Appendix B
Measuring Change in Ecosystems: Research and Monitorning Strategies

A Workshop Report



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 97
Opportunities in Applied Environmental Research and Development Appendix B Measuring Change in Ecosystems: Research and Monitorning Strategies A Workshop Report

OCR for page 97
Opportunities in Applied Environmental Research and Development This page in the original is blank.

OCR for page 97
Opportunities in Applied Environmental Research and Development Summary Scientists began sustained measurements of ecosystems several decades ago. Classic studies, such as those of Hubbard Brook in New Hampshire, and recent studies, such as that of Chesapeake Bay, demonstrate the evolution of ecosystem science. These studies have heightened scientists' awareness of the need to understand better the continued degradation of many of the country's ecosystems, particularly where the muses of such degradation are anthropogenic. The Science Advisory Board of the Environmental Protection Agency (EPA) noted this need in its subcommittee report, Future Risk: Research Strategies for the 1990's (EPA, 1988). An increasing number of resource agencies also recognized the need to develop programs for evaluating the consequences of their management decisions and the public's use of the ecosystems under their control Unfortunately, as indicated in the 1977 National Research Council (NRC) report, Environmental Impacts of Resource Management, efforts among government and private institutions to develop programs and methodologies to address ecosystem degradation have suffered from lack of top-level support and coordination. An NRC workshop was held in Warrenton, Virginia, on March 2-3, 1989, to develop ideas for research that EPA and other agencies could use to address ecosystem and landscape issues. Through this report of the workshop, the workshop participants, while recognizing problems inherent in dealing with the wide diversity of systems found in this country, hope that their efforts will encourage stronger coordination among agencies. According to workshop participants, this absence of top-level support and coordination continues today while our nation's natural systems continue to deteriorate. New approaches, such as ecological risk assessment, have developed over the past decade. These approaches, combined with increasing effort in ecosystem studies, may give us the necessary tools to retard the rapid degradation of many of our nation's natural treasures as well as the systems on which we depend for our well being. To improve the nation's ability to anticipate future ecosystem and landscape changes in response to natural and anthropogenic perturbations, workshop participants developed a strategy that included the following principal elements: Establish the principle of maintaining ecosystem and landscape integrity and sustainability as an integrative management policy. Select indicator variables of ecosystem integrity and sustainability on a regional or landscape basis. Develop standard methods of data collection and analysis for ecosystem monitoring. Establish an integrated, large-scale, long-term national program for regionally focused ecosystem monitoring, research, and risk assessment. A program that included these four aspects could assist in solving regional issues through the use of regional experts; coordinate regional and local monitoring programs and synthesize data collected from these programs; rank research activities necessary to provide tools for the national program; provide information for management and decision making; and, through annual reports, inform the public about short-and long-term environmental changes, assessment needs, and management recommendations.

OCR for page 97
Opportunities in Applied Environmental Research and Development This page in the original is blank.

OCR for page 97
Opportunities in Applied Environmental Research and Development Measuring Change in Ecosystems: Research and Monitoring Strategies INTRODUCTION The nation is facing concerns about many environmental problems that involve long-term, large-scale environmental degradation (e.g., complex chemical pollution, regional air pollution, coastal degradation, wetland losses, and loss of biotic diversity). Moreover, major ecosystems such as the Chesapeake Bay, Puget Sound, the Great Lakes, the Grand Canyon, and the Great Smoky Mountains are subject to multiple anthropogenic disturbances. Effective conservation and management of these systems will require identification of resources in jeopardy as well as knowledge of the muses of degradation. Assessing and managing risks of this magnitude require an effective strategy for environmental research, monitoring, and assessment. A workshop on ecosystem risk assessment and monitoring was held March 2-3, 1989, near Warrenton, Virginia, to address these research and monitoring needs. This report presents ideas developed at the workshop on ecosystem risk assessment and monitoring for the purpose of understanding ecosystem change or degradation. Strategies to implement a research and monitoring program in a consistent, cost-effective, and scientifically credible manner were addressed. EPA's Environmental Monitoring and Assessment Program (EMAP) was being developed when the workshop was held. Several of the workshop participants had a role in developing EMAP, and the similarity of EMAP as currently described by EPA to many of the workshop's recommendations reflects that dual role. Several workshop participants indicated that successful environmental protection will require greater efforts by federal agencies to understand and thus to predict ecosystem-and landscape-level consequences of environmental disturbance and contamination. The federal government's approach to environmental threats has been focused largely on risks to human health, endangered species, and a few critical habitats. This approach is insufficient, although significant progress has been made in identifying and controlling some major air and water pollutant sources. Assessments of the response of ecosystems and landscapes (large spatial traits with interacting ecosystems) to anthropogenic perturbations are performed in a piecemeal, fragmented manner because responsibilities are divided among many agencies; such assessments usually fail to provide adequate estimates of environmental threats according to a number of workshop participants. To address inadequacies, a coordinated research program is needed that is directed toward identifying key indicators of ecosystem integrity (structure, function, and stability) and toward using these indicators to provide information about the natural variability of ecosystems. Such a research program is vital for improving our knowledge of the kinds of perturbations at the ecosystem and landscape levels that are likely to be significant risks to the stability or long-term viability of a variety of terrestrial and aquatic ecosystems. A number of workshop participants indicated that subtle ecosystem perturbations may have long-term regional and even global consequences. Various attempts have been made to improve understanding of the structure and function of large ecosystems, such as the International Biological Program, the International Geosphere-Biosphere Program, the Long-Term Ecological Research Program, and the Global Emissions

