Click for next page ( 82


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 81
ECOSYSTEMS Humanity depends on the continued functioning of eco- systems for food, timber, and other renewable resources and for recreation and aesthetic amenities. Ecological processes also assimilate wastes, recycle nutrients, and regenerate clean air and water. To protect these largely irreplaceable values, it is very important to know the effects of human activities on ecosystems. Evaluation of the potential ecological impact of a given environmental change requires information on effects on units of biological organization above the level of the population or species, we need to examine structural fea- tures of biological communities, such as diversity and abundance of species and interactions among populations, and functional characteristics of ecosystems, such as the transfer of energy, production of biomass, and cycling of oxygen, water, and mineral elements. The stability and resilience of these characteristics over time is another important property of the system. Any factor, such as pollution, that significantly alters the rates of one or more critical processes within a system can shift the system to a new equilibrium state. Spatial heterogeneity, the mobility cf organisms, dor- mant structures, and other mecharisms give most ecosystems a capacity to recover from many kinds cf perturbations, even those that have severe immediate impacts. Some envi- ronmental impacts, however, might produce long-lasting or irreversible detrimental ecological changes. It is there- fore important to know not only the kinds of effects that may occur, but also the ability cf the system to recover from those impacts it may absorb. A great deal of research on the structure and functions of ecosystems has been conducted in the last decade under the auspices of the International Biological Program (IBP), the NSF-PANK Environmental Systems Program, and other major projects. Much less has been dore to study the responses of ecosystems to specific pollution stresses. Available approaches for investigating such problems include labora- - 81 -

OCR for page 81
tory model systems, or microcosms; field studies of experi- mentally perturbed communities of limited dimensions; mathe- matical (computer) simulation models of ecosystems; and field studies to measure effects on the scale of a regional ecosystem. The predictive accuracy of each approach alone is distinctly limited; taken together, they can produce research results that may approximate measurements of ecc- systcm effects. LABOPATOFX MICFCCCSKS cpsms should be refined and acclied, but the limited ut ility^of _suchL- systems for A number of bioassay techniques have recently been developed that employ simplified microcosms of biological communities. Small populations cf six or seven "typical" organisms representing different ecological roles are kept in a small aquarium or terrarium; a sample of a chemical to be tested is introduced, and its behavior and effects are observed (Metcalf 1975). Microcosm studies are appealing because of the speed with which they can be used to test mary chemicals. The systems are relatively simple and can be mass-produced. It is. likely that such models will be increasingly used in evaluations of the potential environmental impacts of new chemicals and other pollutants. It is very important, therefore, to recognize the inherent limitations of the technique for simulating the processes cf actual ecosys- tems. Some microcosm studies have been useful for predict- ing the behavior of chemicals in the environment. Results showing the extent to which substances iray be bioaccumula- ted or degraded, and the nature cf the breakdown products, have been relatively accurate, as ccnfirmed by studies of the same chemicals in the outdoor environment. Microcosm studies, therefore, can be very useful screening devices for identifying substances whose behavior may indicate potential risks of biological effects (NFC 1975). The laboratory microcosm is rot, however, an adequate xrcdel for predicting either the quality cr the magnitude of effects on ecosystems. Because of their structural sim- plicity, microcosms lack homeostatic mechanisms present in real, complex systems. They are thus subject to random changes, and to instabilities and perturbations that might - 82 -

OCR for page 81
rot cccur in nature because of built-in resilience. Furthermore, it is generally not possible to project the probable ecological consequences of effects that may be observed in microcosms, because the mechanisms of effects cn real ecosystems are simply not well enough understood yet. In short, while microcosm studies may be useful to screen substances for further study, reliable predictions of effects on ecosystems require more elaborate tests. EXPERIMENTAL PERTURBATIONS OF COMMUNITIES Experimental perturbations of ecosystems are a very useful tool for producing important information about eco- logical effects of environmental changes. In such studies a number of nearly identical plots, ponds, or streams within a given system are used in statistically designed controlled experiments to examine the effects cf specific pertur- bations on the structural integrity and stability of biolog- ical communities. The parameters measured include species diversity, predator-prey interactions, productivity, energy transfer, and other selected characteristics of the system. To control changes due to the cyclic nature of many biolog- ical processes, measurements should be made under different appropriate seasonal and environmental conditions. Experimental manipulations of communities and ecosystems have been used successfully in recent ecological research; examples include the Hubbard Brock experimental watershed and the Brockhaven irradiated forest (Eormann et al. 1974, Dayton 1971, Hall et al. 1970, Paine 1966, Simberloff and Wilson 1969). Such studies involved selective removal cf key species, introductions of new species, eradication cf whole trophic levels, and intentional changes in inputs of nutrients, toxic substances, radiation, or heat. Techniques both for making experimental perturbations and for measuring biological responses are relatively well-developed. The chief advantage of field studies of this kind is that they measure effects under actual environmental condi- tions; there is no other reliable way to obtain much of the needed information. However, such studies are elaborate and costly, and can look at only relatively small spatial ur.its cf ecosystems, which may not be representative of the tctal system in their response to stresses. - 83 -

