Wild Ideas: an online exploration of the wild

The Calyx: Wild Sexuality The Commons: Wild Politics Return to Wild Ideas home page

In The Forest:

Book Reviews
Web Reviews

Stay informed — join WildNews, our announcement list:

E-mail Address:


You are here: Wild Ideas > Forest > Library >

Quantitative Questions
in Biological Conservation:
The Case of Amphibian Declines

Copyright 1996 by Lynna Landstreet. Please don't reproduce or redistribute this paper without asking me first. I'll probably say yes if you do ask, but I do like to know where my work is going.

If you wish to cite this paper in your own work, the following format is suggested: Landstreet, Lynna. "Quantitative Questions in Biological Conservation: The Case of Amphibian Declines." Unpublished paper, available electronically from Wild Ideas (http://www.wildideas.net).

This is better than using the specific URL of this document, because I may reorganize the directory structure of this site from time to time, causing the addresses of specific pages to change.

Despite the interdisciplinary orientation of York University's Faculty of Environmental Studies (FES), and the multifaceted nature of the problem of biological conservation, the approaches taken to that question within FES tend to be limited in range to the sociopolitical and cultural dimensions of what has been termed the biodiversity crisis. In addition, the predominant inquiry paradigm used appears to be quite firmly qualitative.

But regardless of the theoretical approach taken, the actual practice of biological conservation cannot get very far without encountering the need for quantitative research. By its very nature, biodiversity cannot be assessed in a purely qualitative fashion, no matter how much some theorists may wish it could. Ultimately, the task of conserving biological diversity must address the question of precisely which species, which ecosystems, are at greatest risk, the nature and level of risk they face, and how that risk can best be minimized.

None of these questions can be adequately addressed without recourse to quantitative methods -- specifically, to science. As much as some environmentalists may feel that science has played a role in distancing humans from the rest of nature or consolidating human control over other species, the project of undoing the damage we have done will ultimately rely heavily on science. Postmodernism has never restored a degraded streambed; Marxist theory will not bring back the Eastern Cougar.

Interdisciplinarity is still important; not only must science be informed by an understanding of the cultural and political dynamics underlying our relationship with the natural world, but the science required must in and of itself be interdisciplinary. Conservation biology, the relatively new field that has arisen in response to these concerns, combines elements of ecology, zoology, botany, genetics, biogeography and other scientific disciplines in the project of assessing biodiversity and the threats it faces and responding to these threats.

In short, any effective attempt to conserve biological diversity must be based on hard biological data, and that inevitably means that quantitative research methods are of paramount importance. But in the course of creating or consuming quantitative data, many questions come up, some of which are of relevance in any field, and some of which are relatively unique to the "crisis discipline" (Soulé 1986) of conservation biology.

Any number of examples might serve to illustrate these problems, but the one I will focus on herein is the controversy surrounding the question of declining amphibian populations. A plethora of stories in both the scientific and popular press over the last several years have raised the spectre of a worldwide amphibian decline. Anecdotal evidence and short-term, single-species studies are plentiful, but the long-range, wide-area studies that are required to establish the presence or absence of a global trend are few and far between.

In this paper, I propose to review the literature surrounding the question of amphibian declines with a view toward the statistical and research design questions it raises. The three major areas I propose to explore are:

  1. How can we assess, with any degree of certainty, whether an observed decline in a given population is of anthropogenic origin, or simply part of a natural fluctuation pattern that has as yet gone unrecorded? Many populations fluctuate widely, either at random or in a cyclic pattern. How, precisely, do we define a "decline"?
  2. In many instances where long-term studies have been done and population dynamics are to some extent understood, significant declines have been recorded. But this is true across all taxa. Wilson (1992) and others have stated that we are in the midst of the biggest extinction spasm since the demise of the dinosaurs. How significant does the decline of a particular class of creatures -- in this case, amphibians -- have to be in order to stand out from the general crisis in biodiversity?
  3. Both of these questions imply a third: just how certain do we need to be that a particular type of organism is declining due to human impacts, and at a rate severe enough to stand out as distinct, before we sound the alarm? What degree of precision is required? Science has traditionally preferred to err on the side of conservatism, risking a type II error -- accepting the null hypothesis when it should have been rejected -- rather than a type I -- rejecting it when it should have been accepted. Given the stakes in conservation biology -- potentially, the termination of entire evolutionary lines, if we assume no emergency where one in fact exists -- should this preference still stand?

