for minke whale

stocks.

5.9. The objectives of JARPA II comprise a mixture of ecological monitoring
and modeling (Objectives 1 and 2), field work (Objective 3), and management
(Objective 4) with little,if any, intellectual connection. These objectives
demonstrate confusion between monitoring (which may be important iftied to

management, but cannot be considered research since there is no focused question
or hypothesis) and management on the one hand, and alleged scientific
investigation on the other.

5.10. Since lethal take without demonstrated scientific need is involved, the

objectives of JARPA II blur potential scientific exploration and resource
exploitation. Furthermore, as with JARPA, the objectives are so broad asto allow
almost any activity.

The 'krill surplus' hypothesis

5.11. As described above (Para 4.9ff), the application of an overarching
conceptual framework should lead to a set of focused questions or hypotheses to

be investigated since without clear questions the likelihood of developing new
knowledge is slight.

5.12. However, the only clearly identifiable hypothesis in JARPA or JARPA II
is the krill surplus hypothesis, according to which the over-harvesting of the great
whales lead to a krill surplus, which in turn lead to an abundance of minke whales.

For example, Tamura and Konishi (2009) wrote "[t]his rapid decreasing of large
baleen whales species provided the annual surplus of krill as muchas 150 million
tonnes (Laws, 1977ab). This surplus became available for other krill

predators... This phenomenon is called 'krill surplus from the depletion ofbaleen
whales"' (pg 23).

5.13. Among JARPA workers, the krill surplus hypothesis quickly went from
hypothesis (that is, an idea to be investigated and possibly rejected) to theorem
(that is, a demonstrated result whose truthfulness is known). For example in the

review of JARPA undertaken by the government of Japan, Murase et al (2006)
wrote '"[k]rill surplus' caused by intensive commercial harvesting of large
whales... has been central theorem ofthe Antarctic ecosystem study" (pg 1). In

describing the possible 'krillsurplus' as a 'centraltheorem', Murase et al suggest
that it has already been proven. This is simply not the case, as will be explained
below.

5.14. The Antarctic continues to be unveiled as an ecosystem of intriguing
complexity in which simple predictions fail (Karentz and Bosch 2001). The figure

362below (Mackintosh 1965) illustrates the network of interactions between predators
and prey (arrows going from prey to predator).

5.15. From this figure, it is clear that the krill surplus hypothesis as applied in
JARPA and JARPA II deals with only a small part of the entire Southern Ocean
ecosystem. In addition, neither JARPA nor JARPA II are capable oftesting it

(Paras 5.36-5.37).

Data col/ected and the RMP

5.16. The literature concerning JARPA and JARPA II contains a variety of

references to whaling policy that will be based on Maximum Sustainable Yield
(MSY), which, as described above, has effectively been discarded by the IWC
(Para 3.23ft). As noted, the RMP explicitly avoids the use of lethal-take data as a
means of estimating abundance, instead placing a strong emphasis on data

obtained by means of sighting surveys.

3635.17. JARPA and JARPA II provide no demonstration ofhow the fieldwork
undertaken in those programs would actually contribute to the analysis ofMSY,
MSYR, orto improving flaws in the RMP. In particular, JARPA II does not make

clear how the improvement of management procedures for minke whale stocks
can be considered scientific research, which might be appropriate if the RMP had
been shawn to be seriously flawed. However, neither JARPA nor JARPA II has
demonstrated the existence of serious problems with the RMP.

5.18. JARPA was not relevant to the RMP, which (unlike the NMP) deliberately

does not depend on accurate estimates of demographie parameters. In spite of this,
JARPA II continues along the same path as JARPA. In particular, the collection
of demographie parameters of whales by lethal take remains central in JARPA II,
but has no relevance to the RMP.

Ecosystem mode/

5.19. At the meeting of the IWC that followed the 2006 Intersessional

Workshop (IWC 2007b), "Japan re-iterated the goal of JARPA Il, i.e. to develop
an ecosystem modelleading to sustainable use through multi-species
management" (IWC 2007b, pg 41). Ecosystem-level models refer to conceptual,
mathematical, or statistical models that include many components of the

ecosystem, rather than a focus on a single species.

5.20. Although the development of ecosystem-level models is a foundation for
Ecosystem Based Fisheries Management (Mangel2010a), the contribution of
JARPA or JARPA II as field programs to ecosystem-level management models is
never made clear. Even though JARPA Il's objectives have changed, its practice

has not been altered so asto collect the type of data required for a far broader
ecological study (see Paras 5.36-5.37).

5.21. One of the justifications of JARPA and JARPA II is that they will provide
the scientific information thatis required for the resumption of commercial
whaling. Itis remarkable that the JARPA and JARPA II documents lack even the

beginnings of a bioeconomic mode! providing investigation about the required
biological and economie circumstances to make commercial whaling on minke
whales feasible, although it had been recognized long before JARPA began that a
bioeconomic madel would provide key insights into the future commercial

whaling ofminke whales (Lockyer 1976). Such models are clearly the province
of scientific research since they provide the biological, economie and
mathematical foundations for effective conservation and management as time­

dependent phenomena (Clark 2010).
5.22. In summary,

• It is difficult to impossible to clearly identify the hypotheses of either

JARPA or JARPA II;

364 e Both programs offer broad objectives that conflate science, management,

and exploitation;

e Their stated objectives could be used to justify almost any activity that
Japan wished to pursue.

e Their contribution to management remains undemonstrated and the
potential of JARPA II to bring new knowledge about the conservation and
management of whales is very low, if it indeed exists at ali.

In my opinion, JARPA II fails to meet the essential first characteristic of a

program for the purposes of scientific research in the context of conservation and
management ofwhales.

A program for the purposes of scientific research in the context of
conservation and management of whales employs appropria te
methods likely to achieve the stated objectivesf including (a) use of

lethal methods on/y where the objectives of the research cannot be
achieved by any other means (i.e. by the analysis of existing data
and/or the use of non-lethal research techniques); (b) setting sample

sizes using accepted statistical methodo/ogy; and (c) linking
mathematical and statistical models to data consistent/y

Appropriate empirical tools

5.23. Scientific research on whales in the Southern Ocean canuse a variety of
tools for empirical research including:

a) sighting surveys in which whales are counted from ships or aircraft
(including photo-identification of individuals);

b) lethal take;

c) DNA analyses based on biopsies;

d) biochemical analyses; and

e) satellite tagging.

Sightings surveys
5.24. Sightings surveys are a common feature ofresearch in ali global whale

populations and if conducted appropriately may be a useful empirical tool for
assessing the abundance and distribution ofwhales. Recent JARPA II cruises
sighted blue, fin, sei, minke, humpback, southern right, sperm, and southern

bottlenose whales (e.g. Ishikawa et al2008).

3655.25. Sighting surveys can provide information on population density (Burt and
Borchers 1997), movement (Bannister et al 1999, Rock et al2006), the

relationship between physical habitat and whale distribution (Kasamatsu et al
2000) and the relationship between the abundance of food (krill) and whale
distribution (Murase et al 2002).

5.26. As described above (Para 3.23ff), the RMP uses sighting surveys in arder
to estimate abundance and does not rely on lethally acquired information.
However, sorne of the sighting surveys in JARPA and JARPA II are compromised

because their methods involve bath counting whales and preparation for lethal
take.

Lethal take

5.27. In contrast to sighting surveys, lethal take is not a common feature of
research in ali global whale populations. While it cannat be excluded that there
may be situations in which lethal take may contribute to a program for purposes of

scientific research in the context of conservation and management of whales,
JARPA and JARPA II simply assert but do not demonstrate that lethal take is
required. In addition, lethal methodology is a disproportionate focus in JARPA
and JARPA II.

5.28. Japan sought to justify lethal take as a means of obtaining age estimates

that could then inform the rate ofnatural mortality (required for the NMP but not
the RMP), but, as noted in the final review of JARPA, the effort failed.

5.29. This is because there are significant problems with the lethally derived
data used for aging. Bar plugs of whales have a structure of alternating light and
dark bands. Thus, in principle the age of a whale can be determined by counting
the bands, muchas with tree rings (Morris 1972, Roe 1967, Lockyer 1974, de la

Mare 1985). However, the difficulties in the interpretation of growth layers make
ear plug growth layers only somewhat reliable indicators of age. Furthermore,
there are problems with reading the ear plugs at ali and often a large number of

the killed animais do not provide readable ear-plugs (Lockyer 2010).
5.30. As described in Para 4.14 a tool should only be selected for use after

evaluating its effectiveness in achieving the stated objectives. Japan conducted no
such evaluation. For ear plugs such an evaluation was done only after nearly 25
years of JARPA and JARPA II (Lockyer 2010) and ear plugs failed to provide
information about the age dependence of the rate ofnatural mortality. Whether

alternatives exist or not for aging, the approach of JARPA had demonstrably
failed, but JARPA II continues along this track.

366Other too/s

5.31. Other common tools used in the study of populations of whales include

a) DNA analysis based on biopsies;

b) biochemical analyses; and

c) satellite tagging.

5.32. In the last 20 years, DNA and other molecular technologies for population
studies have advanced enormously. A small sample of tissue now yields a
sufficient amount of DNA for many different types of analyses on population
structure, animal gender, inter-relatedness of individuals and other population

scale parameters (Waples and Gaggiotti 2006).

5.33. Similarly, it is now possible to measure the concentration of many
pollutants in whales by taking non-lethal biopsy samples (Kunito et al 2002) and
to assess reproductive status from hormone concentrations in the blubber of minke
whales (Mansour et al 2002). Awruch et al (2008) demonstrated that size at

maturity in a shark could be obtained using blood samples measuring hormones.
These papers suggest a promising avenue of investigation for a non-lethal method
of determining reproductive status in whales.

5.34. Over the last 20 years, tagging whales with radio transmitters and
associated technology for collecting and analyzing data has progressed rapidly

(Fedak 2004, Freitas et al 2008). Satellite tags have an antenna that protrudes
through the skin, so that the whale eventually rejects them much like a splinter is
ejected. Mate et al (2007) reviewed the advances in satellite tags. Currently, tags
last long enough to cover either leg of the annual migration or the whole feeding

season and within the decade willlikely last for multiple seasons. These
longevities are sufficient to answer the critical questions about stock structure
required to apply the RMP multi-stock rules.

5.35. In 1987, the Government of Japan (Japan 1987, pg 43) noted that "[i]f
mark [or tag] and mark recapture could be available both in the low latitude

(breeding ground) and the high latitude (feeding ground), this method [mark­
recapture] would certainly produce information with the highest accuracy ever
obtained by any other methods ever adopted in the past for ascertainment of stock
movement, migration, and identification". At the time that this was written, the

longevity oftags was only about 3 weeks. However, this gold standard of
methodology - called for by Japan nearly 25 years ago - can now be achieved.
That is, non-lethal means are nowa practicable way for determining stock
structure.

367Linking methods to objectives

5.36. Japan has suggested that JARPA and JARPA II can test the krill surplus

hypothesis. (Japan 2000, pg 1). However, neither JARPA nor JARPA II is
sufficiently broad or deep to be able to test the krill surplus hypothesis as a
scientific hypothesis. Indeed, it may be impossible to test the krill surplus
hypothesis at all (Ainley et al2007). Nicol et al (2007) observed

It [the krill surplus hypothesis] isjust difficult to support or refute
without appropriate long-term, systematically collected, dataset on

krill and its major predators. With a few notable exceptions, we
are not in a position to be able to indicate whether most of the
major krill consumers have globally increased or decreased as a

resultof the demise of the great whales, nor how these predators
might now be responding to the recovery of sorne of these whale
populations. Furthermore, we remain unable to estimate robustly
global krill consumption now or in the past; data which are

essential for examining the krill surplus hypothesis (pg 292).

5.37. That is, given the enormous difficulties and the many uncertainties
surrounding the krill surplus hypothesis, a program intended to investigate it
would need to begin with a broad focus on the interactions between different
predators of krill and krill consumption by all such predators (see image in Para

5.14). JARPA and JARPA II do not do this since their narrow focus is
purportedly on three (in practice, essentially just one) species ofwhale. Rather,
JARPA and JARPA II have used the krill surplus hypothesis to conflate research
and exploitation.

Setting samp/e sizes

5.38. In the case ofboth JARPA and JARPA II, sample size indicates the
number ofwhales to be killed. It is very difficult to understand the statistical basis

for setting the levelof lethal take in either JARPA or JARPA Il.

5.39. Barly in JARPA, Tanaka et al (1992) computed the sample size (lethal
take) associated with the error in an estimate of the rate of natural mortality. The
resulting curves (Tanaka et al 1992, Figure 6) were very flat. 1have reproduced
one panel below (the others are similar).

5.40. The x-axis in this figure is the sample size (the number ofwhales killed)

and the y-axis is a measure oferror in the estimate ofthe rate ofnatural mortality.
The different curves in this figure show the error in the estimate of the rate of
natural mortality plotted as a function of the sample size under various
assumptions about the accuracy of the data.

368 A T
~

._
0
WN ~ 1
cno
d i 8
L :~.

=··

200 400

Samplesize

5.41.The arrows denote the sample size that gives the minimum standard error.
Ail else being equal, these arrows would point to the sample size that one would
choose were one's goal to minimize the errorate.
5.42.However, the curves are very flat, suggesting that many fewer whales
could be taken with only a minimallossy. For example, using the
curve marked T=lO, note that taking 100 whales rather than 300 whales only
marginally decreases the accuracyimate ofnatural mortality rate,
suggesting that many fewer whales could be taken without compromising the
resulting analysis. Thatay, many fewer whales killed will produce virtually
the same leve! of accuracy.

5.43.Indeed, Tanaka et1992) themselves noted that takes in the range of
200-400 whales provided the same accuracy, but then stated that "[h]owever, in
the actual research, other factors should be taken into consideration" (pg 419) to
increase sample size. The other factors are not explained in the context of
objectives nor are they justified through any statistical considerations. There is no
analysis provided to show how either knowledge or management would be
improved by having the marginally improved estimate of the rate of natural
mortality associated with a take of 300 rather than 100 individuals.
5.44.Lack ofstatistical clarity continues. For example, in
responding to the discussion of the proposai for JARPA II by Childerhouse et al
(2006), Hatanaka et al (2006) wrote that catches "under JARPA II have been
calculated as the minimum required to obtain statistically significant data. Given
that the stocks sampled are abundant and,for humpback andfin whales,

369increasing rapidly, it is quite logical that the sample size is correspondingly
large" (italics added).

5.45. This conclusion is not logical at ali. According to generally accepted
scientific and statistical methodology, the determination of a sample size must be

grounded in statistical reasoning. Whether the stocks are sufficiently abundant
may affect the practicability oftaking a particular sample size, but it should in no
way affect the actual determination of the sample size.

5.46. I consider that the spatial distribution of the lethal take is also important.
The IWC has divided the Southern Ocean into six sectors for reporting catches

and other data. In the years between the 1963/64 season and the 1985/1986
season the vast majority of Japanese minke whaling take was in IWC Areas IV
and V [which are the areas closest to Japan], with very few whales taken from
Areas I and II [where much more fuel and time would be needed to operate]

(Ohsumi 1979).

