ECOTOXICOLOGY: A Comprehensive Treatment - Chapter 28 pot

Clements: “3357_c028” — 2007/11/9 — 12:43 — page 603 — #1 28 Conclusions 28.1 GENERAL The study of community ecology is primarily concerned with understanding how biotic and abiotic factors influence patterns of distribution, abundance, and species diversity. Deriving mechanistic explanations for patterns observed in nature, including the log normal distribution of abundance, species–area relationships, and the spatial changes that occur across latitudinal or elevational gradi- ents, has occupied the attention of community ecologists for several decades. Our treatment of community ecotoxicology has attempted to explain how communities will respond to one major abiotic factor: contaminants. Direct effects of contaminants may result in reduction or elimination of local populations and are generally easier to interpret than indirect effects. An autecological understanding of life history characteristics and species-specific sensitivity to contaminants may be sufficient for predicting direct effects on populations. In contrast, indirect effects of contaminants are often subtle and difficult to predict without conducting manipulative experiments. Our limited understanding of the indirect effects of contaminants is surprising, especially given the prominent role that research on species interactions (e.g., competition, predation, and mutualism) has played in basic community ecology. If ecotoxicologists accept the notion that species interactions are import- ant in regulating natural communities, then focused research should be directed at understanding the influence of contaminants on these interactions. 28.2 SOME PARTICULARLY KEY CONCEPTS 28.2.1 I MPROVEMENTS IN EXPERIMENTAL TECHNIQUES One of the greatest challenges in community ecotoxicology is separating contaminant-induced changes in species diversity and composition from variation owing to natural factors. While observational studies can provide critical support for hypotheses concerning relationships between contaminants and community responses, some researchers feel that experiments are the only way to demonstrate causation. The lack of true replication and random assignment of treatments to experi- mental units limits the use of inferential statistics in most observational studies. The recent emphasis on experimental approaches for assessing the effects of contaminants is seen as a major development in community ecotoxicology. If the ability to test hypotheses with critical experiments truly defines the maturity of a science (Popper 1972), we suggest that experimental approaches should play an increasingly important role in ecotoxicology. Our enthusiasm for experimental approaches is somewhat tempered by recognition of the import- ant tradeoffs between replication and ecological realism. Some ecologists argue that the degree of control afforded by small-scale experiments does not compensate for the lack of realism (Carpenter 1996). We feel that the importance of spatiotemporal considerations for predicting how communities will respond to contaminants should be treated like any other scientific hypothesis (Resetarits and Fauth 1998). Thus, an emerging area of research in microcosm and mesocosm testing is to identify those key ecological processes that must be accurately reproduced in order to have an adequate representation of nature. Another major research goal for community ecotoxicology should be to determine the context dependency of community responses to contaminants. Because communities from different locations 603 © 2008 by Taylor & Francis Group, LLC Clements: “3357_c028” — 2007/11/9 — 12:43 — page 604 — #2 604 Ecotoxicology: A Comprehensive Treatment will vary naturally in diversity, abundance, and history of disturbance, effects of stressors on these communities are also likely to differ. Experimental approaches may be the only way to invest- igate context-dependent responses to contaminants. Relatively simple experimental designs allow researchers to manipulate several variables simultaneously and investigate the importance of dir- ect and indirect effects. The U.S. Environmental Protection Agency (EPA)’s decision to abandon mesocosm testing for pesticide registration is seen by many as a missed opportunity to improve our understanding of context-dependent responses and indirect effects (Pratt et al. 1997, Taub 1997). Finally, while improvements in experimental approaches have strengthened our ability to demon- strate causation, ecotoxicological experiments should not be conducted without an appreciation of natural history or in isolation from underlying ecological theory. We do not consider descriptive and experimental approaches as opposing ends of a continuum, but rather advocate a research program in which well-designed experiments are integrated with observational and theoretical approaches. When small-scale experiments are linked with observational studies, ecotoxicologists will gain a more realistic understanding of how communities respond to contaminants and other stressors. 