All three of these questions have implications for the broader field of conservation biology, beyond the immediate question of amphibian declines, and yet the amphibian question provides a highly suitable context for their exploration. It is my hope that these questions, and the findings I draw from the literature on amphibian declines, will be useful in addressing these broader concerns as well.

First, some background on the issue of declining amphibian populations is in order. The various anecdotal reports of declines and local extinctions had been accumulating for some time, but seemed to accelerate in the late 1980's. At the First World Congress on Herpetology, held in Canterbury, England, in the summer of 1989, a number of scientists compared notes and found that they had all observed marked declines in the populations of the various amphibian species they had been studying. In response, a two-day workshop was held in Irvine, California, in February, 1990, and attended by approximately 40 scientists with specialties ranging from herpetology to climatology to pathology. The workshop attendees agreed that anecdotal evidence, at least, suggested that a variety of amphibian species around the world were declining dramatically in population. (Phillips 1990)

One of the outcomes of the Irvine workshop was the establishment of the Declining Amphibian Populations Task Force (DAPTF) by the International Union for the Conservation of Nature (IUCN)'s Species Survival Commission. Coordinated by the Ecology and Conservation Research Group at the Open University in Milton Keynes, UK, DAPTF is an international association of researchers working in the area of amphibian declines.

In the five years since its origin, it has grown to include working groups in areas as diverse as Canada, Belize, Denmark, Sri Lanka, Turkey and the Commonwealth of Independent States, as well as specialized working groups on different issues such as climatic and atmospheric change, chemical contaminants, and invasive species, and publishes a quarterly newsletter, Froglog. (DAPTF 1994a-1995d) Other results of the task force's work have been the North American Amphibian Monitoring Program, a project involving scientists and student volunteers from across the continent in monitoring a wide variety of amphibian populations with the aim of assessing long-term population trends, and a book, Measuring and Monitoring Biological Diversity: Standards Methods for Amphibians. (Heyer et al. 1994) However, despite all this activity, there is still considerable controversy within the herpetological field as to the existence and precise nature of the amphibian decline phenomenon.

That many individual species are declining is beyond doubt. Blaustein et al. (1994a) reviewed 16 long-term studies of particular amphibian species and found that six -- primarily toads -- were declining significantly; one of these had experienced the extinction of six local populations. Of the remaining ten, three appeared stable, six -- primarily salamanders -- were fluctuating to the point where a definite upward or downward trend was impossible to pinpoint, and one appeared to be increasing, although this could be explained by a local hurricane having provided an increase in retreat sites and decrease in predators. Phillips (1990) lists seven other frog and toad species that have been recorded as declining or becoming locally extinct.

However, there is nothing intrinsically unnatural about population declines, or even extinctions. Evolution, by its very nature, involves ongoing cycles of speciation and extinction; the question is whether these particular declines and extinctions are unusual. Do they exceed the level of fluctuation that might naturally be expected of amphibian populations? Do they represent a global phenomenon? Can they be traced to anthropogenic causes?

Amphibian populations can fluctuate wildly under entirely natural circumstances; in their 11-year study of three salamander species and one frog species at Rainbow Bay in South Carolina, Pechmann et al. (1991) found that female breeding population sizes fluctuated by over three orders of magnitude among years, and juvenile recruitment by over five, with no overall trend toward increase or decrease being discernible. Given this level of variation, attempting to assess the uniqueness of any particular species decline can be very problematic.