5.47. The proposai for JARPA (Japan 1987) noted that "very little information
was obtained [from commercial activity] for Area I and Area II" (pg 8). One
might therefore expect the focus of a program for purposes of scientific research
to be on Areas I and Il, in order to gain more information about those regions, but

instead the focus in both JARPA and JARPA II is in Areas IV and V because it
"makes the research more efficient" (pg 8). That may be true if one measures
efficiency in tenns ofwhales killed per effort, but less so if efficiency is measured
in terms of new knowledge.

5.48. That is, because the effort in JARPA II is in regions in which Japan

traditionally whaled, JARPA Il is collecting data that in large part already exist
from commercial whaling (before JARPA) and JARPA itself. The potential
development of new knowledge in this situation is very low.

Linking mathematica/ models ta data

5.49. Ecosystem models are one of the objectives of JARPA Il, but the JARPA
II proposai (Japan 2005, pg 11) discusses modeling competition among whale
species and future management objectives with no reference to other components

ofthe ecosystem. The models used in Appendix 9 ofthe proposai for JARPA II
(pg 81-82) do not require the detailed information that JARPA II sets out to
collect.

5.50. Indeed, neither JARPA nor JARPA II offer explanation for the assertion
made under those programs that to obtain the necessary data for the models

requires lethal take, nordo they offer an explanation or indication asto how those
data are to be used in the models. After nearly 20 years of JARPA effort, Mori
and Butterworth (2006) offered a "first step towards modeling the krill-predator

370dynamics of the Antarctic ecosystem". Their model (pg 225ft) does not require
the data from lethal take that is purportedly essential under JARPA II.

5.51. In summary,

o Although a variety of empirical methods are in princip le employed in
JARPA II, a majority of effort is devoted to lethal take for which there are

other, non-lethal methods that can provide nearly all of the same
information.

o The lethal take data are not required for the RMP.

o There are problems with the lethally derived data and many animais are
killed without providing any useable data.

o Other tools (DNA and biochemical analyses from skin biopsies, satellite

tagging) can provide much the same information as that provided by lethal
take.

o Japan has not demonstrated that its objective of developing an ecosystem
model (Para 5.8) is attainable with the datait collects through lethal

research.

o JARPA II is insufficiently broad to test the krill surplus hypothesis, which
has been treated not as a hypothesis but as a proved theorem in most of
JARPA and ali of JARPA II.

o The reasoning that underlies the setting of sample sizes (the number of

animais killed) or the distribution of sampling effort is vague, unclear or
simply wrong at times.

In my opinion, JARPA II fails when measured against the second essential
characteristic of a program for purposes of scientific research in the context of

conservation and management of whales.

371A program for purposes of scientific research has periodic review of
research proposais and resu/ts and adjustment in response to those
reviews.

5.52. In the development of a program for the purposes of scientific research in
any applied context, the responsibility is on the proposers to demonstrate that the
objectives are important and attainable with the methods proposed and will
contribute to the applied problem. This should be done through peer review of

proposais and resulting papers. Although the proposais for JARPA and JARPA II
had sorne form ofreview within the Scientific Committee ofthe IWC, there is no
evidence that they went through rigorous and anonymous peer-review by experts

in the field or that the proposais were substantially changed in response to the
comments obtained in review.

5.53. Workers involved in JARPA began, and those in JARPA II continue with
and consistently defend the position that 'lethal take is required' (Ohsumi 1995)
with no demonstration of ability to change their minds or respond to feedback
when lethal take is discussed.

5.54. For example, in 1998 JARPA workers argued "Genetic analyses using

DNA can be conducted using biopsy sampling. However, the number of samples
required in studies on stock identification in the case of the southern minke whale
is large, and consideration of sampling collection should be taken into account"
(IWC 1998, pg 412). DNA technology has changed so much since 1998 that this

is no longer the case (Para 5.32) but there has been no change in the position of
workers in JARPA II to reflect this.

5.55. The vague justifications for setting sample sizes(Paras 5.39-5.48) and the
justification of lethal takes as a means of cost recovery (Ohsumi 1995) are
examples of assumptions and policies that come from outside the scientific
sphere. However, workers involved in JARPA and JARPA II have not exposed

these assumptions and policies.

5.56. In 2010, Japan submitted a list of the scientific contributions of JARPA
and JARPA II (and the north Pacifie equivalents) to the IWC (Japan 2010). This
list shows 195 IWC Scientific Committee and other meeting documents and 107
peer-reviewed journal publications listed for JARPA and JARPA II.

5.57. I divided the roughly one-third (107 of302) ofthe publications that were
peer-reviewed into categories of management (including genetic methods for

stock identification and humane killing, ecology (including environmental
toxicology), evolution and population genetics, and reproductive physiology or
lipid biochemistry. The papers on management and ecology are potentially

relevant to the objectives of JARPA and JARPA II, those in evolution Jess so, and

372those in reproductive physiology or biochemistry not even mentioned in the
objectives.

5.58. Of these peer-reviewed publications, slightly Jessthan half (51 of 107)
deal with management or eco!ogy. That is, only about one-sixth (one-half of one­

third, or around 15%) ofthe articles are peer-reviewed and potentially relevant to
the broadly stated objectives. Only about one-fourth of the papers in management
or ecology appear in the ecologicalliterature outside of IWC publications. In
short, 1/12 (one-quarter of one-third) of the publications dealing with whale

ecology and management have appeared in literature outside IWC publications.

5.59. Nearly 40% (39 of 107) of the peer-reviewed articles relate to reproductive
physiology, or lipid biochemistry, which could be viewed as representing an
opportunistic use of samples obtained because of the use oflethal methods.
However, it is not clear how the knowledge about in vitro fertilization of minke

whale eggs with previously frozen sperm, attempts to inject sperm into frozen and
then thawed eggs, or to mature minke whale eggs in vitro are even remotely
relevant to the objectives of JARPA or JARPA II as set out in the proposais for
those programs.

5.60. Japan has asserted (IWC 2007) "that for ethical reasons, many western

scientific journals refuse to accept papers based on lethal studies of whales". Since
there is no supporting information for this assertion, it is difficult to determine
how frequently such refusais occur and if they are based on the purported ethical
reasons or the objective quality of the submitted worlc

5.61. I note, however, that the IWC'sjournal The Journal ofCetacean Research

and Management and Marine Mammal Science (see Para 4.36) both publish
papers based on lethal studies, as long as the work is of sufficient quality and was
conducted legally. Sorne more general journals, such as Animal Behavior do not
consider work based on lethal studies, but others, for example Oecologia or Polar

Bio/ogy, do consider work based on lethal studies.

5.62. In summary,
• The review of proposais for JARPA and JARPA II has been weak and the

response to reviews even weaker.

• Workers in JARPA and JARPA II have not demonstrated an ability to
respond to criticism orto admit being wrong.

• Workers in JARPA and JARPA II have not exposed assumption and
policies that come from outside sources.

• Only about 15% ofthe papers produced by JARPA and JARPA II appear
in peer-reviewed literature and are relevant to the objectives as laid out in

the proposais.

373In my opinion, JARPA II fails when measured against the third essential
characteristicof a program for purposes of scientific research in the context of
conservation and management of whales.

A program for purposes of scientific research in the context of
conservation and management of whales is designed to avoid

adverse effects on the stocks being studied

5.63. Estimates ofthe number ofminke whales in the Southern Ocean have
fluctuated considerably and are still highly uncertain, but for the purposes of this
paper one may consider the estimate to be of the order of magnitude of 300,000-

500,000 individuals (Gambell 1999, Gillespie 2005). Minke whale takes from
1974 to 1984 were of the order of 5,000 per year, and takes during the eighteen
years of JARPA averaged just over 435 individuals per year. Under JARPA II, the

takes number around 550 animais per year.
5.64. It was assumed in JARPA, and continues to be assumed in JARPA II, that

lethal takes will have no effect on the dynamics of the stock (Nakamura 1991,
1993; Nakamura et al 1993). The danger is that this is a self-fulfilling prophecy:
when one analyzes data assuming that there is no effect of the catch on the

dynamics of the stock, then one will be forced to draw that conclusion since it is
built into the analysis itself. Put another way: the assumption of no effect of
research takes on populations is a preconceived conclusion.

5.65. Furthermore, ifthere are multiple local popÜlationswithin the sampling
area of JARPA II, the possibility exists that takes are unevenly distributed across
different local populations, which can lead to different impacts. While unlikely,

some depletion of small populations could be occurring and JARPA II would not
be able to measure or monitor such impacts.

5.66. Using the estimated overall population size may be misleading for another
reason. JARPA and JARPA II sample minke whale schools that are typically 1-4
individuals, containing a mixture of mature and immature individuals (Kato et al

1989, Fujise et al1993, Kasamatsu et al 1993, Nishiwaki et al2005). In
population biology there is a phenomenon known as the Allee effect (Courchamp
et al 2008, Mangel et al 2010) in which once the size of the population becomes
sufficiently small (for example through anthropogenic effects) the population

continues to decline, even if the original reason for the decline is removed. There
are many causes for of Allee effects (Courchamp et al2008), one ofwhich is the
disruption of social structure as would happen by removing individuals from

small schools. The importance of social structure in minke whale feeding schools
is still uncertain, but there is no mention in any ofthe JARPA or JARPA II
literature of Allee effects, and nor of any efforts made to confirm that the
populations under consideration do not show Allee effects.

3745.67. In summary,

" Japan has not shawn that JARPA II will not adversely affect the stocks,

instead, it simply assumes that this will be so;

e There may be a whole range of indirect effects on the populations that are
not even considered in JARPA Il.

e A well-designed program of research would recognize these possibilities
and check for them, even if the likelihood of an adverse effect on the

overall population were small.
Thus, 1consider that JARPA II is inconsistent with the fourth characteristic of a

progratn for purposes of scientific research in the context of conservation and
management ofwhales.

3756. CONCLUSION

6.1. A program for purposes of scientific research in the context of

conservation and management of whales must do much more than simply collect
data; the data must be capable offorming the basis of new knowledge. Indeed,
methods that generate the most data often do not generate the most knowledge.
JARPA is an example of an activity that collected data but which failed to

generate additional knowledge. On the basis of the materials I have reviewed, I
consider that JARPA II will continue as an activity for the collectionf data but,
similarly to JARPA, will contribute little new knowledge relevant to the
conservation and management of whales.

6.2. Scientific research work should begin with a question as opposed to an

answer, since retrofitting a problem to a solutions almost never a good approach.
Most importantly, the collection of data should never begin until one knows how
it will be analyzed and used. Both JARPA and JARPA II began with an answer
that lethal take is required and without clear planshow data were to be/or will

be analyzed or used.

Defined and achievable objectives that aim to contribute knowledge that is
important to the conservation and management of wha/es

6.3. In2005 the objectives of JARPA II were:

a) monitoring of the Antarctic ecosystem (including whales, krill and the
feeding ecology ofwhales, and the effects of contaminants of cetaceans,
monitoring of cetacean habitat);

b) modeling competition among whale species (including constructing a
model of competition among whale species and new management
objectives including the restorationf the cetacean ecosystem);

c) elucidation of temporal and spatial changes in stock structure; and

d) improving the management procedure for Antarctic minke whale stocks.

6.4. These objectives are based on considerable science by assertion,in which
claims are stated as if they were demonstrated through rigorous study but actually
are not.

6.5. The objectives of JARPA II are extremely broad and Jackfocus.

Experience with JARPA suggests that the broad and vague objectives of JARPA
II effectively allow any activity, and are used to provide justification for lethal
take.

3766.6. The RMP of the IWC provides a practical and well-tested approach for the
management of future Southern Ocean whaling. Itis an excellent compromise
between the complexity of the mode! and the availability of data, and is capable of
dealing with the high levels of uncertainty in the Southem Ocean ecosystem.

6.7. The data that are proposed for collection during JARPA II are not required

for the RMP and the information on stock mixing (which is today better collected
through combinations of satellite tagging and genetic analysis) will only
peripherally contribute to any reconsideration ofiWC regulations concerning
stocks. Thus, the potential applicability of JARPA II to the RMP is low, if it

exists at ail.
6.8. In my opinion, JARPA II fails to meet the first criterion to characterize it

as program for purposes of scientific research in the context of conservation and
management ofwhales.

Appropriate methods that are like/y ta achieve the stated objectives,
including:

i. lethal methods on/y where the objectives of the research cannat be
achieved by any other means (for example, by the ana/ysis of

existing data and/or the use of non-lethal research techniques);
ii. setting sample sizes using accepted statistica/ methodology; and

iii. finking mathematical and statistica/ models ta data consistent/y;

6.9. The methodology of JARPA II includes modeling, sighting surveys,
biopsies, and lethal take. However, the expressed requirement for lethal take is
science by assertion and the contribution of JARPA II as a field program to

management models is not demonstrated.

6.1 O. The mathematical models proposed in association with JARPA II are, to a
very large extent, independent of the field data collected in JARPA II, especially
the lethal data.

6.11. Sighting surveys, biopsies, and modeling are ali effective empirical tools
that are available to address the currently stated objectives. Lethal take is not
required to meet the objectives of JARPA II.

6.12. Even if lethal take were required, the process for setting sample sizes of

lethal take in JARPA II is not based on solid statistical reasoning or analyses of
the accuracy required to meet objectives.

6.13. In my opinion, JARPA II fails to meet the second criterion to characterize
it as program for purposes of scientific research in the context of the conservation
and management of whales.

377Periodic review of research proposais and results and adjustment in
response to such review.

6.14. Most of the work conducted in association with JARPA and JARPA II is
published outside the standard peer-review process. Much of the worlcthat is
published in standard peer-reviewed literature is on physiology and biochemistry

of reproduction in whales, topics irrelevant to the stated objectives of JARPA and
JARPA II. Only about 15% of the papers resulting from JARPA and JARPA II
are both peer-reviewed and relevant to stated objectives.

6.15. Scientists in JARPA and JARPA II have demonstrated an unwillingness to
change their minds, particularly with respect to the asserted requirement for lethal
take.

6.16. In my opinion, JARPA II fails to meet the third criterion characterizing a

program for purposes of scientific research in the context of conservation and
management ofwhales.

/s designed to avoid adverse effects on the stocks being studied

6.17. There is no record that JARPA II is designed with any attention directed to
avoiding unintended consequences.

6.18. In my opinion, JARPA II fails to meet the fourth criterion characterizing a
program for purposes of scientific research in the context of conservation and
management ofwhales.

Overa/1Assessment of JARPA Il

6.19. JARPA II is an activity that collects data in the Southern Ocean.
However, by reference to standard accepted practice of science and the IWC
Special Permit criteria, it is not a program for purposes of scientific research in the

context of conservation and management of whales.

3787. LITERATURE CITED

Ainley, D.G., Clarke, E.D., Arrigo, K., Fraser, W.R., Kato, A., Barton, K.J., and P.R.
Wilson. 2005. Decadal-scale changes in the climate and biota of the Pacifie sector of the
Southem Ocean, 1950s to the 1990s. Antarctic Science 17:171-182

Ainley, D., Ballard, G., Ackley, S., Blight, L.K., Eastman, J.T., Emslie, S.D., Lescroel, A.
Olmastroni, S., Townsend, S.E., Tynan, C.T., Wilson, P. andE. Woehler. 2007.