28.2.2 U SE OF MULTIMETRIC AND MULTIVARIATE APPROACHES TO ASSESS COMMUNITY-LEVEL RESPONSES Community-level data used to assess effects of contaminants range from simple lists that reflect the presence or absence of species to more sophisticated compilations that include abundance, trophic structure, life history characteristics, and measures of species-specific sensitivity. Because the occurrence of an indicator species is influenced by numerous factors other than contaminants, presence–absence data alone are insufficient for assessing all but the most severe forms of pollution. Although there has been significant progress in biomonitoring research since the development of the Saprobien system of classification, most community-level assessments rely on the assumption that species vary in their sensitivity to a particular stressor and that community responses will reflect this variation. Multimetric and multivariate approaches are particularly useful for community-level studies because they reduce the typically complex, multidimensional data to readily interpretable patterns. Unfortunately, the complex and often unwieldy statistical algorithms of many multivariate approaches are considered major obstacles to their widespread application. The new generation of software packages designed to perform multivariate analyses has increased the use of these approaches; however, the widespread availability of “point-and-click” software does not eliminate the obligation of users to fully understand the output. Multimetric approaches are computationally simple, but are data-intensive and often require a comprehensive understanding of ecology and nat- ural history. These approaches have been especially effective for assessing impacts of contaminants in aquatic systems. The development of multimetric indices for other taxonomic groups, particularly for those in terrestrial communities, is seen as an important research need. Although they were developed independently, multimetric and multivariate approaches are complementary and can be used together to assess biological integrity (Reynoldson 1997). For example, output from multivariate analyses could be used to identify sensitive metrics in a multi- metric index. Conversely, multivariate analyses could be conducted using traditional metrics from a multimetric index. By selecting metrics that respond to different classes of stressors, results of multivariate analyses may be useful for identifying specific stressors in systems receiving multiple perturbations. 28.2.3 D ISTURBANCE ECOLOGY AND COMMUNITY ECOTOXICOLOGY One of the most significant contributions of basic ecology to ecotoxicology is the application of disturbance theory in the study of community responses to contaminants. Disturbance is considered a major regulator of community structure and has been the subject of intense debate for several © 2008 by Taylor & Francis Group, LLC Clements: “3357_c028” — 2007/11/9 — 12:43 — page 605 — #3 Conclusions 605 decades. Assuming that responses to natural and anthropogenic disturbance are somewhat analog- ous, theoretical and empirical studies of resistance and resilience may help ecotoxicologists predict effects of contaminants. Rapport’s Ecosystem Distress Syndrome (Rapport et al. 1985) provides an important framework for understanding how communities respond to and recover from natural and anthropogenic disturbances. Basic research in disturbance ecology may also help explain the significant variation in responses to contaminants observed among communities. If diversity in some communities is enhanced under moderate levels of contaminant stress, as predicted by the intermediate disturbance hypothesis, we would not expect that concentration–response relationships between contaminants and species diversity to be linear. In addition, communities subjected to natural disturbance may be preadapted to anthropogenic stressors, thus reducing their sensitivity to contaminants. The ability of communit- ies to tolerate contaminants forms the basis of the pollution-induced community tolerance (PICT) hypothesis (Blanck and Wangberg 1988). In contrast to community composition, which can vary significantly owing to natural factors, increased community tolerance for a particular contaminant is considered to be a direct result of exposure. It is likely that the same biological factors that determine the rate of recovery from natural disturbance will also determine how quickly communities recover from exposure to contamin- ants. Thus, an understanding of colonization abilities of dominant species, proximity of colonists, and life history characteristics may help ecotoxicologists predict recovery times following remedi- ation. Because of natural changes in community composition and species turnover over time, there remains considerable uncertainty in our ability to identify specific endpoints of recovery. For example, we know that some measures of recovery, such as number of species, can quickly return to predisturbance conditions despite persistent differences in community composition. Similarly, because of redundancy in many communities, functional measures may recover faster than struc- tural measures. Our understanding of recovery is further complicated by the uncritical acceptance of equilibrium theories and our failure to recognize the role of historic events. The community- conditioning hypothesis (Landis et al. 1996, Matthews et al. 1996) acknowledges that communities are a reflection oftheiruniquehistory. Because this history may includeexposuretocontaminantsand other anthropogenic disturbances, traditional models of recovery based on equilibrium conditions may not apply. Finally, throughout this section we have attempted to make a strong case for the importance of long-term research. The National Science Foundation’s Long Term Ecological Research (LTER) programs have contributed significantly to our understanding of basic ecology. Unfortunately, there are relatively few examples where long-term studies have been conducted to assess recov- ery from contaminants. Bruce Wallace’s research on responses of headwater streams to pesticides (Wallace et al. 1982) and David Schindler’s long-term research on acidification in lakes (Schind- ler 1988) described in Chapter 23 are two of the more prominent examples. We suggest that a national program monitoring responses to anthropogenic disturbance, analogous to NSF’s LTER program would greatly enhance our understanding of biotic and abiotic factors that determine recovery. 28.2.4 AN IMPROVED UNDERSTANDING OF TROPHIC INTERACTIONS Few topics in basic community ecology are as relevant to understanding ecotoxicological effects of contaminants as food webs and trophic interactions. Quantitative approaches used by ecologists to measure energy flow have been modified to estimate potential transport of contaminants. The primary focus of this effort has been on measuring the concentrations of contaminants in organisms and attempting to quantify contaminant transport among biotic and abiotic compartments. What is generally lacking from many ecotoxicological investigations has been a critical understanding of the ecological factors that influence contaminant transport. In addition to information on physiochemical properties of contaminants (e.g., molecular structure, lipophilicity), ecotoxicologists now realize © 2008 by Taylor & Francis Group, LLC Clements: “3357_c028” — 2007/11/9 — 12:43 — page 606 — #4 606 Ecotoxicology: A Comprehensive Treatment that predicting contaminant transport also requires an understanding of ecological characteristics (e.g., feeding habits, food chain length and complexity, and habitat use). Our ability to quantify the importance of ecological factors on contaminant transport has been greatly improved by the application of stable isotopes analyses. Few advances in the study of food webs have had as great an impact on our understanding of feeding relationships. Time-integrated estimates of energy flow, trophic position, and carbon sources can be obtained by comparing the unique isotopic signatures of consumers and resources. Despite thebroadinterestin quantifyingcontaminanttransport in ecotoxicologicalinvestigations, relatively few studies have considered the effects of contaminants on trophic structure or have used food web characteristics as endpoints in assessments of ecological integrity. We feel that understanding the effects of contaminants on food web length, complexity, and trophic structure is a significant researchneed in ecotoxicology. The ecologicaleffects ofcontaminants ontrophicstructure and the transport of contaminants to higher trophic levels is dependent