PV A Critiques

262
Conservation Biology, Pages 262263

Volume 16, No. 1, February 2002

Critiques of PVA Ask the Wrong Questions: Throwing
the Heuristic Baby Out with the Numerical Bath Water

BARRY W. BROOK,* MARK A. BURGMAN, H. RESIT AKAKAYA,
JULIAN J. OGRADY, AND RICHARD FRANKHAM

*Key Centre for Tropical Wildlife Management, Northern Territory University, Darwin, Northern Territory 0909, Australia
School of Botany, University of Melbourne, Parkville, Victoria 3052, Australia
Applied Biomathematics, 100 North Country Road, Setauket, NY 11733, U.S.A.
Key Centre for Biodiversity and Bioresources, Macquarie University, New South Wales 2109, Australia

The analyses of Ellner et al. (2002 [this issue]), like those
of Ludwig (1999) and Fieberg and Ellner (2000), suggest
that although the predictions of population viability
analyses (PVA) may be unbiased, the estimates of extinc-
tion risk are usually too imprecise to be worthwhile. Ell-
ner et al. correctly reconcile the superficial discrepan-
cies between Ludwig (1999) and Brook et al. (2000). But
they then allude to the recent review by Coulson et al.
(2001) that restates the widely held view (see also Har-
court 1995; Beissinger & Westphal 1998) that in circum-
stances where data are sparse or of low quality (com-
monly the case for threatened species), PVAs have little
useful predictive value and should be dispensed with in
favor of alternative methods. The trouble is that none
of these authors have specified why these alternatives
would be superior to PVA. It is our view that even when
PVAs perform poorly against some vaguely defined abso-
lute standard, they still perform better than alternatives
that are even more vague, are less able to deal with un-
certainty, are considerably less transparent in their reli-
ability, and do not use all the available information.
Conservation biologists are aware that ours is a crisis
discipline (Soul 1985) and that decisions must be made
quickly and in the face of incomplete data. In such
cases, any method of assessment will be uncertain. Al-
though critiques such as those of Ellner et al. are entirely
appropriate, there is a growing body of opinion (e.g.,
Coulson et al. 2001) that recommends dropping PVAs,
without suggesting an alternative. At the moment, PVAs
are the best tool available in many circumstances, and al-
though the current debate serves to emphasize the point
that they work better for some problems than for others,
the results of Brook et al. (2000) demonstrate that PVAs
have value even when uncertainty exists. The options in
decision-making are to use the best scientific informa-
tion and tools available (e.g., PVA). The alternatives are
subjective decisions made by humans, which are notori-
ously inaccurate (Zeckhauser & Viscusi 1990), or deci-
sions made by bureaucrats and politicians with little or
no scientific input.
We agree with Ellner et al. that in any single applica-
tion PVA predictions may be unreliable, even over short
time frames (see Chapman et al. 2001). This reiterates a
point made earlier by Taylor (1995) that there is a statis-
tical distribution associated with any prediction and that
the confidence intervals may be broad. But PVA pro-
vides substantial information content that makes it a use-
ful tool in applications to single species. The question of
how much information is sufficient to provide a basis for
reliable or useful predictions depends largely on the
question being asked. For individual cases, the only ob-
jective way to assess whether data quality is high
enough is to systematically incorporate uncertainties
and evaluate whether the results are too uncertain to ad-
dress the question at hand (Akakaya & Raphael 1998;
Caswell 2001). Even with uncertain data, or with data
from related species (Gaillard et al. 2000), one might be
able to conclude that extinction risk is very high (or neg-
ligible), which can be a useful result for management. In
fact, the Red List system of the World Conservation
Union operates on this basis (Mace & Hudson 1999).
Ellner et al. describe PVA as one of a variety of possi-
ble tools. But the three alternatives Ellner et al. men-
tionhistorical and predicted future habitat loss, recent
population trends, and genetic considerationsare actu-
ally only types of information that may be used in a PVA.
Akakaya and Sjgren-Gulve (2000) describe the limita-
tions of other decision-support tools. Reserve-selection

email barry.brook@ntu.edu.au
Paper submitted August 30, 2001; revised manuscript accepted Sep-
tember 19, 2001.
Conservation Biology
Volume 16, No. 1, February 2002