Paradigm !ost, oris tbp-down forcing no longer significant in the Antarctic marine
ecosystem? Antarctic Science 19:283-290

Awruch, C.A., Frusher, S.D., Pankhurst, N.W., and J.O. Stevens. 2008. Non-lethal
assessment of reproductive characteristics for management and conservation of sharks.
Marine Ecology Progress Series 355:277-285

Bannister, J.L.2002. Baleen whales. Pg 62-72 in Encyclopedia of Marine Mammals.

(W.F. Perin, B. Wursig, and J.G.M. Thewissen, editors). Academie Press, San Diego

Bannister, J.L., Pastene, L.A. and Burnell,S.R. 1999. First record ofmovement of a
southern right whale (Eubalaena australis) between warm water breeding grounds and
the Antarctic ocean, South of 60°S. Marine Mammal Science 15(4): 1337-1342

Branch, T.A., Matsuoka, K., and T. Miyashita. 2004. Evidence for increases in Antarctic
blue whales based on Bayesian modeling. Marine Mammal Science 20:726-754

Burt, M and Borchers D.l. Minke whale abundance estimated from the 1991/92 and
1992/93 JARPA sighting surveys. Documents submitted to the 1997 interim JARPA

review. SC/M97 /21. SC/M97 /23

Clapham, P.J. and C.S. Baker. 2002. Whaling, modern. Pg 1328-1332 in Encyclopedia of
Marine Mammals. (W.F. Perin, B. Wursig, and J.G.M. Thewissen, editors). Academie
Press, San Diego

Childerhouse, S., Gales, N., Baker, C. S., Bass, C., Berggren, P., Bickham, J., Breiwick,
J., Brownell, R., Carlson, C., Charrassin, J-B., Cipriano, F., Clapham, P., Collins, T.,

Cooke, J., Cozzi, B., Dinter, W., Engel, M., Findlay, K., Fortuna, C., Funahashi, N.,
Gedamke, J., Groch, K., Iniguez, M., Kasuya, T., Kell, L., Kock, K-H., Krahn, M.,
Leaper, R., LeDuc, R., Mattila, D., Moore, S., Northridge, S., Olavarria, C., Palazzo, J.,

Panigada, S., Parsons, C., Perrin, W., Pomilla, C., Porter, L., Reijnders, P., Ridoux, V.,
Ritter, F., Rabbins,J.,Rogan, E., Rojas, L., Rose, N., Rosenbaum, H., Rowles, T., Sadler,
L., Secchi, E., Senn, D., Simmonds, M., Sironi, M., Stachowitsch, M., Thiele, D., Urban,
J., Wade, P., Van Waerebeek, K., Waples, R., Weinrich, M., Williams, R., Wilson, B.,

Zerbini, A. 2006. Comments on the govemment of Japan's proposai for a second phase of
special permit whaling in Antarctic (JARP Ail). J.Cetacean Res. Manage. 8
(Supplement): 260-261

Clark, C.W. 2006. The Worldwide Crisis in Fisheries. Economie Models and Human
Behavior. Cambridge University Press, Cambridge, UK

379Clark, C.W. 2010. Mathematical Bioeconomics. The Mathematics of Conservation. Third

Edition. John Wiley & Sons, New York

Cooke, J. G. 1995. The International Whaling Commission's Revised Management
Procedure as an example of a new approach to fishery management. Developments in
Marine Biology 4 (Whales, seals, fish and man [AS Blix, L Wall0e, and 0. Ulltang,
editors], Elsevier, Amsterdam): 646-657

Courchamp, F., Berek, L. and J. Gascoigne. 2008. Allee Effects in Eco/ogy and

Conservation. Oxford University Press, Oxford, UK

de la Mare, W.K. 1985. On the estimation ofmortality t'tesfrom whale age data, with
paraticular reference to Minke Whales (Balaenoptera acutorostrata) in the Southern
Hemisphere. Rep. int. Whal. Commn 35:239-250

de la Mare, W.K. 1986a. The sensitivity ofMSY to the parameters ofthe Baleen whale
mode!. Rep. int. Whal. Commn 36:425-427

de la Mare, W.K. 1986b. Fitting population models to time series of abundance data. Rep.

int. Whal. Commn 36:399-418
de la Mare, W.K. 1986c. Simulation studies of management procedures. Rep. int. Whal.

Commn 36:429-450

Donovan,G.P. 2002. International Whaling Commission. Pg 637-642 in Encyclopedia of
Marine Mammals. (W.F. Perin, B. Wursig, and J.G.M Thewissen, editors). Academie
Press, San Diego

Fedak, M. Marine animais as platforms for oceanographie sampling: a 'win/win' situation
for biology and operational oceanography. Memoirs of the National Institute of Polar

Research, Special Issue, 58:133-147

Freitas, C., Lydersen, C., Fedak, M.A., and K.M. Kovacs. 2008. A simple new algorithm
to filter marine mammal Argos locations. Marine Mammal Science 24:315-325

Fujise, Y., Ishikawa, H., Saino, S., Nagano, M., Ishii, K., Kawaguchi, S., Tanifuji, S.,
Kawashima, S. and Miyakoshi H. 1993. Cruise report ofthe 1991/92 Japanese research in
Area IV under the special permit for Southern Hemisphere minke whales. Rep. int. Whal.

Commn 43: 357-371

Gambell, R. 1999. The International Whaling Commission and the contemporary
whaling debate. pg 179-198 in Conservation and Management of Marine Mammals (J.R.
Twiss and R.R. Reeves, editors). Smithsonian InstitutionPress, Washington, DC

Gillespie, A.2005. Whaling Diplomacy. Edward Elgar, Cheltenham UK &

Northamption, MA, USA
Gopnik, A. 2009. Angels and Ages. Alfred A. Knopf, New York

Grinnell, F. Everday Practice of Science. Where Intuition and Passion Meet Objectivity

and Logic. Oxford University Press, Oxford and New York.

Harrison, S., Quinn, J.F., Baughman, J.F., Murphy, D.D., and P.R. Ehrlich. 1991.
Estimating the effectsof scientific study on two butterfly populations. American

380Naturalist 137:227-243

Hatanaka, H., Morishita, J., Goodman, D., Pastene, L.A., and Y.Fujise.06. Response
to Appendix 2 [Childerhouse et al2006). J. Cetacean Res. Manage. 8 (Supplement): 262-
264

Hilborn, R. and M. Mangel. 1997. The Ecological Detective. Confronting Models with
Data. Princeton University Press, Princeton, NJ

Holt,S. 2004. Forward to the 2004 printing. Pg i-xxiii in R.J.H. Beverton and S.J. Holt.

On The Dynamics of Exploited Fish Populations. Blackburn Press. Caldwell, NJ, USA

Ishikawa, H., Goto, M., Ogawa, T., Bando, T., Kiwada, H., Isoda, T., Kumagai, S., Mori,
M., Tsunekawa, M., Ohsawa, T., Fukutome, T.,K., Koyanagi, T., Kandabashi, S.,
Kawabe, S., Sotomura, N., Matsukura, R., Kato, K., Matsumoto, A., Nakai, K.,
Hasegawa, M., Mori, T., Yoshioka, S., and T. Yoshida. 2008. Cruise Report of the

Second Phase of the Japanese Whale Research Program under Special Permit in the
Antarctic (JARPAII) in 2007/2008', Japan, 2000 estimated, IWC Doc. SC/60/04

IWC. 1995. 1994 Report of the Scientific Committee, Rep. int. Whal. Commn 45, 1995,
80-85

IWC. 1998. Repmt of the Intersessional Working Group to Review Data and Results

from Special Permit Research on Minke Whales in the Antarctic, Tokyo, 12-16 May
1997. (SC/49/Rep 1):377-412

IWC. 2007. Report of the 58thAnnual Meeting.

IWC 2007a. 'Report of the Intersessional Workshop to Review Data and Results from
Special Permit Research on Minke Whales in the Antarctic, Tokyo 4-8 December 2006',
IWC Doc. SC/59/REPl

IWC 2007b. Report of the 58thAnnual Meeting, pp 6, 16-17, 39-45

IWC 2009. Process for the review of special permit proposais and research results from
existing and completed permits. J. Cetacean Res. Manage. 11 (Suppl). 2009: 398-401

Japan. 1987. 'The Program for Research on the Southern Hemisphere Minke Whale and

for Preliminary Research on the Marine Ecosystem in the Antarctic', Japan, March 1987,
IWC Doc. SC/39/04

Japan. 1987a. 'The Research Plan for the Feasibility Study on "The Program for Research
on the Southern Hemisphere Minke Whale and for Preliminary Research on the Marine
Ecosystem in the Antarctic'", Japan, October 1987, IWC Doc. SC/D87/1

Japan 2000. 'The 2000/2001 Research Plan for the Japanese Whale Research Program

Under Special Permit in the Antarctic (JARPA)', Japan, April 2000, IWC Doc. SC/52/02

Japan. 2005. 'Plan for the Second Phase of the Japanese Whale Research Program under
Special Permit in the Antarctic (JARPA II)- Monitoring of the Antarctic Ecosystem and
Development of New Management Objectives for Whale Resources', Japan, 2005

estimated, IWC Doc. SC/57/01

381Japan. 2010. Scientific contributions of JARPA/JARPA II and JARPN/JARPN II. IWC

Doc. 62/20.

Jenkins, S.H. 2004. How Science Works. Evaluating Evidence in Biology and Medicine.
Oxford University Press, New York

Karentz D, Bosch I. 2001. Influence of ozone-related increases in ultraviolet radiation on
Antarctic marine organisms. American Zoologist 41:3-16.

Kasamatsu, F., Yamamoto, Y., Zenitani, R., Ishikawa, H., Ishibashi, T., Sato, H.,
Takashima, K. and Tanifuji, S. 1993. Report ofthe 1990/91 southem minke whale

research emise under scientific permit in Area V. Rep. int. Whal. Commn 43: 505-522
Kato, H., Hiroyama, H., Fujise, Y. and Ono, K. 1989. Preliminary report ofthe 1987/88

Japanese feasibility studyof the special permit proposai for Southem Hemisphere minke
whales. Rep. int. Whal. Commn 39: 235-248

Kirkwood, G.P. 1992. Background to the development ofRevised Management
Procedures. Reports of the International Whaling Commission 42:236-243

Kunito, T., Watanabe, l., Yasunaga, G., Fujise, Y. and Tanabe, S. 2002. Using trace
elements in skin to discriminate the populationsof minke whale in southern hemisphere.

Marine Environmental Research 53: 175-1

Larkin, P. 1977. An epitaph for the concept of maximum sustained yield. Transactions of
the American Fisheries Society 106:1-11

Laws, R.M. 1977a. Seals and whales ofthe Southern Ocean. Philosophical Transactions
of the Royal Society B 279:81-96

Laws, R.M. 1977b. The significance of vertebrates in the Antarctic Marine Ecosystem.
Pg 411-438 in Adaptations Within Antarctic Ecosystems. Proceedings of the 3rdSCAR

Symposium on Antarctic Biollogy. G.A. Llano editor. Smithsonian Institution,
Washington, DC.

Lockyer, C. 1974. Investigation of the ear plug ofthe southern sei whale, Balaenoptera
borealis, as a valid means of determining age. Journal of the International Councilfor
the Exploration ofthe Seas 36:71-81

Lockyer, C. 1976. Body weights ofsome species of large whales. Journal ofthe

International Councilfor the Exploration of the Seas 36:259-273-81

Lockyer, C. 2010. Report of the Antarctic Minke whale ear plug experiment. IWC Doc.
SC/62/IA11 (with an associated spreadsheet)

Ludwig, D. and C.J. Walters. 1985. Are age-structured models appropriate for catch­
effort data? Canadian Journal of Fisheries and Aquatic Sciences 42:1066-1072

Mackintosh, N.A. 1965. The Stocks ofWhales. Fishing News Books, London

Mange!, M. 201O.Scientific inference and experiment in Ecosystem Based Fishery
Management, with application to Steller sea lions in the Bering Sea and Western Gulf of
Alaska. Marine Policy 34:836-843

382Mange!, M. 2010a. Review ofC.W. Fowler Systemic Management: Sustainable Human

Interactions with Ecosystems and the Biosphere Marine Mammal Science 26:757-760
Mange!, M., Marinovic, B., Pomeroy, C., and D. Croll. 2002. Requiem for Ricker:

Unpacking MSY. Bulletin of Marine Science 70: 763-781

Mange!, M., Brodziak J.K.T., and G. DiNardo, 2010. Accounting for Variation in Mortality
and Allee Effects When Computing Steepness for Strategie Fisheries Management. UCSC­
SOE-1 0-33. Available at http://www.soe.ucsc.edu/research/report?ID=l596

Mansour, A.A.H., McKay, D.W., Lien, J., Orr, J.C., Banoub, J.H. 0ien, N. and S.
Stention2002. Detenniniation of pregnancy status from blubber samples in minke whales

(Balaenoptera acutorostrata) Marine Mammal Science 18:112-120

Mate, B., Mesecar, R., and B. Lagerquist. 2007. The evolution of satellite-monitored radio
tags for large whales: One laboratory's experience. Deep-Sea Research Il 54:224-227

Mori, M. and Butterworth, D.S. 2006. A first step towards Modeling the krill-predator
dynamics of the Antarctic ecosystem. CCAMLR Science 13: 217-277

Morris, P. 1972, A review ofmammalian age determination methods. Mammal Review
2:69-104

Murase, H., Matsuoka, K., Ichii, T. and Nishiwaki, S. 2002. Relationship between the

distribution of euphausiids and baleen whales in the Antarctic (35°E-145°W). Polar Biol
25: 135-145. document are available on the Institute ofCetacean Research website
(http://www.icrwhale.org/JARP AReview2.htm)

Murase, H., Nishiwaki, S., Ishikawa, H., Kiwada, H., Yoshida, T. and lto, S., 2006. Results

of the cetacean prey survey using echo sounder in JARPA from 1998/99 to 2004/2005.
IWC Doc. SC/D06/J21

Nakamura, T. 1991. A new look at a Bayesian cohort mode! for time-series data obtained
from research talees of whales. Rep. int. Whal. Commn 41: 345-348

Nakamura, T. 1993. Two-stage Bayesian cohort mode! for time-series data to reduce bias in
the estimate of mean natural mortality rate. Rep. int. Whal. Commn 43: 343-348

Nicol, S., Croxall, J., Trathan, P., Gales, N., andE. Murphy. 2007. Paradigm misplaced?
Antarctic marine ecosystems are affected by climate change as well as biological processes

and harvesting. Antarctic Science 19:291-295

Nishiwaki, S., lshikawa, H. and Fujise, Y., 2005. 'Review of general methodology and
survey procedure under the JARPA', Japan, 2005 estimated, Government of Japan
document number JA/J05/JR2. available on the Institute ofCetacean Research website

(http://www.icrwhale.org/J ARPAReview2.htm
Ohsumi, S. 1979. Population assessment ofthe Antarctic minke whale. Rep. int. Whal.