on the number of trophic levels and whether the system is regulated by top-down or bottom-up factors. These ecological factors have important applications for the management of sport fisheries. Because contaminant transfer is greatly influenced by food chain length and other aspects of trophic structure (Rasmussen et al. 1990), size- selective stocking and other fisheries management programs may influence contaminant levels in game species (Jackson 1997). The most important research limitation in food web ecotoxicology has been the inability to relate concentrations of contaminants measured in different trophic levels with biologically important effects. Bioenergetic approaches may provide the conceptual framework to quantify biologically significant responses associated with contaminant uptake. Because energy is a common currency that unifies all biological systems (Carlisle2000), studying the effects of contaminants on energy flow provides an opportunity to integrate responses across levels of organization. Integrating contaminant transport models with bioenergetic models will allow researchers to link exposure with ecologically significant effects. 28.2.5 I NTERACTIONS BETWEEN CONTAMINANTS AND GLOBAL ATMOSPHERIC STRESSORS Although generally not included in discussions of ecotoxicology, global climate change, increased UV-B radiation, and acidic deposition represent three of the most serious threats to ecological communities. Assessing the direct effects of these stressors is complicated by their large geographic extent, which requires extrapolation across broad spatial and temporal scales. Evidence that global climate change, increased UV-B radiation, and acidification are directly related to anthropogenic emissions has been obtained from a variety of sources. However, the direct and indirect effects on aquatic and terrestrial communities are largely uncertain. The coarse spatial scale of most general circulation models (GCMs) limits our ability to accurately predict regional responses to climate change. For example, although global declines of amphibians have been related to increased levels of UV-B (Blaustein and Wake 1990) and the worldwide degradation of forest health has been attributed to acidification (Ollinger et al. 1993), there is tremendous uncertainty in these relationships. Because of the pervasive and widespread distribution of global atmospheric stressors, ecotox- icologists cannot continue to study the effects of contaminants in isolation. Interactions between atmospheric and local stressors complicate our ability to predict effects of climate change, UV-B, and acidification on communities. It is quite likely that increased temperatures will have a signi- ficant influence on physiochemical characteristics of contaminants and also influence community responses to contaminants. In addition, the structure of communities and their susceptibility to contaminants will most likely change in a warmer climate. Synergistic interactions between UV-B radiation and polycyclic aromatic hydrocarbons (Oris and Giesy 1986) and between acidification and heavy metals (Genter 1995) are well documented. To predict effects of contaminants on communit- ies simultaneously subjected to increased temperature, greater UV-B radiation, and/or increased © 2008 by Taylor & Francis Group, LLC Clements: “3357_c028” — 2007/11/9 — 12:43 — page 607 — #5 Conclusions 607 acidification will require a better appreciation for the impacts of multiple stressors. Currently, our ability to predict interactive effects of multiple stressors on natural systems is greatly limited. It is likely that interactions among stressors will be common in natural systems, and therefore, com- munity ecotoxicologists should anticipate the “ecological surprises” resulting from these interactions (Paine et al. 1998). 28.3 SUMMARY We hope that our treatment of community ecotoxicology has convinced the reader of the importance of understanding how contaminants may affect distribution, abundance, and species diversity. Within the context of the hierarchical arrangement of living systems, communities are intermediate between populations and ecosystems. Although the responses of individual species to contaminants will influence patterns of diversity and abundance in nature, community responses often transcend those observed in populations. Recognition of the emergent properties of communities and higher levels of biological organization remains a significant point of contention between proponents of reductionism and holism (Odum 1984). The emergence of ecotoxicology as