OCR for page 97
Opportunities in Applied Environmental Research and Development Monitoring System. In addition, NRC studies (1981, 1986) have identified the need for more baseline data on ecosystems and their natural variability to improve the ability to predict responses to natural and anthropogenic perturbations. Public and national concern over the possibility of regional and global environmental consequences of such perturbations has encouraged ecologists and resource managers to increase their efforts to understand the dynamics of ecosystems and landscapes and to predict with greater reliability ecosystem responses to natural and anthropogenic perturbations. Examples of studies of natural resource perturbations at the landscape level are in the literature (Barrett, 1985) and were also presented at the workshop (Patten, 1989). Unfortunately, the available information on ecosystems, including their natural variability, continues to be insufficient to permit rigorous risk assessments, especially at the landscape level. According to a number of workshop participants, recommendations for action to reduce risks to ecosystem integrity presuppose knowledge sufficient to distinguish changes that would have serious adverse consequences from changes that would be either beneficial or insignificant. It is precisely this kind of knowledge that is still inadequate or improperly synthesized. This workshop was in large part an effort to address this lack of synthesized information for ecosystem research and monitoring strategies. Workshop participants attempted to develop a general research strategy that would lead to better understanding of ecosystem and landscape processes relevant to ecosystem risk assessment and would address long-term environmental issues that are likely to confront the nation over the next decade or more. PRINCIPAL FINDINGS During the 2-day working group and plenary discussions, workshop participants indicated that regulatory and resource management agencies should consider the following strategy to improve the nation's understanding of landscape changes resulting from anthropogenic and natural perturbations. Establish the principle of maintaining ecosystem and landscape integrity and sustainability as an integrated management policy. Select indicators of ecosystem integrity and sustainability on a regional or landscape basis. Develop standard methods of data collection and analysis for ecosystem monitoring. Establish an integrated, large-scale, long-term national program for regionally focused ecosystem monitoring, research, and risk assessment. The elements of each of these is described further below:. 1. Establish the principle of maintaining ecosystem and landscape integrity and sustainability as an integrated management policy. Management of ecosystems and landscapes to prevent continued deterioration of natural resources requires a research program dedicated to that objective according to a number of workshop participants. Sustainability requires maintenance of essential ecosystem resources and processes (e.g., available moisture and nutrients and productivity) above threshold levels determined to be necessary to maintain essential system integrity and vitality. An integrated national monitoring and research program should provide periodic information, based on the analysis of critical indicators, to assess whether this objective is being achieved. To help provide direction for implementing this suggestion, several workshop participants developed a strategy for assessing and managing risks of large-scale ecological change so that decisions regarding choices among regulatory action, additional monitoring and research, or no action could be made within a risk assessment framework analogous to that now used to make other regulatory decisions (such as decreasing the speed limit on highways or banning use of certain chemicals as food additives). The key component of the assessment strategy is an ecological inventory, monitoring, and research program to characterize the ''health,'' or integrity, of the environment (Shaeffer et al., 1988) and to quantify the influence of potential environmental perturbations. The data from this program, together with appropriate statistical and