Commn 29:407-420

Ohsumi, S. 1995. The Necessity of Employing Lethal Methods in the Study of Whale
Resources. From 'Research on Whales, ICR, 1995). Available at
http://luna.pos.to/whale/icr _rw_oh.html

383Ohsumi, S. 1997. Development of objectives in the JARPA. Paper in the submitted in
the interim review of JARPA

Pickett, S.T.A., Kolasa, J., and C.G., Jones. 2007. Ecological Understanding. The Nature

ofTheory and the Theory of Nature. Academie Press/Elsevier, Amsterdam

Polanyi, M. 1969. The Republic of Science: Its political and economie them·y.Pg 52-72
in Knowing and Being (M. Greene, editor). University of Chicago Press, Chicago, IL

Raymo, C. 1991. The Virgin and the Mousetrap. Essays in Search of the Sou! of Science.
Viking, New Yorle

Resnik, D.B. 2011. A troubled tradition. American Scientist 99:24-27

Ricker, W.E. 1975. Computation and Interpretation ofBiological Statistics ofFish
Populations. Bulletin 191, Fisheries Research Board of Canada

Roe, H.S.J. 1967. Seasonal formation oflaminae in the ear plug ofthe fin whale.
Discovery Reports 35:1-30

Rock, J.,Pastene, L.A., Kaufman, G., Forestell, P., Matsuoka, K. and Allen, J. 2006. A

note on East Australia Group V Stock humpback whale movement between feeding and
breeding areas based on photo-identification. J.Cetacean Res. Manage. 8(3): 301-305

Tamura, T. and Konishi, K 2009. Feeding Habits and Prey Consumption of Antarctic
MinlΠWhale (Balaenoptera bonaerensis) in the Southern Ocean. Journal ofNorthwest
Atlantic Fishery Science 42: 13-25

Tanaka, S., Kasamatsu, F. and Fujise, Y. 1992. Likely precision of estimates of natural

mortality rates from Japanese research data for Southern Hemisphere minlŒwhales. Rep.
int. Whal. Commn 42: 413-420

T0nnessen, J.N. and A.O. Johnsen. 1982. The History of Modern Whaling. University of
California Press, Berkeley and Los Angeles, CA

Valiela, 1.2001. Doing Science. Design, analysis, and communication of scientific
research. Oxford University Press, New York

Waples, R.S. and O.Gaggiotti. 2006. What is a population? An empirical evaluation of

sorne genetic methods for identifying the number of gene pools and their degree of
connectivity. Molecular Ecology 15:1419-1439

Wiedenmann, J., Creswell, K., and M. Mange!. 2008. Temperature-dependent growth of
Antarctic krill: predictions for a changing climate from a cohort mode!. Marine Ecology

Progress Series 358:191-202

3848. APPENDICES

Appendix A. Biographica/ Details Concerning Marc Mange/

(Web page: http://www.soe.ucsc.edu/~msmangel/)

Marc Mange! is Distinguished Professor of Applied Mathematics and Statistics, Jack
Baskin Endowed Professor ofTechnology and Information Management, and Director of
the Center for Stock Assessment Research at the University of California Santa Cruz,

where he has served since 1996. At Santa Cruz, he also directed the Geographie
Information Systems Laboratory (1996-1999), served as Associate Vice Chancellor,
Planning and Programs (1997-1999) and chaired the Department of Applied Mathematics

and Statistics (2007-09). Since 1 July 2010, he has also chaired the Program in
Technology and Information Management.

From 1980-1996, Mange! was at the University ofCalifornia Davis, where he served as
Assistant, Associate and Full Professor for eight years in the Department of Mathematics
and eight years in the Department ofZoology/Section of Evolution and Ecology. He

chaired the Department ofMathematics (1984-1989) and was founding Director of the
Center for Population Biology (1989-1993).

His awards include the Koopman Paper Prize from the Operations Research Society of
America, 1982; JASA Applications Paper from the American Statistical Association,
1983; Joseph MyerhoffFellowship, Weizmann Institute of Science, 1987; John Simon

Guggenheim Memorial Fellowship, 1987; Fulbright Senior Research Fellowship, Oxford
University, 1988; George Gund Foundation Distinguished Environmental Scholar, Case
Western Reserve University, 1992; Distinguished Statistical Ecologist, International
Association for Ecology, 1998; Mote Eminent Scholar, Florida State University, 2000;

Fellow, California Academy of Sciences, 2000; Fellow American Association for the
Advancement of Science, 2003; UCSC Academie Senate Excellence in Teaching Award,
2003; Frohlich Fellow, CSIRO Hobart, 2006; Astor Lecturer, University of Oxford, 2007;

Kaeser Lecturer University of Wisconsin, 2008; Fellow of the Royal Society of
Edinburgh, 2009; the award for the best paper (out of95) published in The Transactions
of the American Fisheries Society for 2009, for their work on !ife history models of
steelhead trout on the Central Coast of California, and Lamberson Ecology Trust Lecturer

Humboldt State University, 2010.

Mange! has numerous journal publications and books that inelude Decision and Control
in Uncertain Resource Systems (1985, Academie), Dynamic Modeling in Behavioral
Ecology (with Colin Clark, 1988, Princeton), The Ecological Detective. Confronting

models with data (with Ray Hilborn, 1997, Princeton University Press), Dynamic State
Variable Models in Ecology: Methods and Applications (with Colin Clark, 2000, Oxford
University Press), and The Theoretical Biologist's Toolbox. Quantitative methods for
ecology and evolutionary biology (2006, Cambridge, University Press). He edited

Classics ofTheoretical Biology (A Special Issuef the Bulletin ofMathematical Biology.
Part 1:Volume 52 Numbers 1,2. Part II: Volume 53, Numbers 1,2), Sex Allocation and

385Sex Change: Experiments and Models (Lectures on Mathematics in the Life Sciences,
Volume 22) and Proceedings of the Second International Symposium on Krill (Canadian
Journal ofFisheries and Aquatic Sciences 57(Supplement 3). He has supervised more

than 50 undergraduate research projects or senior theses, 20 PhD students and 28 post­
doctoral colleagues.

Mangel and Douglas Butterworth were the first two invited experts to the Scientific
Committee of the Commission for the Conservation ofMarine Living Resources
(CCAMLR) and he served on the US delegation to CCAMLR in 1991. His worlcon

southern ocean krill has been supported by NOAA Fisheries (1994-97), the US National
Science Foundation (1998-2002) and the Lenfest Ocean Program (2006-2010). Mangel
served for six years (1990-1996) on the Committee of Scientific Advisors of the US

Marine Marnmal Commission and in that role lead the effort to update the Principles for
the Conservation of Wild Living Resources (Mangel et al 1996). He served on the
Special Committee on Seals for the Natural Environment Research Council of the UK
from 2004-2011, chairing it from 2008-2011.

386Appendix B. Terms of Reference provided by the Government of

Australia

The focus ofyour report should be on the Second Phase of the Japanese Whale Research
Program under Special Permit in the Antarctic (JARPA II). However, your report should

draw on references to the First Phase of the Japanese Whale Research Program under
Special Permit in the Antarctic (JARPA), where it is relevant.

Your report should reflect your honest belief as to the issues and questions posed.

Please address the following matters/questions in your report:

a) identif)r and outline the essential characteristics of a program undertaken for
purposes of scientific research; and

b) provide a critical analysis of the objectives, methodologies and other features of
JARPA II and, in so doing, assess whether JARPA II has the essential
characteristics referred to in paragraph (a).

387Appendix C- Background Materia/ Provided by the Government of
Australia

The Government of Australia provided the following material:

• The International Convention for the Regulation of Whaling, 1946;

e A range ofiWC documents relating to special permit whaling, including
o resolutions of the Commission concerning special permit whaling and

JARPA and JARPA II from 1987 to 2007;

o relevant extracts of the annual reports of the Commission and Scientific
Committee from 1985 to 2009, including discussions on special permit
whaling and the RMP;

o reports of the IWC interim and final reviews of JARPA;

o summary of special pe1mits issued 1951 to 1987;

• A collection of documents prepared by the Government of Japan, including

o research proposais in relation to JARPA and JARPA II, from 1987 to
2005;

o cruise reports in relation to JARPA and JARPA II, from 1988 to 2010;

o special permits issued in relation to JARPA and JARPA II;

o reportofthe Govermnent of Japan review meeting of JARPA, 18-20
January 2005;

o documents submitted to the IWC interim and final reviews of JARPA;
and

o publications listed on the Instituteetacean Research website.

In addition, I was provided with a range ofscientific papers and publications. Any paper
cited appears in the list ofLiterature Cited.

388 INTERNATIONAL COURT OF ruSTICE

WHALING IN THE ANTARCTIC
(AUSTRALIA v.lAPAN)

Supplement to

An Assessment of Japanese Whale Research Programs

Under Special Pennit in the Antarctic (JARPA, JARPA II)

as Programs for Purposes of Scientific Research in the

Context of Conservation and Management ofWhales*

Marc Mangel

University of California

Santa Cruz

15April2013

•1have prepared this supplementary report pursuant to the request by the Court, communicated
by letter dated 17 October 2012, for full texts of expert statements to be submitted by 15 April
2013. This statement supplementsriginal expert opinion ('An Assessment of Japanese
Whale Research Programs Under Special Permit in the Antarctic (JARPA, JARPA II) as
Programs for Purposes of Scientific Research in the Context of Conservation and Management
of Whales', Mange! 2011; Appendix 2 in the Memorial of Australia). These two documents
together constitute my full expert statement to the Court.

1TABLE OF CONTENTS PAGE

1. INTRODUCTION .........................................................................
.........................3

2. GENERAL OBSERVATIONS ABOUT THE COUNTER-MEMORIAL ........5

3. JARPAH IS NOT FOR 'PURPOSES OF SCIENTIFIC RESEARCH' ...........7
JARPAII LACKS TEST ABLE HYPOTHESES .........................................................7..............
.......................

SAMPLE SIZES IN JARP A IlARE SET WITHOUT CONSISTENT APPLICATION OF ACCEPTED
METHODOLOGY ........................................................................
........9.......................................................
MO DELS ANDDATA- ESPECIALL Y LETHALLY COLLECTED DATA- ARE NOT CONNECTED IN JARP A II
........................................................................
...................11...................................................
...............

JARP A IIGENERALLY LACKS BONA FIDE PEER-REVIEW AND MOST PEER-REVIEWED PAPERS ARE NOT
RELEVANT TO THE CONSERVATION AND MANAGEMENT OF WHALES ......................................14............

4. THE DATA OBTAINED DURING 26 YEARS OF JARPA AND JARPA II
HAVE NOT CONTRIBUTED TO THE RMP ..................................................17

GOAL OF THE RMP ...........·····.·..........................................................17...........
........................................
THE RMP USES MANAGEMENT AND OPERATING MODELS ..............................................18....................

REQUIRED AND ANCILLARY DATA ................................................................19......
.................................
IMPLEMENTATION SIMULATION TRIALS AND LETHAL TAIΠ...........................................19.....................
REFINEMENT AND REVISION OF THE RMP.........................................................................
...................

5. THE DATA OBTAINED BY LETHAL MEANS COULD BE OBTAINED BY

OTHER METHODS .........................................................................
.2.................2

TAGGING ........................................................................
...........22...........................................................
BIOPSY ........................................................................
............24..........................................................
...
PHOTOGRAPHY ........................................................................
.......25......................................................
SUMMARY ........................................................................
...........25.........................................................

6. REASSESSMENT OF THE OBJECTIVES OF JARPA H..............................27

7. CONCLUSION .........................................................................
....0.......................3

LITERATURE CITED .........................................................................
....2..................3

APPENDIX 1: CORRESPONDENCE WITH PROF BRUCE MATE
CONCERNING TAGGING WHALES 16 NOV 2010 ...................................... 36

APPENDIX 2: CORRESPONDENCE WITH DR NICK GALES CONCERNING

TAGGING WHALES 20 MARCH 2013 ............................................................ 37

21. INTRODUCTION

1.1. I have been asked by the Government of Australia to consider my assessment of

Japanese Whale Research Programs Under Special Permit in the Antarctic (JARPA,

JARPA II) as programs for 'purposes of scientific research' in the context of

conservation and management of whales in light of the Counter-Memorial (CM) of

Japan, dated 9 March 2012.

1.2. It is not possible to discuss JARPA II as a program for 'purposes of scientific

research' without having a definition of such activity. The Scientific Committee of the

International Whaling Commission (SC-IWC) has never provided such a definition.

Taking into account the International Convention for the Regulation of Whaling

(ICRW), the practice of the SC-IWC, the Revised Management Procedure (RMP),
1
drawing on my own experience in basic and applied ecology and consistent with the

general practice of science, I previously concluded that a program for 'purposes of

scientific research' in the context of conservation and management of whales (Mangel

2011, para 4.39):

(a) has defined and achievable objectives that aim to contribute knowledge that is

importantto the conservationand managementof whale stocks;
(b) uses appropriate methods that are likely to achieve the stated objectives,

including:

(i) lethal methodsonly wherethe objectivesofthe researchcannot be achieved
by any other means (for example,by the analysisof existingdata and/orthe
use of non-lethalresearchtechniques);

(ii) settingsample sizes usingacceptedstatisticalmethodology; and

(iii)linkingmathematicaland statisticalmodelsto dataconsistently;
(c) includes periodic review of research proposais and results, and adjustment in

responseto such review;and
(d) is designedto avoid adverseeffects on the stocks beingstudied.

1.3. These are minimum criteria that reflect established practice, and also take into

account the approach and criteria of the IWC. Ali of these criteria are required for an

activity to be considered a program for 'purposes of scientific research' in the context of

1 My experience includes working on Southern Ocean krill sinee 1986 (I am one of the first two invited
experts to CCAMLR and served on the US delegation to CCAMLR in 1991, as a visitor in 2006, and
organizing a 2007 meeting on resolving uncertainties in krill fishery management models), membership
of the Committee of Scientific Advisors of the US Marine Mammal Commission 1990-1996, member of
the delegation of the United States to the Scientific Review of Large-Scale Pelagie Driftnetting, member
(2004-20 11) and chair (2008-2011) of the Special Committee On Seals, a statutory committee that
advises the British Government on the conservation and management of seals, and numerous publications

in both basic and applied ecology.

3conservation and management of whales. I am unaware of authoritative alternative

views (e.g. that science does not require testable hypotheses or that sample sizes should
not be set in a consistent manner using appropriate statistical methodology); such views

would be far from the mainstream of modern science.

1.4. A central issue brought into foc us by the CM is whether each Party to the ICRW

can decide for itself, on the basisof its own subjective criteria, what is a program for

'purposes of scientific research' in the context of conservation and management of
whales, or whether this can be done only on the basis of objective criteria. The proper

characterization of an activity as a program for 'purposes of scientific research' must be

based on sound and accepted principles and not subjective assertion. The determination

of what constitutes 'science' is to be assessed by reference to boundaries of practice and
principle, applying objective criteria; from a scientific perspective, if those boundaries

are crossed the practice cannot amount to 'science'.