OCR for page 81
Some studies of this sort should be undertaken as con- trolled experiments; that is, a pollutant or other environ- mental stress should be introduced as part of the study design. In addition, research should continue tc take advantage of subsequent effects cf accidents (large oil or chemical spills, for example) and of existing pollution- stressed systems (such as the forests arcund a lead smelter or an aluminum refinery), where compariscn studies can be conducted after the fact. Eecause of seasonal and annual variability, the resil- ience of natural systems, and the time scales of many eco- logical processes, several years of observations may be required before reliable evidence of ecological effects can be obtained. The substantial costs cf long-term field ecological research make it important to develop a rela- tively small number of sites that can be sustained until useful results are derived. SIMULATION MODELS CF ECOSYSTEMS gon^inyed resfarch_is^needed_to_in)Erove .Jgi§-Si- l2_£ti£_§ .changes Mathematical simulation techniques, usually computer- based, can be very useful in modeling the behavior of at least portions of ecosystems. Like laboratory microcosms, such simulation techniques cannot, by themselves, predict ecological effects with much certainty. They can, however, contribute to support for environmental management decisions when used in concert with the other approaches described here. Computer simulations have the considerable advantages cf speed and flexibility; the significance of new assump- tions or data can be tested almost immediately. The chief disadvantages of theoretical models, however, are the in- ability of models to include all relevant variables, and of modelers to describe intricate biological relationships in precise mathematical terms. Simulation techniques are most useful for predicting the spatial and temporal distribution, movement, transformation, and accumulaticn of substances in the environment. To the extent that these phenomena depend en well-understood physical and biological processes, prop-* erly designed modeling studies can predict with some con- fidence the exposures tc various contaminants that organ- isms or spatial units in an ecosystem will encounter. The biological mechanisms of effects on populations or ecosys-

OCR for page 81
tens, however, involve many variables and are much less well understood; consequently, predictions of the responses cf whole ecosystems to environmental stresses are beyond the capability cf most existing nodels (Levin 1974) . A model must account for the distribution of responses in bcth space and tiire in order to predict effects cn whcle ecosystems. Present models can simulate with some accuracy structural and functional changes in seme subunits of eco- systems through time. The greatest need now is to develop models that can also inccrporate the spatial variability (heterogeneity) of ecosystems. Fefinement of ecological models is a major objective of current research, and advances in ecosystem modeling would be very valuable for environmental decision making. If simulations of the spatial and temporal distribution cf tcxic agents in an ecosystem could be integrated with models identifying the locations and time scales of criti- cal structural or functional properties of the system, the result would be a "map" that could point to particularly vulnerable, as well as especially resilient, elements of the system. Such predictions would need to be verified by field measurements, but models cculd assist that effort immensely by indicating what, where, and when to measure. MONITORING EFFECTS ON ECOSYSTEMS 2I_sv.s t e.m_chao£t er i s t i c s . The only reliable way of knowing whether deleterious changes are occurring in polluted ecosystems is to Treasure the status cf systems in the field. Carefully chosen indi- cators of critical structural or functional elements of eco- systems (such as species diversity, nutrient cycling, erergy flows, or primary productivity) should be monitored. Ir addition, some parameters should be measured on a scale larger than the local ecosystem. Fcr instance, regional impacts on agricultural productivity, changes in water yield or water quality, or changes in the heterogeneity of wild- life habitat cannot usually be detected in small experimen- tal plots, but rather might be anticipated through the com- bined knowledge .gained from experimental perturbation and theoretical modeling, and then detected by field observa- tion, - 85 -