OCR for page 97
Opportunities in Applied Environmental Research and Development mechanistic models, would be used to (1) identify potential adverse changes before major or irreversible damage has occurred, (2) identify causes for the observed changes, and (3) predict ultimate consequences—leading to a rational basis for management decisions (Figure 1). Selection of end points is the first step in any risk assessment. In health risk assessment terminology, an end point is most often defined as an undesirable event (e.g., contracting cancer or being injured in an automobile accident), and the objective of the assessment is to quantify the risk of occurrence of this event. The appropriate end points for human-health risk assessment are usually negative and obvious (e.g, mortality, morbidity, and teratogenesis). Some negative population-level end points for ecological risk assessment, such as extinction of endangered species or reductions in fish or timber yield, can be readily defined (Barnthouse et al., 1988). However, knowledge of population-level end points is not sufficient to make adequately informed environmental decisions when dealing with ecosystem and landscape perturbations; knowledge of intrinsic variables that indicate changes in ecosystem structure, function, or stability is required. According to several workshop participants, some critical attributes are difficult to measure on ecosystem and landscape levels, aside from using aerial photographs or satellite observations. Therefore, to perform a risk assessment, changes in those attributes usually will need to be inferred from more readily measured variables. For each critical attribute identified for ecosystems, one or more measurement variables are needed. These variables should be readily measurable, sensitive to different types of perturbation, indicative of current status, and indicative of short-and long-term changes. Suter (1989) defined "measurement end points" as the measurements from which changes in assessment end points are inferred. For example, maintenance of lake trophic status is a common assessment end point (indicator variable) for regulation of discharges into a lake. However, regulation is not usually based on measurements of effects of effluents on receiving systems, such as changes in the trophic status. Instead, regulation is based on measurement of nutrient loading and published water quality standards. The second step is establishment of a monitoring program to evaluate the integrity of ecosystems or landscapes through their critical attributes. Such a program should provide periodic quantitative information on the measurement variables (e.g., density of critical resource species) and other diagnostic variables (e.g., contaminant distributions, climatic data, and land-use characteristics), thereby assessing the condition of the ecosystem. These measurement variables would be chosen to be region-specific (regional indicator variables). Third, explicit statistical or mechanistic models are needed to predict changes in the critical attributes based on measured changes in the variables. Statistical extrapolation models, simulation models, or other quantitative risk assessment models could be used to translate variable measurements into estimated values or changes in values of the critical attributes. Measurements of diagnostic variables would help to elucidate the causes of change. Because of the spatial variability in ecosystems and potential multiple cumulative effects, geographic information systems may be crucial to accurate prediction of the effects of different threats on regional landscapes (Brown and Norris, 1988). Management decisions of regulatory action, additional monitoring and research, or no action could then be made within a risk assessment framework analogous to that now used to make other regulatory decisions. A number of workshop participants also indicated that the assessment and evaluation of critical attributes of ecosystems, analyzed within system boundaries (e.g., changes in biodiversity or rates of nutrient flux) and between ecosystems (e.g., rates of genetic dispersal or of nutrient exchange between systems), should clarify ecosystem self-regulatory and recovery processes and, consequently, suggest strategies for ecosystem management and land-use planning on a regional basis. In regional land-use planning, retention and fluxes between various ecosystems (e.g., between forest and agricultural plots) can be properly evaluated mainly at the landscape level. Changes in regional land-use patterns likely will alter individual ecosystem dynamics within the landscape mosaic. Further, several workshop participants indicated a landscape perspective for ecosystem analysis may also increase our

OCR for page 97
Opportunities in Applied Environmental Research and Development FIGURE 1 Ecosystem assessment strategy