1.5. In this supplement, I first provide general observations about the CM, and then

explain that: (i) JARPA II is not for 'purposes of scientific research'; (ii) the data

obtained by lethal means over a 26 year period have not contributed to the RMP and are
not likely to contribute to it in the future; and (iii) the data obtained by lethal means

could be obtained by other methods. I then return to the Objectives of JARPA II

(CM para 5.20) and provide a reassessment of them in light of my previous report, the

CM, and this current report. I conclude once again that JARPA II is not a program for

'purposes of scientific research' in the context of conservation and management of
whales.

42. GENERAL OBSERVATIONS ABOUT THE COUNTER-MEMORIAL

2.1. As a practicing scientist actively involved in peer-review, both as author and

editor, I had expected a point-by-point response in the CM to my analysis. With the

exception of my comments about the Allee effect (Mange! 2011, paras 5.64-5.67; CM

para 5.86), the CM lacks such a response. Furthermore, the CM introduces a number of

points that I consider to be in·elevant or erroneous, including by omission.

2.2. Among the most substantive missing or irrelevant points in the CM are the

following:

0
The response in the CM to the fundamental criteria in para 1.2 above is cursory.

The CM describes them as 'arbitrary' (CM, para III.15) and suggests that the criteria are
no more than my persona! opinion (CM, para 9.10). However, the CM puts forward no

authority for these assettions other than stating that Japan too has scientists. The CM

fails to set out any alternative criteria for what qualifies as a program for 'purposes of

scientific research'. With this approach, any activity that Japan chooses to undertake

can be characterized as science.

• The response in the CM to my conclusion that JARPA II generally lacks

hypotheses, except for the krill hypothesis (which is not so much tested in JARPA II as

assumed to be true) is

JARPA II, however, does not purport to verify the validity of the krill surplus
hypothesis. What it tries to do is to incorporatedata on other animalslfishthat prey on
krillin order to develop a "model of competitionamong whale species", considering

severa!hypothesesexplainingc657ges in abundanceof baleen whale species including
the krillsurplushypothesis • Australia'sallegationsthat JARPA II is not designedor
not conductedto verifythe validityof the krillsurplushypothesis 658are thus besidethe
point.(para 5.31)

However, none of these 'severa! hypotheses' are described, leaving the reader to guess

what they might be.

e The CM provides no analyses illustrating why lethal take is necessary; rather this is

simply asserted, as has been done since the outset of JARPA.

2 1 currently serve on the editorial board of the following scientific journals: The American Naturalist,
Environmental and Ecological Statistics, Evolutionmy Ecology Research, Israel Journal of Ecology and
Evolution, Oecologia, and Theoretical Population Biology, and am on the Senior Advisory Council,
Natural Resources Modeling. From 1994 to 1999, 1served as co-editor of Behavioral Ecology. I held a
variety of editorial positions previous to that.

5e The CM introduces extraneous information that distracts from the main points. The

question of whether JARPA II is a program for 'purposes of scientific research' in the

context of conservation and management of whales is not informed by discussion in the

CM on: (i) whether minke whales in the Antarctic can sustain a take, or are endangered

or threatened with extinction (CM, para 40); (ii) whether minke whales should be
harvested or not; this is an ethical question (as weil as a policy question) not amenable

to the methods of science (e.g. Weinberg 1972); or (iii) the policy position of Australia

on the resumption of commercial whaling (CM, Footnote 318).

e The elaborate description of.telomeres (the ends of DNA molecules) as a means of
aging (CM para 4.67) and why they will not work is the needless introduction of a straw

man, since 1 did not raise the possibility and telomeres are not mentioned elsewhere in

the Memorial of Australia.

e On the other hand, the CM is silent about using blood samples to assess

reproductive hormones (for example, CM para 4.70) and brushed away the notion of
using biopsy to measure pollutants with the comment that biopsy sampling "would not

be possible for ali contaminants" (for example, CM para 4.79). However, lacking a

conceptual framework through testable hypotheses we have no idea which contaminants

are important or why.

63. JARPA II IS NOT FOR 'PURPOSES OF SCIENTIFIC RESEARCH'

JARPA II Lacks Testable Hypotheses

3.1. JARPA II follows in the tradition of JARPA as a program for the collection of
data with the assertion that it will somehow inform the conservation and management of

whales in the Southem Ocean, but without the requisite conceptual framework to make

this outcome even likely.

3.2. Nearly ail of the quotations provided in the CM in support of JARPA and

JARPA II as programs of science are actually more consistent with it being nothing

more than concept free data collection, without defined and achievable objectives or
based upon clearly testable hypotheses. 3

3.3. However, collecting data can only be a part of scientific research when such

collection is associated with a hypothesis. As I previously noted (Mange! 2011, paras

4.9-4.13), a program for 'plll·poses of scientific research' requires a testable and

operationally defined hypothesis. In other words, one needs both a hypothesis (or

hypotheses) and a means to test it so that one can leam about nature.

3.4. Lacking these requirements, the program may collect data but is not one of
scientific research. Simply taking such measurements does not constitute a program for

'purposes of scientific research'. Even induction (the process of forming or developing

the relevant question as data collection proceeds, rather than having it clearly articulated

from the outset) requires sorne sense of question if it is to be anything other than the

mere collection of data.

3.5. For example, serious bird-watchers are well known to assemble '!ife Iists' ofbirds
that they have seen, yet nobody would consider such !ife lists any kind of scientific

program exactly because such lists Jackhypotheses and a conceptual framework.

3.6. Similarly, it is weil understood that monitoring is an essential component of

environmental protection. Although such monitoring for environmental protection

helps safeguard both human health and the environment, monitoring itself is never

suggested by proponents to be a program for 'purposes of scientific research'.

3 Examples in the CM include paras 4.15, 4.30, 4.49, 4.54, 4.90, 4.114, 4.152, 5.98, 5.101, 8.68.

73.7. This principle is weil accepted in the scientific community. One long-term goal

of the philosophy of science in the 20th century was to establish how to demarcate

science from non-science. 4 It is now generally accepted that "[t]he common thread in

ali science is the ability to produce and test hypotheses based on systematically

collected empirical data (via experiments or observations)" (Pigliucci 2010, p. 23).

That is, it is nowa well-established principle accepted in the scientific community that a

testable hypothesis or conceptual framework is an indispensable component of scientific

research. The mere collection of empirical data without a testable hypothesis simply

cannat be treated as 'scientific research'. By formulating a problem, hypothesis or

rigorous conceptual framework that can be tested, we create scientific propositions.

However, "[p]ropositions [i.e. hypotheses or objectives] that are so loosely framed asto

be untestable are very slippery indeed" (Foster and Huber 1999, p. 233) since they can

never be assessed.

3.8. It is thus weil accepted in the scientific community that the mere collection of data

does not amount to science. Rather, one requires a hypothesis and a means for testing

the hypothesis in arder to learn from the outcome. Peters (1991, p. 223) noted that the

first step for the scientist "is to identify a relevant question or hypothesis to test as the

goal of the research". Similarly, Karban and Huntziger (2006, p. 60) write

The first stepin doing researchis to have a clear questionor hypothesisin your mind. If
you are vaguely interested in a system (an organism or an interaction), you are not
ready.... You must be able to fonnulate your ideas into a clear question. Without a
clear question, there is no end to the data (relevant or otherwise) that you may fee!
5
compelledto collect.

3.9. Without the conceptual framework created by hypotheses, one cannat do science.

One could, of course, collect lots of data, as in JARPA and JARPA II, but as Platt

(1964, p. 349) noted, "years and decades can easily be wasted on the usual type of' law­

information' observations or experiments if one does not think carefully in advance

about what the most important and conclusive experiments [or observations] would be".

This point has been well understood and accepted by the scientific community long

before JARPA began 26 years ago.

4 This philosophical effort is associated with the intellectual giants Karl Popper, Thomas Kuhn, Imre
Lakatos, and MichaelPolanyi, among others.
5 Other authorities supporting this view include Angier (2007, p. 32), Casti (1989, p. 11-14), Chalmers

(1999, p. 59-73), Cromer (1993, p. 20), Giere (1997, p. 29-38), Rigler and Peters (1995, p. 16), Shermer
(2001, p. 133), Ziman (1991, p. 32).

83.10. In summary, lacking testable hypotheses, long-term programs may collect data

(even in considerable amounts), but they cannat properly be considered to be programs

for 'purposes of scientific research'. As I have stated previously (Mange! 2011), and

which was not rebutted in the CM, JARPA II Jacks testable hypotheses that are the
foundation of science (Mangel 2011, para 5.22); it is, at best, nothing more than a

program for the collectionof data (Mangel2011, paras 6.1, 6.19).

Sample Sizes in JARPA II Are Set Without Consistent Application of Accepted
Methodology

3.11. Lacking any testable hypotheses, it is impossible to set sample sizes in a proper

manner, because sample size is properly set by reference to what is required to answer

the focal question. That is, without a focal question, one Jacks the necessary touchstone

to determine how big the sample needs to be. I noted that JARPA and JARPA II lacked

clarityin the setting of sample sizes (Mange! 2011, paras 5.38-5.48). Although the CM
gives more detail on sample sizes (para 5.57-5.71), there is no improvement in clarity.

3.12. The additional detail in the CM concerning sample size is a reference to a table

published by The Research Advisors (and available at http://research­

advisors.com/tools/SampleSize.htm) (CM Footnote 712) and Appendices of the

JARPA II proposai (CM Footnote 715).

3.13. The table from The Research Advisors is used in CM para 5.58 to show that if a

population size of 761,000 is assumed, along with a margin of error of 3.5%, and a

confidence interval of 95%, then the required sampied size is 783 - which happens to be

very close to the sample size used in JARPA II. However, in this simple example, no

explanation is given of why a margin of error of 3.5% is selected - that is, why would
one need this leve! of margin of error when other choices are available? For example, if

the margin of error were 1.0%, then the sample size will be about 9400, but if the

margin of error were 5.0% then the required sample size would only be 384- nearly 25

times different. And if the margin of error were 10.0% (while not included in the table,

this can be a legitimate scientific choice for certain purposes), then the required sample
would be even smaller. Even though this is an example, Japan offers no explanation as

to how and why one would decide that a margin of error of 3.5% is chosen to be

appropriate. Without having a focused question, one cannat decide on the appropriate

9choice; the appropriate margin of error can only be selected in light of what is needed to

answer the focused question.

3.14. Simply put, knowing which margin of error to apply requires knowing the
question that is intended to be answered, which in turn requires a hypothesis.

3.15. The CM then turns (paras 5.59 ft) to three examples (age at maturity, pregnancy

rate, and rateof change ofblubber thickness) illustrating how sample size is determined.

Without reference to hypotheses that would indicate tolerable margins of error, the CM
concludes that the sample sizes for these are, respectively, from 594 to 1288 [CM para

5.67], from 663 to 1617 [CM para 5.68] and from 818 to 971 [CM para 5.69].

3.16. The CM continues that for just these three items the necessary sample size is

somewhere between 594 and 1617, that for most of the myriad of data collected (more

than 100 per whale; CM para 5.59) the necessary sample size is around 800 per year
(CM para 5.70) and that 850 was finally adopted as the JARPA II sample size (CM para

5.71). For those items for which this sample size is too small, the CM asserts that the

compromised accuracy will be mitigated by "comprehensively integrating many

different data and analyses" (CM para 5.71).

3.17. Although a sample size of 850 might seem reasonable from a superficial analysis
(and roughly matches sample sizes of JARPA), the choice is not fully explained and

cannat therefore be taken as being correct. Rather, the upper limit of the range of

sample sizes identified for each research item-namely 1617-would be the value

chosen by a program for 'pm·posesof scientific research' since any smaller value will
result in samples that do not meet the criteria of margin of error or confidence interval

for at least some of the research items. By using a sample size smaller than the upper

limit, oneis essentially changing either the margin of error or the confidence interval for

those research items Japan had previously concluded required sample sizes above 850 in
an arbitrary and ad hoc fashion.

3.18. To be clear: 1am not suggesting that more whales should be killed, because it is in

fact not necessary to kill any whales for 'purposes of scientific research' in the context

of conservation and management of whales. Rather, I am explaining the lack of

consistency in the application of methodology.

103.19. These examples illustrate the flexibility that is misused in the statistical methods

adopted by Japan. For example, the table referred to above specifies a required sample
size as a function of the size of the population sampled, the margin of error one is

willing to accept (1%, 2%, 3.5%, or 5%), and the confidence that one is correct (95% or

99%). Flexibility arises because the margin of error and leve! of confidence are matters

of judgment and choice. Although there are accepted approaches regarding their choice
(as stated above, this must be linked to the question being asked), Japan provides no

explanation regarding its choices.

3.20. The same kind of flexibility in terms of exercising judgment and choice, and

which is equally capable of misuse, applies to the more complicated examples given

conceming age at maturity, pregnancy rate, and blubber thickness. Indeed, there is even
more flexibility in these more complicated examples because, in addition to the choices

in the previous paragraph, there is the anticipated leve! of change and the interval over

which that change is intended to be detected. The overall result is that without a

conceptual framework in which data collected are embedded, one can select almost any

sample size and describe it as being required by way of retrospective reference to
unexplained choices of each of the parameters. This appears to be what is done in

JARPA II.

3.21. The actual takes, reported in CM para 5.72, are far below the target of 850 except

in 2005/06, thus suggesting that the entire set of samples from 2006/07 onwards are

compromised. Although the CM tries to explain the reasons for the reduced take, it
makes no effort to explain how the comprised accuracy will be mitigated by

"comprehensively integrating many different data and analyses" (CM para 5.71).

3.22. In summary, it is still unclear how sample sizes are set in JARPA II. The only

point that can be made clearly is that they are not set in a manner consistent with the
proper conduct of a program for 'purposes of scientific research'.

Models and Data - Especially Lethally Collected Data - are Not Connected in
JARPA II

3.23. JARPA II also fails to meet the requirement of a program for 'purposes of
scientific research' in the context of conservation and management of whales that

models and data are to be linked in a consistent fashion.

113.24. In ecology, models are used in a variety of ways, including: (i) for synthesis and

integration of data; (ii) to provide guidance to empiricists; and (iii)for prediction.
6
Complicated models are not necessarily more useful, or better, than simpler models.

Indeed, in many cases more complex models will often give Jess accurate answers than
simpler variants (Ludwig and Walters 1985).

3.25. The CM is not consistent in its approach to models and data: Japan asserts one

thing in its objectives (the development of a mode) of the ecosystem), and then does

another in the methods it chooses and the actual conduct of its program (focusing on a

small component of the ecosystem). A simplified version of the Southern Ocean

ecosystem is found in Figure 5-1 of the CM. I have reproduced that figure here

(Figure 1 below).

Figure 1: The simplified view of the Southern Ocean ecosystem, from the CM and
based on Beddington and May (1982). Note that JARPA II collects data on only a

single species of one component of the ecosystem (the baleen whales, red circle).

3.26. The CM reinforces the inconsistent approach within JARPA and JARPA II

regarding the development of the ecosystem model. Japan asserts: (i) that JARPA II

6 Additional authorities suppoiiing my expert opinion include Giere (2006, p. 60), Johnson (2001,
p. 105 ft), Karban and Huntzinger (2006, p. 25-27), Oreskes (2003), Rastetter (2003), Taper and Leie
(2004), van Fraassen (2010, p. 13), Ziman (1991, p. 77 ff).