OCR for page 81
A great deal of past effort has been expended in gathering data (such as measurements of pesticide resi- dues in soils, plants, animals, and humans) with no real sense of the possible ecological significance of the infor- maticn being gathered. The usefulness cf collected data will depend on the soundness of the theoretical basis for choosing particular indicators. If the status of selected organisms is to be used to assess effects on ecosystems, the crganisms chosen should be valid representatives of critical functional guilds, such as primary producers, de- composers, herbivores, predators, and the like. If, on the other hand, the objective is to use wildlife as sentinels fcr potential environmental hazards to man, a sound theoret- ical basis must exist for drawing analogies between specific responses ir those organisisms and risks to humans. In each case, it is critical to know the time delays between exposure to an environmental agert and the appearance of a detectable effect. An example of an important theoretical consideration in selecting parameters to measure to assess ecosystem damage is the potential significance of impacts on organisms that rank low in numerical abundance in biological communities. Monitoring has generally focused on the most common, easily sampled representatives of a given functional guild. The resilience of an ecosystem depends to a large extent, how- ever, on the ability of less numerous species to increase their populations to fill an ecological role if the popu- lation of the primary species occupying that role should decline. Thus, if an environmental contaminant were selec- tively toxic to the most abundant species, the impact or the system might be less severe than if it increased mortality randomly in all species within the functional group. It is important, therefore, to obtain information on the effects of an environmental change on some cf the many numerically uncommon species, as well as on the more abundant ones, and to assess the impact on the biological community of mortal- ity in relatively rare species. Measurement of effects on the functions of ecosystems at the regional scale should be based en monitoring of important processes, as well as observations of effects on representative organisms. Some critical processes to monitor include the hydrologic cycle, the transfer of energy and production of biomass, the natural (biological) regulation of "pest" populations, and the cycling of nutri- ents and ether elements. Research on cne of these (nutrient cycles) is discussed in detail below as an example; but po- tential effects on any of these functions (and others net mentioned) deserve research attention. - 86 -

OCR for page 81
Effects on Nutrient Cycles made tg ?-fcudv the effects of pollutants One of the critical functional characteristics of eco- systems is the cyclic movement of nutrients, pollutants, and other elements among the atmosphere, soil, water, and living things. In both terrestrial and aquatic ecosystems, processes carried out largely by microorganisms synthesize, transform, and decompose the microchemical species of many stages of biogeochemical cycles. It is possible that pol- lutants or other stresses could, through effects on the environment itself or on the organises involved, alter either the products or the rates of some of the myriad chemical and biological processes that make up cycles. Even small changes in the rates cf a few critical pro- cesses could produce, over time, substantial shifts in the pools or rates of flow- of important nutrients or of undesirable ty-products or pollutants. For example, a change in the rate of denitrif icaticn in soils or aquatic systems could have adverse or bereficial impacts on pro- ductivity, and on the generation of nitrous oxide, a crit- ical factor in the depletion of stratospheric ozone (Council for Agricultural Science and Technology 1976) . Cur present understanding of potential effects cf environmental contaminants on bicgecchemical cycles is too crude to assess the resulting risks cf significant long-term impacts on ecosystems cr climate. Most inves- tigations of ecological effects cf pollution have not examined impacts on biogeochemical cycles or underlying microbial processes. One exception is the extensive body cf research on effects cf fertilizers and pesticides on soil organisms; but even in this well-studied field, systematic understanding of impacts cn ccmplex elemental cycles has proved to be elusive. Cn the other hand, recent large-scale ecological research efforts?" such as the IEP have produced a substan- tial amcunt of basic knowledge abcut the ways important nutrient cycles function in a variety of terrestrial and aquatic systems. Some general principles have emerged, based on repeated empirical observations rather than on sophisticated theoretical constructs. For instance, it has been noted that, in general, undisturbed ecosystems tend to retain nutrients, while perturbed systems tend to lose nutrients (Eormann et al. 1974, Neuhold and Ruggiero 1977). Although it is not yet clear why this is so, this - 87 -