One might hypothesize that this very variability in numbers might make amphibians more vulnerable than other taxa to extinction, but in fact the relationship between population fluctuations and extinction is far from straightforward. In many species, variability in numbers increases with population density, and hence variability may be inversely rather than positively correlated with the likelihood of extinction. (Blaustein et al. 1994a)

In addition, local extinctions are not necessarily a disaster. There is an increasing tendency to view species in terms of metapopulations, particularly where numerous small local populations exist due to either human-caused habitat fragmentation or stochastic processes. Rather than focusing on the individual populations, researchers are tending to look at the broader picture and the cycles of extinction and recolonization of local populations that maintain the metapopulation of the species over a broader area. This has been termed the nonequilibrium model of community dynamics, and it functions as yet another complicating factor in attempting to assess population growth or decline patterns. Merely studying individual populations, as has traditionally been done in the past, may not tell the whole story. (Blaustein et al. 1994a)

There are also statistical problems to be addressed in attempting to tell whether a particular population trend is "unnatural." Pechmann and Wilbur (1994) raise the question of serial autocorrelation in environmental factors. Each new year is not a clean slate with an equal chance of seeing a population increase or decrease; climatic patterns often follow discernible multi-year cycles. This applies to biological factors as well, they point out: a year of heavy predation on a particular prey population by a particular predator is likely to be followed by another such year (as the predator population increases) and so on until the predator population crashes due to lack of prey.

In addition to all this, there are problems involved simply in attempting to count some species in the first place. Salamanders, in particular, are known to burrow underground during drought conditions, and according to some researchers may stay there for years. Others, however, dispute this, arguing that since salamanders feed almost exclusively above ground, they would starve if they remained below ground for as long as has been claimed. A fierce debate has arisen amongst herpetologists on this topic, particularly as regards the effects of clearcut logging on salamanders, and no clear consensus has emerged. The precise mechanics of salamanders' underground dormancy are not known, and are very difficult to study. (Cohn 1994) It is entirely possible that many of the perceived fluctuations in salamander population may in fact be due to large numbers hiding, not dying off. Crump et al. (1992) have observed similar behaviour in the golden toad, and have hypothesized that its apparent extinction may in fact merely indicate a dormant period in response to unfavourable climatic conditions.

Pechmann & Wilbur (1994) discuss the complexity of temporal and spatial scales, and the problem of deciding over just what span of space and time population dynamics should be monitored in order to assess whether declines are occurring and whether their extent is unprecedented, and arrive at the following seven questions:

  1. Is the proper null hypothesis that amphibian populations are declining, or that they are not declining?
  2. In the absence of adequate long-term census data to estimate natural variation in population sizes, how large a decline should be considered unnatural, or what is an appropriate "effect size" to use?
  3. Over what scales of time should one expect changes in population sizes to average zero under the null hypothesis of no change, and over what temporal scales does important serial autocorrelation in population sizes or trends occur?
  4. Over what spatial scales should one expect changes in population or metapopulation sizes to average zero, and what is the relative importance of immigration and emigration vis-à-vis births and deaths?
  5. At what scales of space and time should naturally occurring extinctions be expected for amphibians, and over what scales should recolonization equal extinction?
  6. How sensitive is total population growth rate to variation in each individual component, such as fecundity, egg survival and adult survival?
  7. Given known natural variations, how many amphibian populations and species within a region or around the world should one expect to be declining at any given time, and what spatial and temporal covariances should be expected?

Implied in several of these questions is the matter of context. Given that, as stated earlier, we are in the midst of an extinction crisis of historic proportions, the question of amphibian declines should perhaps not be evaluated in terms of whether amphibians as a taxon are declining at all, but whether they are declining any more than other taxa. The various lists of species at risk produced by environmental organizations around the world list organisms of every sort; if amphibians, alone among all taxa, could be shown not to be declining as a result of human impacts, it would surely be cause for amazement. So the second major question that needs to be addressed is: if amphibians (in at least some instances) are declining precipitously, is this a distinct phenomenon, or merely part of the overall biodiversity crisis?

In a sense, this is a false question, because any diminishment of biodiversity is intrinsically part of the biodiversity crisis. So perhaps a better way to address it would be: what are the factors, if any, that make this particular part of the biodiversity crisis unique?