12will lead to the development of a management mode! for the whales of the Southern

Ocean ecosystem; and (ii) that lethal take of minke whales is required for the

development of this madel because non-lethal methods are insufficiently accurate. 7

3.27. However, JARPA II almost exclusively focuses data collection on minke whales

(sorne data are also collected on fin whales, and on some occasions krill), which is but

one small component of the ecosystem. The outcome is that JARPA II workers will

have a large amount of supposedly necessary data on that single species in its ecosystem

madel, while nearly ali of the other components of the madel (e.g. the other baleen

whales, birds, mammals) will be characterized in some unknown way. By the logic of

the CM, the resulting madel cannat possibly provide any useful results in assessing the

ecosystem.

3.28. Second, JARPA II workers are themselves inconsistent in the assertion that

generally accepted non-lethal methods are not accurate enough for the purposes of

ecosystem modeling. For example, Tamura and Konishi (2006), two JARPA II

workers, compared estimates of krill consumption using daily changes in stomach

content (a lethal method) with estimates using classic (Kleiber 1947) methods that do

not require lethal take and concluded that the "results showed that estimated daily prey

consomptions were similar between the above two methods" (p. 1) and that although

both methods were subject to en·ors "the estimates based on two independent methods

were coïncidence well [sic] each other" (p. 6-7). This paper was submitted for the

JARPA review in 2006. A somewhat different version was published later in a peer­

reviewed journal (Tamura and Konishi 2009). Although data on stomach contents are

discussed in the later paper, only the non-lethal method is used to compute an estimate

of prey consumption. Thus, JARPA II workers themselves have shawn that non-lethal

methods are equally as accurate as the lethal ones and when writing for scientific

colleagues chose to use the non-lethal ones.

7 For example, the CM states that: "Non-lethal allometric technican only produce rough and
indirect estimates of food consumption, which are not reliable enough for use as input data in ecosystem
models" (CM para 4.72; a similar comment is made in para 4.76).

13JARPA II Generally Lacks Bona Fide Peer-Review and Most Peer-reviewed Papers are
Not Relevant to the Conservation and Management of Whales

3.29. As described previously (Mange! 2011), true peer-review (the periodic review of

research proposais and the results, and adjustment in response to such review) is an

essential component of a program for 'purposes of scientific research'. The reason is

that science is the development of consensual knowledge. Ziman (1991, p. 3) wrote

"... scientific knowledge is the product of a collective human enterprise to which

scientists make individual contributions which are purified and extended by mutual

criticism and intellectual cooperation ... the goal of science is a consensus of rational

opinion over the widest possible field". Peer-review plays a key role in the

development of such consens ua}knowledge.

3.30. A key component of proper peer-review is that it is independent and lacks bias.

This is important because

The experts in a particular field can become so indoctrinated and so committed to the
current paradigm that their critical and imaginative powers are inhibited, and they

cannot 'seebeyond their own noses'. ln this circumstancescientific progress may come
to a halt -- knowledge may even regress -- until intellectual intruders come through the
interdisciplinaryfrontiers and look at the field without preconceptions.(Ziman 1991, p.
134).

3.31. The indispensable nature of adequate peer-review to a program such as JARPA

and JARPA II is evident in the following statement:

Every scientific paper and report has togo through the critical scrutiny of other experts:
peer review. Scientific authors are required to take reviewers' comments and criticisms

seriously, and to fix any mistakesthat may have been found. It's a foundationalethic of
scientific work: no claim can be considered valid- not evenpotentially valid - until it
has passed peer review(Oreskesand Conway2010, p. 3-4).

3.32. Oreskes and Conway (2010, p. 269) also note "[i]n science, you don't get to keep

harping on a subject until your opponents just give up in exhaustion". Yet this pretty

much seems to be what happens in the SC-IWC. Additional authorities supporting the
essential nature of unbiased peer-review are found throughout the entire scientific

commumty.. 8

3.33. The CM considers that peer-review outside of the SC-IWC is time consuming

(paras 4.108, 4.109) and by implication not worth the effort and delay. To be sure,

B For example: Casti (1989, p. 14), Cramer (1993, p. 145), Legendre (2004, p. 53), Shenner (2001, p.

317).

14peer-review may sometimes be lengthy, but it is also the only way we know to establish

unbiased assessment of research worlc The most effective peer-review is anonymous

and often requires mandatory changes before worlc can go forward (in the case of
proposed research) or be published (in the case of completed research). As described in

the CM (paras 4.108, 4.109) and in Clapham et al. (2003), review in the SC-IWC does

not have these features. Indeed, Clapham et al. (2003, p. 212) note that in SC-IWC the

authors of a proposai play a major role in writing the resulting evaluation. JARPA II

fails to be a program for 'purposes of scientific research' on the criterion of peer­
review.

3.34. Even so, the CM (paras 4.112-4.114) describes a total of 195 documents from

JARPA/JARPA II between 1988 and 2009, ofwhich 107 are said to be peer-reviewed

papers. I previously analyzed the material presented in CM Footnote 511 (Mange!

2011, paras 5.58, 5.59). I noted that(i)publications from JARPA/JARPA II generally
do not appear in peer-reviewed journals outside of the IWC; (ii) that only about 1/6 of

the articles were peer-reviewed and broadly relevant to conservation and management

of whales; and (iii) that nearly 40% of the peer-reviewed articles related to lipid

biochemistry or reproductive physiology, thus irrelevant to the conservation and

management ofwhales and to the stated objectives of JARPA II.

3.35. The CM (para 5.99) notes that two peer-reviewed publications from JARPA II

have been published and that "a larger number ... are expected" in the future. The two

papers published at that point, described in CM Footnote 774, are about whale

morphology and reproductive physiology - i.e. they are totally irrelevant to the

conservation and management of whales and to the stated objectives of JARPA II.

3.36. Given the promise of fmthcoming peer-reviewed publications in para 5.99 of the

CM, it is instructive to consider once again the 'scientific contributions' from JARPA

and JARPA II. To do this, I accessed the ICR website

http://www.ictwhale.org/sc.JARPA.html on 4 April 2013 and downloaded the report

Scientific Contribution from JARPAIJARPA II (December 2012), which lists written
documents by year. This is an updated version of the document referred to in

Footnote 511 of the CM. Based on document titles, it is not clear which ones relate to

JARPA or JARPA II, and so I do not distinguish in my analysis. Documents are

15identified as peer-reviewed or as 'unpublished'. In light of the essential nature of peer­

review discussed above, I again focus on Japan's peer-reviewed documents.

3.37. Focusing on the additional 'contributions' after 2009 (I have already analyzed the
contributions up to 2009, see paragraph 3.34 above), it is claimed that

JARPA/JARPA II have produced 25 documents between 2010 and 2012, of which 15

are peer-reviewed. Of these 15, one is in Norwegian and 11 are in Japanese, making

them generally inaccessible to scientific colleagues (no English translation has been

provided). Furthermore, based on the citation information, seven of the 11 papers in
Japanese appear to be two pages and one of the 11 appears to be three pages; they

appear to be nothing more than abstracts ofwork rather than full analyses.

3.38. The three remaining peer-reviewed papers are about: (i) stock structure and

migratory routes of Antarctic minke whales, using lethal and non-lethal methods (as I

describe below in Section 5, this work could have been done with only non-lethal
methods); (ii) abundance estimates of humpback whales from sighting surveys using

non-lethal methods; and (iii) genetic diversityin blue whales, determined using biopsy

samples collected by Japanese vessels during the IDCR/SOWER cruises rather than

from JARPA or JARPA II (Sremba et al. 2012, p. 3). This last paper in particular

demonstrates that when JARPA researchers wish to, they are able to make use of non­
lethal methods (for discussion on non-lethal methods, see section 5 below).

3.39. As such, of the 15 peer-reviewed papers devised from JARPA and JARPA II

between 2010 and 2012, 12 are inaccessible to the scientific community for peer-review

and three eitherin fact use, or could have achieved the same results using, entirely non­

lethal techniques. On this basis the 15 papers do not provide any support to Japan.

3.40. Clapham et al. (2003, p. 211)- ail of whom were members of SC-IWC - writing

about this trendin JARPA note that there are many peer-reviewed articles "on tapies of

no value to management" and that "JARPA's failure to publish in international refereed

journals says much about the quality and motives of its science".

164. THE DATA OBTAINED DURING 26 YEARS OF JARPA AND JARPA II
HAVE NOT CONTRIBUTED TO THE RMP

Goal of the RMP

4.1. The RMP is designed to correct the two major problems of previous management

regimes: (i) to prevent the depletion of whales; white (ii) maintaining the stability of

catches at as high a leve! as possible, consistent with (i). In doing so, the SC-IWC put

aside the disproven and futile notion that Maximum Sustainable Yield (MSY) and
9
Maximum Sustainable Yield Leve! (MSYL) can be measured in the field. Instead, the

SC-IWC recognized that we will always have only approximate knowledge of stock
levels and recruitment curves so that a successful management regime must be robust to

these uncertainties.

4.2. Contrary to the assetiion in the CM, the RMP is not complicated. It can be best

understood as a simulation of whaling using a computer madel. The number of whales

in a population in the following year will be determined by the number this year, the net
productivity (new whales produced minus natural deaths), and the take, if there is one

(for more details on the dynamics of populations, see Mange! 2011, paras 3.8-3.14).

For whale populations in nature, we generally do not know the precise pre-exploitation

population size (the historical population size before whaling commenced), the current

leve! of depletion from the pre-exploitation leve!, or the productivity, and can only
measure current population size inaccurately.

4.3. However, computer simulations allow us to ask: "what would the dynamics of a

whale population be if we assume values for pre-exploitation population size and

productivity as a function of population size?" We can then compare the predicted

dynamics of the whale population with the inaccurately observed estimates of

abundance and, by repeating this process over and over again, learn about more likely
and less likely valuesof productivity and current levels of depletion. The RMP includes

a decision rule (para 4.5 below) that sets the catch based on our estimates of

productivity and population size.

9 For explanation ofthese two terms, see Mangel2011, para 3.12.

17The RMP Uses Management and Operating Models

4.4. The approach taken by SC-IWC when developing the RMP, in which it undertook

computer simulation testing of management procedures to determine the management
measures that would achieve the two goals referred to in para 4.1, is now generally

known as Management Strategy Evaluation (MSE; Smith et al. 1999, Mangel2010). In

MSE, one tests a management madel against a variety of 'operating models'. The idea

is that each of the operating models is a different - and potentially very complex -

description of how the world might worlc We seek a management mode! that is
effective, regardless of which of these operating models best characterizes the natural

world (it is almost a surety that none of the operating models are exactly correct, and

therefore the management mode! must be robust to ali of them).

4.5. The management mode! in the RMP is known as the Catch Limit Algorithm

(CLA). The CLA is based on the recognition that the true state of the stock (level of
depletion/current abundance and productivity) is unknown. For this reason, the

population mode! in the CLA is extremely simple. Neither the population madel nor the

parameter values in it are intended to give an accurate representation of the dynamics of

a whale population; this is not their purpose. Rather, the madel has been shown by
simulation trials to allow robust calculation of catch limits (IWC 1999). Furthermore,

the CLA is designed to learn about the key parameters of the managed stock -

productivity and depletion - as the RMP is applied without the need for lethal take other

than historical records of levels of past commercial catch.

4.6. The CM asserts that biological parameters collected by JARPA/JARPA II are
'essential' for certain elements of the RMP (CM para 4.165). When testing the CLA, it

is possible, of course, to use extremely complicated 'operating models' based upon a

great deal of biological detail. The only constraint on operating models is the creativity

of the workers developing it. However, lethally obtained data are not necessary for the

development of such operating models. By focussing on the operating mode!, as is done
in JARPA II and the CM, an unlimited amount of biological data can be collected by

lethal talŒwith the claim of making a better operating mode!. However, the same can

be achieved without lethal take as explained in Section 5 below.

4.7. The RMP is conservative because a procedure that takes into account large ranges
of uncertainty and a wide range of operating models must be conservative. A policy

18that leaves one 'extra' whale in the ocean, or a policy that kills one whale too many will

both miss the target by one individual. However, in general, these errors are not
symmetrical-in this simplified example, the one extra whale left behind can still be

whaled next year (that is, the error can be corrected), but the one extra whale taken

cannot be replaced (the error cannot be corrected). The RMP captures this asymmetry

as weiL

4.8. The strength of the simulation testing in the RMP is that it eliminates the need for

detailed biological data obtained through lethal take (Kirkwood 1992, Cooke 1995).

Required and Ancillary Data

4.9. The data required for the RMP are: (i) numbers of ali past catches; (ii)relative

abundance data; and (iii) estimates of absolute abundance (IWC 1994, 1999; Kirkwood

1991). That is, the simulations in the RMP (para 4.2, 4.3 above) use only numbers of

individuals (catch and abundance). The SC-IWC considered including additional data
when developing the RMP, but decided against it (Kirkwood 1991).

4.10. In summary, the RMP is a simple and elegant means for leaming about the

uncertainty inherent in the natural world. It operates by collecting appropriate data and

comparing those data with predictions of well-understood models. None of the

biological parameters - such as natural mortality rate, pregnancy rate, age at sexual
maturity - that are collected by JARPA II through lethal take are required by the RMP.

Nor can it be said that they assist in the operation of the RMP, or that they can

contribute to the improvement of the RMP.

Implementation Simulation Trials and Lethal Take

4.11. The IWC management boundaries do not necessarily reflect the actual biology

and stock structure of whales. This is called the 'stock identity problem' and is solved
in the RMP by dividing the management areas into small regions (called 'small areas')

in which the CLA is separately applied. The premise is that the area is small enough

such that either only one stock occupies it or that when multiple stocks are present they

are well mixed within the small area. The SC-IWC has developed a sequence of mufti­

stock rulesto implement the CLA consistently with this premise.

194.12. To test the multi-stock rules, the SC-IWC conducts simulations called

Implementation Simulation Trials (ISTs). These can be viewed as a variant of the
approach that underpins the RMP. One develops a series of hypotheses about how

many stocks are present and how the whales move about and mix. For each hypothesis

one can then determine a catch limit from the CLA. Using computer simulation, one

can then test how the catch limit that has been determined using one hypothesis
performs against alternative hypotheses. This process can be repeated over and over

again, such that each potential catch limit is tested against every alternative hypothesis.

Ultimately, the catch limit arrived at is the one most robust to the range of uncertainty

that is unavoidably present (since the true state of nature is always unknown).

4.13. This is a form of risk analysis in which consequences are explored as a function of
the difference between the true and assumed states of nature. It is a well-regarded and

powerful tool for environmental protection. In the case of the RMP, it allows us to

determine a catch rule that is robust across the potential states of nature.

4.14. Contrary to the assertion in the CM paras 4.164, 4.165 and Figure 4-12, none of
the biological parameters- such as natural mortality rate, pregnancy rate, age at sexual

maturity - that are collected by JARPA II through lethal take are essential for the ISTs.