OCR for page 81
knowledge may make it possible tc use rates of nutrient flow from a system as rcugh indicators of stress or injury to the system. X It would be very valuable, therefore, to make the study of effects on nutrient cycles a majcr fccus of field inves- tigations of the ecological consequences of pollution. Such effects should be examined both for their utility as a meas- ure cf general stress on ecosystems and to determine the mechanisms involved and the ecological significance of any shifts that might occur in cycles. The biogeochemical cycles of nitrogen, carbon, and sulfur are among the most critical and best understood nutrient pathways, and would be the soundest initial choices for study. Cycles cf otter elements known to be either essential for life or particularly toxic should also be considered where feasible. Comparably designed studies should examine nutrient cycles in a variety of polluted terrestrial and aquatic systems. Some environ- mental perturbations that might be investigated include acid rainfall, heavy fertilizer use, pesticide applica- tions, fly ash fallout, fluoride and heavy metal pollu- tion, and .sludge disposal. Some unstressed ecosystems should also be examined, as controls. Field studies should be supported where feasible with laboratory and sample-plot studies. Effective study of a problem of this magnitude will require coordinated efforts involving microbiologists, ecologists, soil scientists, limnologists, and many other specialists. Studies may require five or ten years to obtain definitive results. A unifying overview and close coordination among many projects will be essential if such a program is to fulfill its prcmise. INSTITUTIONAL AFRANGEKEKTS The kinds of research needed to predict and Treasure effects on biological ccmmunities and ecosystems are cur- rently being carried out by many investigators in diverse specialties, located in universities, private research institutes, national laboratories, and federal agencies. Substantial support for ecological research has come through NSF (RANN), IBP, and other sources. Federal agen- cies, including EPA, ERCA, COI, KOAA, USEA, and the Army Corps of Engineers have supported investigations of ecolog- ical problems of specific interest to them. No single agency, however, has taken the effective lead or coordina- ting role that is essential if ecological research is tc be useful in making decisions on managing the environment. - 88 -

OCR for page 81
Much of the research recommended here, such as improve- ments in theoretical ecology, is probably best carried out in uriversities and- national laboratories, with support from traditional sources of funds for basic research in biology and ecology. Because of the importance of basic ecosystem research for environmental protection, the adequacy of current funding levels for this field should be carefully reviewed. The results of most of the kinds of studies described above would be valuable and useful to EFA, ERDA, USDA, and a dozen or more other federal agencies that have responsi- bilities for managing land and water resources, and for pre- venting pollution and enhancing environmental quality. Some of the recommended research is, short-term and decision-cri- ented, and would be appropriately conducted by EPA or other agencies, intramurally or extramurally. This category includes screening chemicals with laboratory microcosms, and using the most recent data and best available modeling techniques to evaluate potential ecological impacts of par- ticular actions. Other work, such as studies of experimen- tal perturbations of ecosystems or monitcring of nutrient cycles, will produce information of great value to a wider group of federal and state agencies, but only after a num- ber of years. Commitment to support several such studies to completion would be difficult for any single agency, because the costs would be a considerable fraction of any agency's research budget, and most operate on a one-year funding system. This research wculd probably best be dene and supported with the participation of several agencies, including EPA. Close coordination of such a program would be required, and might be achieved through a mechanism such as the new White House Office of Science and Technology Policy or the Office of Managemert and Eudget. In order to make studies of ecological damage both coherent and relevant, such an overview and coordinating mechanism is essential. - 89 -

OCR for page 81
RJFERE.NCE.S Hermann, F.H., G.E. Likens, T.G. Siccama, R.S. Fierce, and J. S. Eaton (1974) The export of nutrients and recovery of stable conditions following deforestation at Hubbard Erock. Ecological Monographs UH (3): 1-225 (Entire Issue). Council for Agricultural Science and Technology (1976) Effect of Increased Nitrogen Fixation on Stratospheric Ozone. Report #53. Ames, Iowa: Council for Agricultural Science and Technology. Dayton, P.K. (1971) Competition, disturbance and community organization: the provision and subsequent utilization cf space in a rocky intertidal community. Ecological Monographs 41(4):351-389. Hall, D.J., W.E. Cooper, and E.E. Werner (1970) An experimental approach to the production dynamics and structure of freshwater animal communities. Limnology and Oceanography 15 (6) : 839-928. Levin, S.A., ed. (1974) Ecosystem Analysis and Prediction. Conference Proceedings. SIAM Institute for Mathematics and Society. Alta, Utah: SIAM Institute for Mathematics and Society. Metcalf, R.L. (1975) Evaluation cf a Laboratory Microcosm for Study of Toxic Substances in the Environment. NSF/RA/E-75-116. Urbana-Champaign: University of Illinois. National Research Council (1975) Principles for Evaluating Chemicals in the Environment. Environmental Studies Ecard, Commission or. Natural Resources, and Committee on loxicology. Assembly of Life Sciences. Washington E.C.: National Academy of Sciences. Neuhcld, J.M. and L.F. Ruggiero, eds. (1977) Ecosystem Processes and Organic Contaminants. Rational Science Foundation Directorate for Research Applications, RANN Division of Advanced Environmental Research and Technology. Washington, D.C.: U.S. Government Printing Office. - 90 -

OCR for page 81
Faine, P.T. (1966) Food web complexity ard species diversity, American Naturalist 100:65-75. Simberloff, D.S. and E.C, Wilson (1969) Experimental zoogeography of islands: the colonization of empty islands. Ecology 50 (2) : 278-289. - 91 -