Brook et al. Critiques of PVA Ask Wrong Questions

263

algorithms have a narrow scope (reserve design) and use
only occurrence data (in space). Habitat models and gap
analysis ignore demographic data on population struc-
ture, dynamics, and trends. Rule-based and score-based
methods have limited scope (species ranking or prioriti-
zation) but can use PVA results (e.g., Red List). Estimat-
ing extinction probability from sighting data is retro-
spective, running the risk that the species is already
extinct), and uses only occurrence data (in time). Land-
scape indices such as fractal dimension ignore demo-
graphic data and have questionable relevance to viability
of species. Ecosystem-based methods are vague, not
well-developed, and not transparent.
Although each of these methods is useful for particular
objectives, none can be considered a replacement for
PVA because none uses all available information, most
cannot assess management or impact scenarios, and many
do not incorporate uncertainties. Although we agree that
we should use whatever works, it is the responsibility of
the users of other tools or methodologies to subject them
to at least the level of scrutiny PVA has been subject to be-
fore they are recommended as alternatives.
Although it is often true that even the best available
science may be unable to provide the level of predict-
ability and accuracy we might wish (Ellner et al. 2002),
it is better to use whatever guidance scientific evidence
has to offer than to dispense with it simply because it is
not as precise as we might wish. The latter alternative
throws away knowledge. A final decision on how and
when to act to protect or conserve a species always de-
pends on human judgment. But we believe that PVA
should not be ignored when it helps to set priorities, en-
sures the internal consistency of arguments, and elimi-
nates semantic ambiguities that plague subjective and in-
tuitive interpretations of evidence. Population viability
analyses are applied in a wide variety of contexts, and
numerical precision in forecasts is important in just
some cases ( Burgman & Possingham 2000). In many
cases the value of a PVA is determined by the clarity it
brings to a problem, and predictive reliability is rela-
tively unimportant. Blanket advice that PVAs should be
dropped will risk throwing the heuristic baby out with
the numerical bath water.
The results of a PVA are just one factor in any social
and political decision-making context and should be a
necessary precursor to good judgment. We advocate cau-
tious interpretation of predictions, conditioned on their
reliability, and contend that in the absence of any better
alternative, the heuristic advantages of PVA, together
with its (admittedly limited) predictive reliability, make
it by far the best conservation management tool we have.

Literature Cited

Akakaya, H. R., and M. G. Raphael. 1998. Assessing human impact de-
spite uncertainty: viability of the Northern Spotted Owl metapopu-
lation in the northwestern USA. Biodiversity & Conservation

7:

875894.
Akakaya, H. R., and P. Sjgren-Gulve. 2000. Population viability analy-
sis in conservation planning: an overview. Ecological Bulletins

48:

921.
Beissinger, S. R., and M. I. Westphal. 1998. On the use of demographic
models of population viability in endangered species management.
Journal of Wildlife Management

62:

821841.
Brook, B. W., J. J. OGrady, A. P. Chapman, M. A. Burgman, H. R.
Akakaya, and R. Frankham. 2000. Predictive accuracy of popula-
tion viability analysis in conservation biology. Nature

404:

385387.
Burgman, M., and H. P. Possingham. 2000. Population viability analysis
for conservation: the good, the bad and the undescribed. Pages 97
112 in A. G. Young and G. M. Clarke, editors. Genetics, demogra-
phy and viability of fragmented populations. Cambridge University
Press, London.
Caswell, H. 2001. Matrix population models: construction, analysis,
and interpretation. Sinauer Associates, Sunderland, Massachusetts.
Chapman, A. P., B. W. Brook, T. H. Clutton-Brock, B. T. Grenfell, and
R. Frankham. 2001. Population viability analysis on a cycling popu-
lation: a cautionary tale. Biological Conservation

97:

6169.
Coulson, T., G. M. Mace, E. Hudson, and H. Possingham. 2001. The use
and abuse of population viability analysis. Trends in Ecology & Evo-
lution

16:

219221.
Ellner, S. P., J. Fieberg, D. Ludwig, and C. Wilcox. 2002. Precision of
population viability analysis. Conservation Biology

16:

258261.
Fieberg, J., and S. P. Ellner. 2000. When is it meaningful to estimate an
extinction probability? Ecology

81:

20402047.
Gaillard, J. M., M. Festa-Bianchet, N. G. Yoccoz, A. Loison, and C.
Toigo. 2000. Temporal variation in fitness components and popula-
tion dynamics of large herbivores. Annual Review of Ecology and
Systematics

31:

367393.
Harcourt, A. H. 1995. Population viability estimates: theory and prac-
tice for a wild gorilla population. Conservation Biology

9:

134142.
Ludwig, D. 1999. Is it meaningful to estimate a probability of extinc-
tion? Ecology

80:

298310.
Mace, G. M., and E. J. Hudson. 1999. Attitudes toward sustainability
and extinction. Conservation Biology

13:

242246.
Soul, M. E. 1985. What is conservation biology? BioScience

35:

727734.
Taylor, B. L. 1995. The reliability of using population viability analysis
for risk classification of species. Conservation Biology

9:

551558.
Zeckhauser, R. J., and W. K. Viscusi. 1990. Risk within reason. Science

248:

559564.

PV A Critiques

Uploaded by

Copyright:

Available Formats

PV A Critiques

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PV A Critiques

Uploaded by

Copyright:

Available Formats

262

Conservation Biology, Pages 262263

You might also like