One is the fact that many of the observed declines have taken place in apparently pristine areas, where there are no obvious human impacts. The gastric brooding frog (Rheobatrachus silus) of Australia, the golden toad (Bufo periglenes) of Costa Rica, and several species previously found in the United States' Pacific Northwest region have all declined or disappeared within seemingly undisturbed environments, raising the possibility of an as yet unidentified global influence. Increased ultraviolet radiation and global warming have been frequently hypothesized as the cause of this. (Barinaga 1990; Blaustein and Wake 1990)

Acid deposition is another factor that has been subject to intense attention; an entire symposium on the matter was held by the Journal of Herpetology (Dunson et al., 1992). And while some have cited evidence against either UV or pH being a major influence on amphibian survival, a synergistic effect between the two is a very real possibility. Long et al. found that while neither UV nor acidified water showed any statistically significant effect on leopard frog (Rana pipiens) embryos, the combination of the two reduced survival from 97% to as little as 51%. (1995) However, there may also be in many cases hidden agents affecting local populations in apparently pristine areas. Blaustein et al. (1994b) identify a pathogenic fungus carried by stocked fish as being responsible for the decline of the western toad (Bufo boreas) in seemingly undisturbed areas within Oregon.

Another is the argument made by a number of writers that amphibians may be especially good bioindicators of environmental degradation, because their complex life histories expose them to both terrestrial and aquatic environmental influences, their skins and eggs are permeable to water and many electrolytes, and they live much of their lives either on or in the substrate where many toxic substances accumulate. (Vitt et al. 1990) In particular, amphibians' extreme sensitivity to moisture and temperature conditions may make them an exceptionally sensitive "early warning system" for climatic change (Pounds & Crump 1994). However, while these facts certainly indicate that amphibians have the potential to function as coal-mines canaries indicating a general decline in global environmental health, they do not prove that this is in fact occurring.

But perhaps the most crucial question raised by all this is just how much certainty we need before raising the alarm. Even the researchers working most intensively on the amphibian decline problem agree that there is a serious lack of long term census data of the sort that would enable us to know for certain the nature of the situation. Blaustein et al. (1994a) write that:

It is essential that rigorous census studies of a representative sample of amphibian populations be initiated worldwide as a means of assessing the directions, magnitudes, and agents of change in numbers. But how much information is needed before one can decide whether special efforts should be undertaken to protect or restore declining populations? Adopting the conservative approach of withholding intervention until extinction rates are conclusively demonstrated to be unusually high might result in an unacceptable loss of populations or entire species. Erring in the opposite direction, by mistakenly concluding that a global decline is occurring when populations are simply exhibiting normal ranges of fluctuation will waste resources and political capital. It boils down to the issue of how we should balance the risk of lost credibility, which might seriously compromise future conservation efforts in this and other arenas, against the cost of failing to respond to a serious environmental crisis

In other words, the question is whether, in this context, it is better to risk a type I or type II error, because any decision made regarding standards of proof will inevitably affect the odds of erring to one side or the other. Type II errors (acceptance of a false null hypothesis) have generally been considered the lesser of two evils by the scientific community, but ultimately the costs associated with each type of error must be the deciding factor. Pechmann and Wilbur (1994) correctly point out the political risks of "crying frog", and the dangers of engaging in "pathological science," where researchers "unknowingly lose their objectivity in interpreting data that are near detection limits in cases where much is riding on the results, such as cold fusion." However, Blaustein (1994), points out in a response entitled "Chicken Little or Nero's Fiddle," that "In the broad sense, the sky is falling. We are losing an unprecedented number of species in all taxa per year..." The only thing writers on both sides can agree on is the need for more research.

In conclusion, the research questions that must be addressed -- and quickly -- are many, but fall primarily into three categories:

  1. What are the natural dynamics and levels of variance among amphibian populations and metapopulations? Across what scales of space and time should we expect to see stability -- and have reason to suspect anthropogenic influences if we don't?
  2. While biodiversity is threatened across all taxa, each individual type of organism at risk is also unique. What are the factors that uniquely threaten amphibians? We have briefly looked at their key vulnerabilities, such as permeable skin and a combined terrestrial and aquatic life cycle, but there may be other factors as yet unknown.
  3. What standards of proof are required before intervention is justified, and is it better to err on the side of scientific caution or conservationist caution? In the absence of sufficient hard data, how do we weigh the risks? My own feeling is that it is better to err in favour of conservation, because protective measures directed toward amphibians may well benefit other species at risk as well. Climate change, acidic deposition and habitat degradation threaten many organisms, so no effort directed at preventing these things can truly be said to be wasted. But at the same time, the risk of lost credibility is a very real one.