Rather they are ancillary, since one can develop the hypotheses for the ISTs without

them, using non-lethal data (examples are given in Section 5). Indeed, the assertion in

the CM that lethal take data are essential for the RMP or any element of it is in fact
contrary to the fundamental design of the RMP that lethal take other than catch levels

from past commercial harvest is not required.

Refinement and Revision of the RMP

4.15. The RMP is intended to apply to all baleen whales, rather than just minke whales.

ln order to do so, the starting point of the simulations of population numbers considers a

wide range of productivity, current depletion (ranging from virtually extinct stocks to
unexploited ones), and inaccuracies of surveyed abundance. As the RMP is applied to a

particular stock in a particular area, and data are obtained, there an in-built process in

the RMP of learning about that focal stock (this is called the 'joint posterior' of

depletion and productivity; see IWC 1994, p. 148).

204.16. Even if achievable, using fieldwork to shrink the range of productivity that forms

the starting point of the RMP, which is one of the objectives of JARPA II, would in fact

undermine the applicability of the RMP to other species of baleen whales. That is, we

expect that the different species of whales will have differing productivities, determined

by their differing biology. To constrain the starting point of the RMP so that it 'fits'
minke whales is likely to make it fail for other species ofwhales.

4.17. On the other hand, if Japan wanted to achieve the same outcome of 'improving'

the RMP for Antarctic minke whales only, it could do so by performing repeated 'dry

runs' of the RMP in a small area and allowing the model's inbuilt leaming mechanism
to make any appropriate adjustments. This could be done using entirely non-lethal data,

since only sighting sm·veysfor abundance are required for the RMP.

215. THE DATA OBTAINED BY LETHAL MEANS COULD BE OBTAINED BY
OTHER METHODS

5.1. The CM continues Japan's long-standing tradition (e.g. Ohsumi 1995) of asserting

that lethal methods are required as part of JARPA and JARPA II. At no point, it seems,

has Japan genuinely explored alternative non-lethal methods. Rather, it takes lethal

means as an accepted proposition, and then asserts that lethally acquired data are
essential for the RMP, which they are not. In doing so, JARPA II uses exactly the same

methods as JARPA, as if 25 years of scientific and technological development had not

occurred.

5.2. In the CM, non-lethal methods are presumed in advance not to be workable. The

CM relies on statements from others such as the conclusion that logistics and abundance

"probably preclude their [non-lethal methods] successful application" (CM, para 4.61).
It claims that skin biopsy is not practicable because a new and heavier projectile, for use

in open water rather than coastal areas, would be required and that"[t]he use of heavier

projectile units would necessitate the use of more powerful delivery units in order to

obtain the necessary range and trajectory. However, adding mass and power to the

projectile unit increases the riskof unwarranted penetration and damage to the target
animal 454" (CM para 4.75). It is a perplexing logic that rules out by assertion a method

on the basis that it is preferential to kill an animal with certainty rather than to risk

possible damage to it.

5.3. In contrast to the CM, I consider that there are three particularly important non­

lethal research methods - tagging, biopsy and photography- that have greatly advanced

during the time of JARPA and JARPA II. These three methods are particularly useful
to scientific research in the context of conservation and management of whales

Tagging

5.4. The CM asserts (paras 5.49, 5.50; Footnotes 696, 697) that tagging minke whales

is not practicable, thus requiring lethal take. The JARPA proposai (Japan 1987, p. 43)

noted "[i]f mark and mark recapture could be available both in the low latitude

(breeding ground) and the high latitude (feeding ground), this method [mark-recapture]

would certainly produce information with the highest accuracy ever obtained by any
other methods ever adopted in the past for ascertainment of stock movement, migration,

and identification".

225.5. This older technique of mark and mark recapture has now been long overtaken by

the superior method of satellite tagging, which pursues the same research items of

ascertaining stock movement, migration and identification. In satellite tagging, marine

mammal scientists have achieved the goal posited in the JARPA proposai. Mate et al.

(2007) provided a review of the advances in satellite tags. In Figure 2, I reproduce their

data, which shows that tag longevity has increased over the last decade to the point of

achieving the standard that Japan called for in the original JARPA proposai. Although

they have not yet tried to tag minke whales, Mate et al. (2007) reported the tagging of a

calf humpback, which is about the same size as a minke whale. In particular, in

response to my question about the feasibility oftagging minke whales, Prof Bruce Mate,

Director of the Marine Mammal Institute, University of Oregon wrote to me: "[t]he tags

we are working with now can be made in a smaller length, which may be more

appropriate for minkes. We have not tagged minkes, but you will see in the attached

paper [Mate 2007] having tagged a calf humpback (even with the older and larger tag)

by mistalΠthat seemed to have worked out very weil indeed." (email correspondence

16 November 2010; Appendix 1). Other scientists have used satellite tagging in the

Antarctic to suggest revision of the management boundary for humpback whales

without recourse to lethal take (Dalla Rossa et al. 2008).

200
1ii'

~
~ 150
~
~
c:
..Q 100
0)
<U
1- 50

0
1000 1005 2IJDD

Year
Figure 2: Tag longevity has increased over the last decade to the point of achieving the

standard that Japan called for in the original JARPA proposai. Rather than embracing
these technologies, JARPA II workers have resisted them.

235.6. Lockyer (2007) noted that VHF and satellite tags and time depth recorders have in

recent years been able to collect data remotely from cetaceans on dive patterns, swim
speeds and foraging bouts. Lockyer also noted that ingested transmitters can provide

information about stomach temperature, indicative of feeding.

5.7. In February 2013, a group of scientists including Dr Nick Gales were successful

in tagging 18 Antarctic minke whales with four different types of tags. A further

19 Antarctic minke whales were also biopsied (email correspondence between Dr Gales
and myself of 20 March 2013; Appendix 2).

5.8. Although these advances may seem bath recent and modest (when compared to

the number of whales involved in the lethal take of JARPA II), they do show that these

methods are not only feasible, but successful when effort is put into their development

and implementation. This also shows clearly that, if JARPA II was a program for

'purposes of scientific research' in the context of conservation and management of
whales, it tao would have put considerable efforts into advancing tagging methodology.

Biopsy

5.9. The CM is similarly dismissive of biopsy (para 4.75, Footnote 453), which has

severa! applications including providing non-lethal information on pregnancy status as

well as pollutants (Mangel 2011, para 5.33). Recent work also suggests that the mixture

of fats in the blubber obtained by biopsy tagging can be used to assess age distributions

ofwhales (Herman et al. 2008, 2009).

5.1O.Shortly after the moratorium and barely into JARPA, Hoelzel and Amos (1988)

showed that one could apply methods of molecular genetics to small (about 2-tenths of a

gram) skin samples collected from free-ranging whales using a dart-tipped arrow, fired

from a cross-bow and retrieved with a fishing line (this was the precursor to today's

biopsy techniques). Thus, from the start of JARPA, techniques for identification of
individuals and fundamental information about populations were available, but not

developed by JARP A scientists.

5.11. In the same interview quoted in Footnote 418 of the CM, Dr. J. Zeh (University of

Washington and then Chair of SC-IWC) was asked whether a non-lethal method could

provide the same data that the lethal methods were providing. She answered "[w]ell,

24many scientists are using biopsy sampling, and that works very weil for humpback

whales. It's been a little less successful for minke whales, and I'm not sure that's

because it hasn't been tried sufficiently and the best techniques worked out, or whether

- I suspect that maybe it's somewhat more difficult to biopsy minke whales than

humpback whales". A dozen years later, JARPA II has brought us no closer to the

answer, even though as Clapham et al. (2003 p. 212) - ali of whom were members of

SC-IWC- noted "if a whale can be hit with a harpoon, the same target can just as easily

be struck with a biopsy dart". The viability of taking biopsy samples from Antarctic

minke whales in the Southern Ocean has indeed now been demonstrated, including

through the 19 animais successfully biopsied in the recent research expedition involving

Dr Nick Gales (see para 5.7 above, and email correspondence in Appendix 2).

Photography

5.12. Regarding non-lethal sampling in observation ecology, Sagarin and Pauchard

(2012, p. 98) note

Digital photography offers the opportunityto inexpensively archive many details of
organisms that are then left to go on living in the field. Genetic data can be collected
and archived based on small, non-lethaltissue samples, even in the case of endangered
species. For example, whales can be sampled both by photographing their distinct

markings and by obtaining small samples for genetics research by lancing blubber
samples when they surface.

5.13. Photography is also an important, non-lethal technique that is summarily

dismissed by Japan (CM, paras 4.62, 4.70). Nearly 25 years ago, Hoelzel et al. (1989)

used photographie methods to individually identify minke whales and then study their

individual foraging specializations. JARPA II workers, rather than embracing and

enhancing these new technologies as would happen in a program for 'purposes of

scientific research', resisted them to continue business as usual.

Summary

5.14. Changing tagging technology, biopsy methods, or photography have been and

continue to be technological revolutions that allow us to sharply break with the past

(one needs to simply think about the technological advances between 1990 and today).

A program of science should embrace these technologies rather than summarily dismiss

them to continue business as usual. JARPA II is characterized by a manifest refusai to

innovate by using new tagging technologies and biopsy methods, and instead doggedly

25pursues the continued and unnecessary killing of whales, without assessing whether

alternatives are available. Finally, in the absence of a testable hypothesis it is

impossible to lmow why non-lethal alternatives to lethal take would not be adequate.

5.15. Corkeron (2009), commenting on the lethal field work in JARPN II (the North

Pacifie version of JARPA II), characterizes the basic design of the work as "an
unsophisticated approach to investigating the foraging ecology" of baleen whales, that

the analyses of data "were simplistic" and that non-lethal studies using far fewer

resources "have produced more definitive information" (p. 305). The same is true for

JARPA IL

266. REASSESSMENT OF THE OBJECTIVES OF JARP A II

6.1. It is now possible to reassess the objectives of JARP A II (CM para 5.20) in light

of my previous report, the CM, and this supplementary report.

6.2. The first objective of JARPA II is

(1) Monitoring of the Antarctic ecosystem

(i) Monitoring of whale abundancetrends and biological parameters

(ii) Monitoring of krill abundanceand the feeding ecolgy of whales

(iiiMonitoring of the effects of contaminantson cetaceans

(iv)Monitoring of cetaceanhabitat

6.3. Reassessment: This objective is broad and general, without clear and testable

hypotheses (Mangel2011, paras 5.9-5.10). As I have explained above (paras 3.1-3.10)

monitoring in the absence of clear testable hypotheses is merely the collection of data,

and the collection of data in itself cannat be considered as being for 'purposes of

scientific research'. Fmihermore, of the elements set out in items (i) - (iv) above,

monitoring whale and krill abundance, the feeding ecology of whales, effects of

contaminants and cetacean habitat do not require lethal take. Moreover, monitoring

biological parameters is both entirely unnecessary for the conservation and management

of whales, and has proven to be unattainable to any usefullevel of accuracy.

6.4. The second objective of JARPA II is

(2) Modelling competitionamong whale species and future managementobjectives

(i) Constructing a model of competition amongwhale species

(ii)New managementobjectives includingthe restorationof the cetacean
ecosystem

- Establishingfuture managementobjectives
- Estimating surplus production(and hence allowablecatch) by species,
under some of the managementobjectives

- Contributetowards a multi-whale-speciesmanagement

6.5. Reassessment: As I have explained above (para 3.25-3.27), workers in JARPA II

are collecting insufficient data to achieve item (i), because of the focus on only a small

component of the Southern Ocean ecosystem and the lack of broad collaboration with

other scientific programs. Furthermore, workers in JARPA II themselves have

demonstrated (para 3.28) that lethal take is not required to develop the models. The

sub-items under item (ii) are both vague and not for 'purposes of scientific research' but

27an attempt to revise the RMP outside of the procedures set by the IWC. In light of the

proven inability to determine parameters such as productivity to necessary levels of
accuracy, as the SC-IWC learned through its failed attempts to implement its previous

management regime (the New Management Procedure; NMP), significant elements of

item (ii) are also highly unlikely to be achieved.

6.6. The third objective of JARPA II is

(3) Elucidationof temporalandspatialchangesin stock structure

6.7. Reassessment: As explained above (para 3.1-3.10), lacking testable hypotheses,

this objective cannot be for 'purposes of scientific research'. That is, merely monitoring

changes in stock structure over space and time, without addressing any broader question

that such monitoring is expressly aimed at addressing, does not amount to 'scientific

research'. Furthermore, given the advances that have occurred in tagging (para 5.4-

5.8), biopsy (para 5.9-5.11), and other non-lethal methods (para 5.12) over the last 25

years, there is no need for lethal talee.

6.8. The fourth objective of JARPA II is

(4) Improvingthe managementprocedurefor Antarcticminkewhale stocks

- lmprovementofMSYR (maximumsustainableyield rate) estimatesfor
Antarcticminkewhales

- Re-definitionof appropriatemanagementAreas
- Incorporationof effectsarisingfromthe inter-speciesrelationshipsamong
the whalespecies.

6.9. Reassessment: As I explained above (para 4.1-4.8), the RMP is designed to learn

about MSYR without recourse to lethal talee (except for commercial whaling) and in
fact was designed with the explicit goal of not trying to measure MSYR in the field

(something that both JARPA and the attempts of the SC-IWC to implement the NMP

demonstrated was essentially impossible). Re-definition of the management areas is

again not a question for 'pm·poses of scientific research'. In any event, as explained

above, it does not require lethal talee because of developments in tagging and biopsy.

The last item ('incorporation of effects arising from the inter-species relationships

among the whale species') is extremely vague, but suggests movement towards a

completely new kind of multi-species management procedure. To the extent that this

seeks to look at interactions between whale species, it also suffers from the fatal flaw of

28actually only studying one species (see para 6.5 above). Again, this item is not for

'purposes of scientific research'.

297. CONCLUSION

7.1. JARPA II follows in the tradition of JARPA as a program for the collection of

lethal data with the assertion that it will somehow inform the conservation and

management of whales in the Southern Ocean. However, JARPA II Jackshypotheses, is
inconsistent about the determination of sample sizes, does not connect models and data

appropriately, and generally Jacksbona fide (or, in many respects, any) peer-review.

7.2. The reason that JARPA/JARPA II publications are generally irrelevant to their

own stated objectives is clear: JARPA II is not a program for 'purposes of scientific

research' in the context of conservation and management of whales. Rather it is a

program of data collection. The data collected by lethal means have, after 26 years, not

contributed to the RMP and it is unlikely that they will so contribute in the future.

Furthermore, any data that are in any way relevant to the conservation and management

ofwhales can be collected by non-lethal means.

7.3. Burnett (2012) reviewed the history of failed attempts at science in the
1
20 h century that lead to the destruction of the great whales, particularly the now
discredited Discovery Committee, which was a series of ship-based and land-based

expeditions in the Southern Ocean, with a focus on whales, between 1925 and the

1940s. The core of this work consisted of: (i) flensing platform studies of the anatomy

and reproductive physiology of whales; and (ii) open-ocean 'marking' expeditions in

which whales were shot with a numbered steel dart that was recovered in the course of

commercial whaling ('marking surveys'). Burnett notes that the Discovery work

"eventually made whalemen of scientists" (p. 30).