OCR for page 97
Opportunities in Applied Environmental Research and Development understanding of life cycles of keystone species that may use more than one ecosystem type during their life history. For example, many bird species use natural ecosystems for nesting but use manipulated systems (e.g., agricultural fields or controlled-flow rivers) for foraging. This information should also help to explain changes in biotic diversity within various ecosystems. It may well be that better understanding of the interactions among ecosystems will prove to be more important for risk assessments and ecosystem regulation than the more isolated processes are that ecologists now commonly study, e.g., individual species indices. A research and monitoring program that assesses interactions among systems will require interdisciplinary coordination and integration of research (Barrett, 1984). 2. Select indicators of ecosystem integrity and sustainability on a regional or landscape basis. Ecosystems consist of the interacting biotic and abiotic elements of a defined physical environment (Odum, 1971; Ricklefs, 1976). Ecosystems can be designated by their dominant physical and chemical characteristics, such as glacial lakes or hardwater streams, or by dominant biological populations, such as oak-hickory forests or big bluestem grasslands. A number of workshop participants indicated it is impossible to measure all biotic and abiotic elements of ecosystems for changes in structure or function. The alternative approach suggested by several workshop participants is to define characteristics of ecosystems that are likely to be critical for maintenance of ecosystem integrity and sustainability. These characteristics are called critical attributes in this report. Although critical attributes are only components of ecosystems and landscapes, their status can be used to evaluate the linkages that make an ecosystem function as a whole. These attributes are often complex and difficult to measure, e.g., maintenance of a lake's trophic status; therefore, for each critical attribute, one measurement variable or more are needed that can be measured or monitored during long-term studies of ecosystem health and sustainability. For each measurement variable, indicator variables, such as selected species, should be chosen regionally because of the complexity and variability of ecosystems throughout the country. Table I shows examples of critical attributes likely to be important in assessing ecosystem integrity. Because indicators are specific to ecosystem type and will differ among geographic regions, they must be selected on a regional bask by scientists who best understand the local ecosystems. Regional workshops could be held for scientists to discuss and select these indicators. It is imperative that ecologists most knowledgeable about specific regional ecosystems (whether from academia, the private sector, or state and national environmental centers) be involved actively in the selection of the indicator variables to be studied (e.g., controlling nutrients, critical species, critical trophic-level components for key systems, and species indicative of disturbance). 3. Develop Standard Methods Of Data Collection and Analysis For Ecosystem Monitoring. Three stages in the acquisition of information needed to assess changes in critical attributes of ecosystems are the following: Inventory and compilation of existing information characterizing important features, natural resources, and other environmental baseline data on landscape units or their component ecosystems. This information may be available from a variety of federal and state monitoring programs, natural resource data sets, and literature. Regional or national monitoring programs that will provide information to determine the status of critical attributes, to detect change, and to evaluate the effectiveness of selected critical attributes as diagnostic factors. The focus of research programs should be to develop a set of monitoring methodologies that would derive a set of environmental indices that could be used by decision makers (Boyle, 1987, 1989). Monitoring programs may indicate ecosystem changes; however, research is needed to explain the cause of these changes and to interpret the significance and implications of these changes. Research at the ecosystem and landscape levels that will provide information on critical attributes that are poorly understood, on the

OCR for page 97
Opportunities in Applied Environmental Research and Development TABLE 1. Ecosystem Integrity Attribute Assessment Examples • Critical attribute: Elemental dynamics—inputs, internal transport, and losses of critical plants, animals, and elemental biochemical compounds, including the flux of nutrients essential for primary production and secondary metabolism. Measurement variable: Potentially limiting nutrients, carbon flux, biomass. Regional indicator variable: One or more controlling nutrients in regional ecosystems. • Critical attribute: Energy dynamics (physical)—energy exchange at geological and biological surfaces (e.g., insolation, sensible and latent heat fluxes, and transportation) and mixing processes (e.g, turbulence, convection, and advection). Measurement variable: Microclimatic and hydrogeological processes. Regional indicator variable: Specific microclimatic and hydrogeological processes shown to be controlling variables for selected regional systems. • Critical attribute: Food web (trophic dynamics)—the set of trophic relationships among species in a community. In its simplest form, the food web is an energy-flow diagram connecting each consumer to all species that it consumes. However, in its dynamic form, the concept also includes rates of consumption, preference of food items, and prey switching. A given food web may indicate which species are necessary resources for other species (e.g, a particular valued species), the amount of redundancy in community functions, and the degree to which particular consumers, termed "keystone" species, may control the competitive processes among the species consumed. Measurement variable: Structural characteristics: species density, biomass, and richness; community composition Regional indicator variable: Presence and abundance of species shown to be controlling trophic dynamics in selected regional systems. • Critical attribute: Biodiversity—the number of taxa per unit area as represented by populations, guilds, or life forms, as well as the relative abundance of the various taxa. Measurement variable: Species richness. Regional indicator variable: Populations, guilds, or other selected taxonomic groups. • Critical attribute: Critical species—keystone, resource, and endangered species: keystone species are those that exert influences over other populations in their ecosystem out of proportion to their abundances (NRC, 1986); resource species are species of energetic, economic, or aesthetic importance; endangered species are species in imminent danger of becoming extinct. Measurement variable: Population monitoring of keystone and resource species Regional indicator variable: Species presence, absence, or density.