Ultimately, this merely underscores my original point: that conservation biology, and the hard quantitative data on which it relies, is and must continue to be the foundation of biological conservation. Theory is not a substitute for data. Qualitative methods have their place, and the cultural and political factors underlying the biodiversity crisis are surely worthy of exploration, but when it comes to concrete action to save living biological communities from extinction, quantitative research is essential to success.

Literature Cited:

Blaustein, Andrew R. and David B. Wake (1990). Declining amphibian populations: a global phenomenon? Trends in Ecology & Evolution 5. 203-204.

Blaustein, Andrew R. (1994). Chicken Little or Nero's fiddle? A perspective on declining amphibian populations. Herpetologica 50. 85-97.

Blaustein, Andrew R., David B. Wake and Wayne P. Sousa (1994a). Amphibian declines: judging stability, persistence, and susceptibility of populations to local and global extinctions. Conservation Biology 8. 60-71.

Blaustein, Andrew R., D. Grant Hokit, Richard K. O'Hara and Richard A. Holt (1994b). Pathogenic fungus contributes to amphibian losses in Pacific Northwest. Biological Conservation 67. 251-254.

Cohn, Jeffrey P. (1994). Salamanders slip-sliding away, or too surreptitious to count? BioScience 44. 219-223.

Crump, Martha L., Frank R. Hensley and Kenneth L. Clark (1992). Apparent decline of the golden toad: underground or extinct? Copeia 1992. 413-420.

DAPTF (1994a). Froglog: Newsletter of the Declining Amphibian Populations Task Force. Issue 10, August 1994. Milton Keynes, UK.

DAPTF (1994b). Froglog. Issue 11, December 1994. Milton Keynes, UK.

DAPTF (1995a). Froglog. Issue 12, March 1995. Milton Keynes, UK.

DAPTF (1995b). Froglog. Issue 13, June 1995. Milton Keynes, UK.

DAPTF (1995c). Froglog. Issue 14, September 1995. Milton Keynes, UK.

DAPTF (1995d). Froglog. Issue 15, December 1995. Milton Keynes, UK.

Dunson, William A., Richard L. Wyman, and Edward S. Corbett (1992). A symposium on amphibian declines and habitat acidification. Journal of Herpetology 26. 349-352.

Heyer, W. Ronald, Maureen A. Donnelly, Roy W. McDiarmid, Lee-Ann C. Hayek and Mercedes S. Foster. Measuring and Monitoring Biological Diversity : Standard Methods for Amphibians. Washington: Smithsonian Institution Press, 1994.

Long, Leslie E., Loralei S. Saylor, and Michael E. Soulé (1995). A pH/UV-B synergism in amphibians. Conservation Biology 9. 1301-1303.

Pechmann, Joseph H. K. and David E. Scott, Raymond D. Semlitsch, Janalee P. Caldwell, Laurie J. Vitt, and J. Whitfield Gibbons (1991). Declining amphibians: the problem of separating human impacts from natural fluctuations. Science 253. 892-895.

Pechmann, Joseph H. K., and Henry M. Wilbur (1994). Putting declining amphibian populations in perspective: natural fluctuations and human impacts. Herpetologica 50. 65-84.

Phillips, Kathryn (1990). Where have all the frogs and toads gone? BioScience 40. 422-424.

Pounds, J. Alan, and Martha L. Crump (1994). Amphibian declines and climate disturbance: the case of the Golden Toad and the Harlequin Frog. Conservation Biology 8. 72-85.

Soulé, Michael. Conservation Biology: The Science of Scarcity and Diversity. Sunderland, MA: Sinauer Associates, 1986.

Vitt, Laurie J., and Janalee P. Caldwell, Henry M. Wilbur, and David C. Smith (1990). Amphibians as harbingers of decay. BioScience 40. 418.

Wilson, Edward O. (1992). The Diversity of Life. Cambridge, MA: Harvard University Press.


All content copyright 1999-2006 by the individual authors, where cited, or by Lynna Landstreet where not specifically credited.

Creative Commons License Except where otherwise noted, this site is licensed under a
Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License.

Green Web Hosting by Dreamhost Site design: Spider Silk Design - Toronto web designers
This page last modified: January 29, 2006


Wild Ideas has just undergone a major redesign and restructuring, and may still be a little rough around the edges. Please bear with us as we get things sorted out.