a distinct discipline within the field of toxicology was at least partially a result of criticism of traditional reductionist approaches such as laboratory toxicity tests (Cairns 1983, 1986). Until ecotoxicologists develop a better appreciation for the importance of species interactions and indirect effects of contaminants, extrapolation of laboratory results based on responses of single species will remain tenuous. Predicting the indirect effects of contaminants on species interactions and trophic structure has been a major theme in our discussion of community ecotoxicology. Experience has shown that population surveys of charismatic or economically important species often provide an incomplete picture of how communitieswillrespond to or recover from anthropogenic disturbance(Wiens 1996). Similarly, functional characteristics such as primary productivity, nutrient cycling, and detritus processing, the endpoints typically included in ecosystem-level studies, are often less sensitive to anthropogenic stressors (Schindler 1987). Because the endpoints evaluated in community-level assessments are generally sensitive, ecologically significant, and socially relevant, communities are an appropriate focus for ecotoxicological investigations. 28.3.1 SUMMARY OF FOUNDATION CONCEPTS AND PARADIGMS • The study of community ecology is primarily concerned with understanding how biotic and abiotic factors influence patterns of distribution, abundance, and species diversity. • Direct effects of contaminants may result in reduction or elimination of local populations and are generally easier to interpret than indirect effects. In contrast, indirect effects of contaminants are often subtle and difficult to predict without conducting manipulative experiments. • One of the greatest challenges in community ecotoxicology is separating contaminant- induced changes in species diversity and community composition from variation owing to natural factors. • The recent emphasis on experimental approaches such as microcosms and mesocosms for assessing the effects of contaminants is seen as a major development in community ecotoxicology. • An emerging area of research in microcosm and mesocosm testing is to identify key ecological processes that must be accurately reproduced in order to have an adequate representation of nature. • Because communities from different locations will vary naturally in diversity, abundance, and history of disturbance, effects of stressors on these communities are also likely to differ. © 2008 by Taylor & Francis Group, LLC Clements: “3357_c028” — 2007/11/9 — 12:43 — page 608 — #6 608 Ecotoxicology: A Comprehensive Treatment • Although the development of more ecologically realistic experimental approaches has strengthened our ability to demonstrate causation, these experiments should not be con- ducted without an appreciation of natural history or inisolationfromunderlying ecological theory. • Multimetric and multivariate approaches are particularly useful for community-level studies because they reduce the typically complex, multidimensional data to readily interpretable patterns. • Assuming that responses to natural and anthropogenic disturbance are somewhat analogous, theoretical and empirical studies of resistance and resilience may help ecotoxicologists predict effects of contaminants. • An understanding of colonization abilities of dominant species, proximity of colonists, and life history characteristics may help ecotoxicologists predict recovery times following remediation. • Despite the broad interest in quantifying contaminant transport in ecotoxicological investigations, few studies have considered the effects of contaminants on trophic struc- ture or have used food web characteristics as endpoints in assessments of ecological integrity. • Bioenergetic approaches provide the conceptual framework to relate concentrations of contaminants measured in different trophic levels with biologically important effects. • Because of the pervasive and widespread distribution of global atmospheric stressors such as CO 2 , UV-B radiation, and acidification, ecotoxicologists cannot continue to study the effects of contaminants in isolation. • Interactions between atmospheric and local stressors complicate our ability to predict effects of climate change, UV-B, and acidification on communities. REFERENCES Blanck, H. and Wangberg, S., Induced community tolerance in marine periphyton established under arsenate stress, Can. J. Fish. Aquat. Sci., 45, 1816–1819, 1988. Blaustein, A.R. and Wake, D.B., Declining amphibian populations: A