7.4. JARPA and JARPA II, in which attempts are made to have commercial whaling

look like science (the converse of the Discovery effect), share many of the

characteristics of the work of the Discovery Committee. Based on Burnett's
descriptions of the Discove1y Committee 'sactivities and the early years of the SC-IWC,

I draw out the following characteristics they share with JARPA/JARPA II: (i) the

conflation of 'science' and whaling; 10 (ii) publication outside of general peer-reviewed

outlets; 11 (iii) vagueness about how the data collection would contribute to

10
Burnett(2012), p. 29, 174.
IlIbid. p. 138.

30management; 12 (iv) confusion about the collection of data and the process of science; 13

(v) data collection that could not contribute to the conservation and management of
14
whales and not changing the goals in the face of criticism; (vi) the tradition of no
15
matter what, keep going in the same direction; and (vii) ignoring alternative, non­

lethal methods. 16

7.5. My conclusion remains the same in light of the CM: JARPA II is an activity that

collects data in the Southern Ocean. However, it is not a program for 'purposes of

scientific research' in the context of conservation and management of whales.

12Ibid, p. 173.
13Ibid, p. 430.
14
15Ibid, p. 448, 476.
Ibid, p. 496.
16Ibid, p. 399.

31Literature Cited

Angier, N. 2007. The Canon. Houghton Mifflin Books, New York, NY

Beddington, J.R. and May, R.M. 1982 'The harvesting ofincreasing species in a natural
ecosystem'. Scientific American 247(5):62-69

Burnett, D. G. 2012. The Sounding of the Whale. The University of Chicago Press,
Chicago, IL

Casti, J.L. 1989. Paradigms Lost. Images of Man in the Mirror of Science. William
Morrow and Company, New York

Chalmers, A.F. 1999. What is This Thing Called Science? 3rd Edition. University of
Queensland Press, Queensland, AU

Clapham, P.J., Berggren, P., Childerhouse, S., Friday, N.A., Kasuya, T., Kell, L., Kock,
K-H, ManzanillaO-Naim, S., Notbartolo di Sciara, G., Perrin, W.F., Read, A.J., Reeves,
R.R., Rogan, E., Rojas-Bracho, L., Smith, T.D., Stachowitsch, M., Taylor, B.L., Thiele,
D., Wade, P.R., and Brownell Jr., R.L. 2003. 'Whaling as science'. BioScience 53:210-
212

Cooke, J. G. 1995. 'The International Whaling Commission's Revised Management
Procedure as an example of a new approach to fishery management. Developments in
Marine Biology' 4 (Whales, seals, fish and man [AS Blix, L Wall0e, and 0. Ulltang,
editors], Elsevier, Amsterdam): 646-657

Corkeron, P.J. 2009. 'Reconsidering the science of scientific whaling'. Marine

Ecology Progress Series 375:305-309

Cromer, A. 1993. Uncommon Sense. The Heretical Nature of Science. Oxford
University Press, Oxford and New York

Dalla Rossa, L., Secchi, E.R., Mai, Y.G., Zerbini, A.N., and Heidi-Jorgensen, M.P.
2008. 'Movements of satellite-monitored humpback whales on their feeding ground

along the Antarctic Peninsula'. Polar Biology 31:771-781

Foster, K.R. and Huber, P.W. 1999. Judging Science. Scientific Knowledge and the
Federal Courts. MIT Press, Cambridge, MA

Oiere, R.N. 1997. Understanding Scientific Reasoning. 4th edition. Dryden Press,

Orlando, FL

Giere, R.N. 2006. Scientific Perspectivism. The University of Chicago Press, Chicago,
IL

Herman, D.P., Matkin, C.O., Ylitalo, G.M., Durban, J.W., Hanson, M.B., Dahlheim,
M.E., J.M. Straley, Wade, P.R., Tilbury, K.L., Boyer, R.H., Pearce, R.W., and Krahn,

M.H. 2008. 'Assessing age distributions of killer whale Orcinus orca populations from
the composition of endogenous fatty acids in their outer blubber layers'. Marine
Ecology Progress Series 372:289-302

32Herman, D.P., Ylitalo, G.M., Rabbins, J., Straley, J.M., Gabriele, C.M., Clapham, P.J.,

Boyer, R.H., Tilbury, K.L., Pearce, R.W., and Krahn, M.H. 2009. 'Age determination of
humpback whales Megaptera novaeangliae through blubber fatty acid compositions of
biopsy samples'. Marine Ecology Progress Series 392:277-293

Hoelzel, A.R. and Amos, W. 1988. 'DNA fingerprinting and 'scientific' whaling'.
Nature 333:305

Hoelzel, A.R., Dorsey, EM., and Stern, J. 1989. 'The foraging specializations of
individual minke whales'. Animal Behaviour 38:786-794

IWC. 1994. 'The Revised Management Procedure for baleen whales'. Reports of the
International Whaling Commission 44 (Annex H):145-152

IWC. 1999. 'The Revised Management Procedure for baleen whales'. Journal of
Cetacean Research and Management 1 (Supplement, Annex N): 251-258

Japan. 1987. 'The Program for Research on the Southern Hemisphere Minke Whale and
for Preliminary Research on the Marine Ecosystem in the Antarctic', Japan, March
1987, IWC Doc. SC/39/04

Johnson, D.H. 2001. 'Validating and evaluating models'. p. 105-119 in [T.M. Shenk
and A.B., Franklin, editors] Modeling in Natural Resource Management. Development,
Interpretation, and Application. Island Press, Washington, DC

Karban, R., Huntzinger, M. 2006. How ta Do Ecology - A Concise Handbook.
Princeton University Press, Princeton, NJ.

Kirkwood, G.P. 1991. 'Comprehensive assessment ofwhale stocks. Progress Report on
Development of Revised Management Procedures'. Reports of the International
Whaling Commission 41:213-218

Kirkwood, G.P. 1992. 'Background to the development of Revised Management

Procedures'. Reports of the International Whaling Commission 42:236-243

Kleiber,M. 1947. 'Body size and metabolism'. Physiological Reviews 27:511-541

Legendre, L. 2004. Scientific Research and Discovery: Process, Consequences and
Practice. International Ecology Institute, Oldendorf/Luhe, Germany

Lockyer, C. 2007. 'Ail creatures great and smaller: a study in cetacean !ife history

energetics'. Journalof the Marine Biological Association of the UK 87:1035-1045

Ludwig, D. and Walters, C.J. 1985. 'Are age-structured models appropriate for catch­
effort data?' Canadian Journal of Fisheries and Aquatic Sciences 42:1066-1072

Mange!, M. 201O. 'Scientific inference and experiment in Ecosystem Based Fishery
Management, with application to Steller sea lions in the Bering Sea and Western Gulf of

Alaska'. Marine Policy 34:836-843

Mange!, M. 2011. 'An Assessment of Japanese Whale Research Programs Under
Special Permit in the Antarctic (JARPA, JARPA II) as Programs for Purposes of

33Scientific Research in the Context of Conservation and Management of Whales'.
Appendix 2, Memorial of Australia

Mate, B., Mesecar, R. and Lagerquist, B. 2007. 'The evolution of satellite-monitored
radio tags for large whales: One laboratory's experience'. Deep-Sea Research II 54:224-

227

Ohsumi, S. 1995. 'The Necessity of Employing Lethal Methods in the Study of Whale
Resources'. From Research on Whales, ICR, 1995). Available at
http://luna.pos.to/whale/icr rw oh.html

Oreskes, N. 2003. 'The Role of Quantitative Models in Science'. p. 13-31 in Canham,
C.D., Cole, J.J. and Lauenroth, W.K. Models in Ecosystem Science. Princeton
University Press, Princeton, NJ

Oreskes, N. and Conway, E.M. 2010. Merchants ofDoubt. Bloomsbury Books, New
York

Peters, R.H. 1991. A Critique for Ecology. Cambridge University Press, Cambridge,
UK

Pigliucci, M. 201O. Nonsense on Stilts. How to Tell Science from Bunk. University of
Chicago Press, Chicago, IL

Platt, J.R. 1964. 'Strong inference'. Science 146:347-353

Rastetter, E.B.. 2003. 'The collision of hypotheses: What can be learned from
comparisons of ecosystem models?' p. 211-224 in Canham, C.D., Cole, J.J. and
Lauenroth, W.K. Models in Ecosystem Science. Princeton University Press, Princeton,
NJ

Rigler, F.H., Peters, R.H. 1995. Science and Limnology. Ecology Institute,
Oldendorf/Luhe, Germany

Sagarin, R. and Pauchard, A. 2012. Observation and Ecology. Broadening the Scope of
Science to Understand a Complex World. Island Press, Washington, DC

Shermer, M. 2001. The Borderlands of Science. Where Sense Meets Nonsense. Oxford
University Press, Oxford, UK

Smith A.D.M., Sainsbury K. and Stevens, R. 1999. 'Implementing effective fisheries­
management systems - management strategy evaluation and the Australian partnership
approach'. !CES Journal of Marine Science 56:967-79.

Sremba, A.L., Hancock-Hanser, B., Branch, T.A., LeDuc, R.L. and Baker, C.S. 2012.
'Circumpolar Diversity and Geographie Differentiation of mtDNA in the Critically
Endangered Antarctic Blue Whale (Balaenoptera musculus intermedia)'. FLoS One
7(3): e32579. doi:10.1371/journal.pone.0032579

Tamura, T. and Konishi, K. 2006. 'Food habit and prey consumption of Antarctic minke

whale Balaenoptera bonaerensis in JARPA research area'. SC/D06/J18, presented to

34the Intersessional Workshop to Review Data and Results from Special Permit Research

on Minke Whales in the Antarctic. CM Footnote 553

Tamura, T. and Konishi, K. 2009. 'Feeding habits and prey consumption of Antarctic
minke whale (Balaenoptera bonaerensis) in the Southern Ocean'. Journal ofNorthwest
Atlantic Fishery Science 42:13-135

Taper, M.L. and Leie, S.R. 2004. 'Dynamical models as paths to evidence in ecology'.

p. 275-297 in Taper, M.L. and Leie, S.R. editors, The Nature of Scientific Evidence,
University of Chicago Press, Chicago, IL

van Fraassen, B.C. 2010. Scientific Representation. Oxford University Press, Oxford,
UK.

Weinberg, A.M. 1972. 'Science and trans-science'. Minerva 10:209-222

Ziman, J. 1991. Reliable Knowledge. An Exploration of the Grounds for Belief in
Science. Cambridge University Press, Cambridge, UK

35Appendix 1: Correspondence with Prof Bruce Mate Concerning Tagging Whales

16 Nov 2010

Bruce Mate <[email protected]>
11/16/10

to Marc

Hi Marc,

This should help you get started. The tags we are working with now can be made in a smaller
length, which may be more appropriate for minkes.We have not tagged minkes, but you will see
in the attached paper having tagged a calf humpback (even with the older and larger tag) by
mistake that seemed to have worked out very weil indeed.

1guess 1have to encourage Mike to be in touch so we can move forward with this thoughts.

Bruce

Bruce Mate

Director, Marine Mammallnstitute
Oregon State University
Hatfield Marine Science Center
2030 SE Marine Science Drive
Newport, OR 97365

On Nov 16, 2010, at 7:14AM, Marc Mange! wrote:

Hi Bruce

1writing something in which 1would like to provide an assessment of
putting simple, long-lived tags (e.g. giving, date, whale identifier,
and location) on minke whales. Mike Fedak suggested that 1could contact
you for information on what you have done on some of the larger whales

and your opinion about doing the sameon minkes. If you could send a few
papers that would be terrifie.

Mike says hello, and also that he still would like to try attaching some
of his tags using your attachment methodology.

Thanks in advance

Marc

Marc Mange!
Distinguished Professer, Applied Mathematics and Statistics
Jack Baskin Endowed Chair, Technology and Information Management

Director, Center for Stock Assessment Research

36Appendix 2: Correspondence with Dr Nick Gales Concerning Tagging Whales

20 March 2013

from: Nick Gales <[email protected]>
to: "[email protected]" <[email protected]>
date: Tue, Mar 19, 2013 at 9:14PM
subject: Minke whale tagging [sec=unclassified]

mailed-by: aad.gov.au

Dear Mark,
Further to our discussion, in thismail! provide you with details of the tagging and and other

work we conducted with minke whales in February 2013.
As you know, up until this summer no Antarctic minke whales had been tagged. During
February this year 1spent about 10 days on board the NSF Research Vessel Point Sur working
with a group of US scientists led byDr Ari Friedlaender. The project was supported by the US
Antarctic Program and formed part of a collaborative research project of the IWC's Southern

Ocean Research Partnership.

The aim of the voyage was to deploy location-only satellite tags and suction-cup mounted data­
Joggingtags on humpback and minke whales off the Western Antarctic Peninsula. These tags

had been successfully deployed on humpback whales as part of this project during earlier
seasons. This was the first season that minke whales were targeted.

1successfully deployed 10 location only, blubber implantable satellite tags on minke whales.
The tags were deployed from our standard air-powered gun at ranges from 4-1Om.Ali

deployments were from a small rigid-hulled inflatable boat. The tags themselves were a
shortened modification of implantable tags we routinely deploy on larger whales such as
humpback, blue and southern right whales. They are similar to the type of tags that Bruce Mate
deploys.

Drs Robert Pitman and John Durban were also present on Point Sur during our voyage. They
were studying killer whales and were deploying a type of dorsal-fin mounted tag that is fired
onto the side of the dorsal fin and held in place by two pins which penetrate the cartilage. Dr
Durban deployed 3 of these types of tags on minke whales as part of our work. Two of these

tags included depth sensors, so also transmitted dive summaries along with position
information. The tags were deployed by a cross-bow.

Drs Pitman and Durban had spent sorne time earlier in the summer in the Ross Sea and had
successfully deployed three of these tags on minke whales at that location as weil.

Two suction eup mounted tags were also deployed on minke whales. These tags are designed
to stay on for a short time (<1 day) and provide dense data on the three dimensional
movements of the whale during that period. These data are of sufficient precision to determine

when and how often the whale Jungefeeds during dives. Contemporaneous with the
deployment of these two tags we conducted surveys of the prey-field using scientific
echosounders.
Nineteen minke whale biopsies were also collected as part of this work, but many more samples
wouid have been possible if we had assigned more time to this task.

Thus, in total, 18 minke whales were tagged (3 in the Ross Sea, 15 in the Western Antarctic
Peninsula) with 4 different types of tags. A further 19 were biopsied.

We found that when the whales were in small groups (the larger the better, but generally >2)
and were either feeding or exhibiting some social behaviour, they were relatively easy to
approach in a small boat if we took our time. The habitat type was generally open water with ice
floes and the weather conditions were calm. These type of conditions and habitat can be
encountered throughout Antarctica, and so 1believe this work indicates that tagging is practical

in most Antarctic areas when the wind is low. During research voyages focusing of humpback
and blue whales in East Antarctica, we have shown that launching small beats for tagging work

37on the high seas is practical and successful.

The data from the work this summer will be presented to the IWC during the June meeting on
Jeju Island, South Korea and will then be published in peer reviewed papers.

With best regards,

Nick

Nick Gales

Chief Scientist
Australian Antarctic Program

38