global phenomenon? Trends Ecol. Evol., 5, 203–204, 1990. Cairns, J., Jr., The myth of the most sensitive species, BioScience, 36, 670–672, 1986. Cairns, J., Jr., Are single species toxicity tests alone adequate for estimating environmental hazard? Hydrobiologia, 100, 47–57, 1983. Carlisle, D.M., Bioenergetic food webs as a means of linking toxicological effects across scales of ecological organization, J. Aquat. Ecosys. Stress Recov., 7, 155–165, 2000. Carpenter, S.R., Microcosm experiments have limited relevance for community and ecosystem ecology, Ecology, 77, 677–680, 1996. Genter, R.B., Benthic algal populations respond to aluminum, acid, and aluminum-acid mixtures in artificial streams, Hydrobiologia, 306, 7–19, 1995. Jackson, L.J., Piscivores, predation, and PCBs in Lake Ontario’s pelagic food web, Ecol. Appl., 7, 991–1001, 1997. Landis, W.G., Matthews, R.A., and Matthews, G.B., The layered and historical nature of ecological systems and the risk assessment of pesticides, Environ. Toxicol. Chem., 15, 432–440, 1996. Matthews, R.A., Landis, W.G., and Matthews, G.B., The community conditioning hypothesis and its application to environmental toxicology, Environ. Toxicol. Chem., 15, 597–603, 1996. Odum, E.P., The mesocosm, Bioscience, 34, 558–562, 1984. Ollinger, S.V., Aber, J.D., Lovett, G.M., Millham, S.E., Lathrop, R.G., and Ellis, J.E., A spatial model of atmospheric deposition for the northeastern U.S., Ecol. Appl., 3, 459–472, 1993. Oris, J.T. and Giesy, J.P., Jr., Photoinduced toxicity of anthracene to juvenile bluegill sunfish (Lepomis mac- rochirus rafinesque): Photoperiod effects and predictive hazard evaluation, Environ. Toxicol. Chem.,5, 761–768, 1986. © 2008 by Taylor & Francis Group, LLC Clements: “3357_c028” — 2007/11/9 — 12:43 — page 609 — #7 Conclusions 609 Paine, R.T., Tegner, M.J., and Johnson, E.A., Compoundedperturbations yieldecological surprises, Ecosystems, 1, 535–545, 1998. Popper, K.R., The Logic of Scientific Discovery, 3rd ed., Hutchinson, London, England, 1972. Pratt, J.R., Melendez, A.E., Barreiro, R., and Bowers, N.J., Predicting the ecological effects of herbicides, Ecol. Appl., 7, 1117–1124, 1997. Rapport, D.J., Regier, H.A., and Hutchinson, T.C., Ecosystem behavior under stress, Am. Nat., 125, 617–640, 1985. Rasmussen, J.B., Rowan, D.J., Lean, D.R.S., and Carey, J.H., Food chain structure in Ontario lakes determines PCB levels in lake trout (Salvelinus namaycush) and other pelagic fish, Can. J. Fish. Aquat. Sci., 47, 2030–2038, 1990. Resetarits, W.J., Jr. and Fauth, J.E., From cattle tanks to Carolina bays: The utility of model systems for under- standing natural communities, In Experimental Ecology: Issues and Perspectives, Resetarits, W.H., Jr. and Bernardo, J. (eds.), Oxford University Press, Inc., New York, 1998, pp. 133–151. Reynoldson, T.B., Norris, R.H., Resh, V.H., Day, K.E., and Rosenberg, D.M., The reference condition: A com- parison of multimetric and multivariate approaches to assess water-quality impairment using benthic macroinvertebrates, J. N. Am. Benthol. Soc., 16, 833–852, 1997. Schindler, D.W., Detecting ecosystem responses to anthropogenic stress, Can. J. Fish. Aquat. Sci., (Suppl.), 6–25, 1987. Schindler, D.W., Experimental studies of chemical stressors on whole lake ecosystems, Verh. Internat. Verein. Limnol., 23, 11–41, 1988. Taub, F.B., Are ecological studies relevant to pesticide registration decisions? Ecol. Appl., 7, 1083–1085, 1997. Wallace, J.B., Webster, J.R., and Cuffney, T.F., Stream detritus dynamics: Regulation byinvertebrate consumers, Oecologia, 53, 197–200, 1982. Washington, H.G., Diversity, biotic andsimilarity indices.Areviewwithspecial relevanceto aquatic ecosystems, Water Res., 18, 653–94, 1984. Wiens, J.A., Crist, T.O., Day, R.H., Murphy, S.M., and Hayward, G.D., Effects of the Exxon Valdez oil spill on marine bird communities in Prince William Sound, Ecol. Appl., 6, 828–841, 1996. © 2008 by Taylor & Francis Group, LLC . geographic extent, which requires extrapolation across broad spatial and temporal scales. Evidence that global climate change, increased UV-B radiation, and acidification are directly related to anthropogenic emissions. widespread application. The new generation of software packages designed to perform multivariate analyses has increased the use of these approaches; however, the widespread availability of “point-and-click”. experimental approaches as opposing ends of a continuum, but rather advocate a research program in which well-designed experiments are integrated with observational and theoretical approaches. When small-scale