OCR for page 97
Opportunities in Applied Environmental Research and Development • Critical attribute: Genetic diversity (within critical species)— genetic diversity represents the number and frequency of different genotypes within species. Measurement variable: Biochemical markers. Regional indicator variables: Selected species. • Critical attribute: Dispersal and migration—movements of individuals within and between ecosystems that are crucial to the population's survival and the ecosystem's health, including colonization or dispersal between habitats and movements of individuals to different habitats for food, reproduction, overwintering, or protection from predators. Measurement variable: Dispersal and migration rates. Regional indicator variable: Selected species. • Critical attribute: Natural disturbance—externally driven disturbances, unrelated to human activities, that have major impacts on ecosystem integrity by altering the species composition, trophic structure, or other important ecosystem developmental processes; these disturbances include wind storms, fires, and floods that are the result of weather patterns and are often essential to maintenance of certain ecosystems. Measurement variable: Disturbance events. Regional indicator variable: Presence and abundance of invader species known to be indicative of disturbance. • Critical attribute: Ecosystem development (successional processes)—developmental changes (successional stages) in species composition through time, mediated by biological-physical interrelationships, resulting in a defined ecosystem structure and function. Measurement variable: Structural characteristics: species density, biomass, and richness; community composition; functional characteristics: production/respiration, production/biomass, carbon or nitrogen flux. Regional indicator variable: Distribution and abundance of plant and animal communities indicative of successional stages versus those characteristic of mature or stable ecosystems (e.g,, r-selected versus K-selected species).

OCR for page 97
Opportunities in Applied Environmental Research and Development FIGURE 1 Conceptual model of Mono Basin ecosystem

OCR for page 97
Opportunities in Applied Environmental Research and Development FIGURE 2 Ranges of lake levels affecting resources of the Mono Basin, with three salinities for reference (From National Research Council, 1997)

OCR for page 97
Opportunities in Applied Environmental Research and Development The gaps in information on the Mono Basin ecosystem also might cause some unanticipated changes if the lake levels were allowed to fail Glen Canyon Glen Canyon Dam on the Colorado River was completed in 1963. It took nearly 20 years for Lake Powell to fill behind the dam. During that period the releases through the dam were based on energy needs of southwestern cities and requirements for water supply to lower basin states. Operation of the dam has been controlled by hydroelectric production with other resources playing a secondary role. Recent proposed changes in operation of the dam has required a study of the effects of these changes on the many resources of the canyon (U.S. Department of the Interior 1988). Most of the endpoint resources in Glen Canyon and Grand Canyon that would be affected by the releases are directly related to human use of the canyon. These include water for rafting, beaches, maintenance of a trout fishery as well as the general aesthetics of the canyon. Other resources that are of concern, either because of the National Environmental Policy Act or the Endangered Species Act, include bird and wildlife diversity and native fisheries. An analysis of the effects of changing operations of the dam must take into account the relative risks to the endpoint resources, as well as the receptors, those ecosystem factors that if changed directly or indirectly influence the endpoints (Fig. 3). Flow control at Glen Canyon Dam (Fig. 3) is the controlling variable for ecosystem responses below the dam. The controlled flows are both in quantity and periodicity. These influence the scouring effects, sediment movement, and temperature of the water. Vegetation, wildlife, beaches, fishing, and rafting all are influenced by these changes. How might policy decisions on various flow regimes create risks to all of these resources and how are these risks assessed? As an example, certain flows would allow some warming of water temperatures. The exotic fishery (i.e., trout) is dependent on cool water, while the native fish (e.g., humpback chub) require some warm water for spawning. The two are not totally incompatible but one fishery is at a greater risk when the other is at a lower risk. Another example deals with high flow releases. Periodic high flow releases may scour exotic vegetation creating a more natural environment. However, scouring with sediment-free water from the dam moves a great deal of sediment downstream destroying beaches which are not replaced. Vegetation loss also reduces bird and wildlife diversity and high flows may create a high-risk for river rafting. The decision to chose a recommended flow, both quantity and timing, must be a compromise. Setting of priorities on resources is difficult bemuse of the many interest groups. It is possible to set flow regimes if enough is known about the interactions of the ecosystem components and processes and response curves can be developed for each resource. Unfortunately, information on some species in the canyon is incomplete. Sediment transport dynamics are also not well known. With these gaps in information, evaluation of risk levels to the ecosystem components is not accurate. Further short-and long-term studies are needed, but meanwhile, risk analyses need to be done for the endpoint resources to facilitate recommendations on flow regimes. Relationships To Epa Priorities These two case studies are not closely related to the past priorities of EPA. However, there is an increasing concern for the welfare of the ecosystem on which we are all dependent. These cases demonstrate how, with sufficient information, management or policy decisions can be made based on evaluations of the risks to components of the ecosystem. The Mono Basin study showed a nearly complete assessment while the Glen Canyon study showed an inability to make recommendations because the assessments were incomplete.

OCR for page 97
Opportunities in Applied Environmental Research and Development FIGURE 3 Conceptual scheme of Glen Canyon ecosystem components and theirinteractions under present operations of Glen Canyon Dam. (From National Research Council, 1987. River and dam management, National Academy Press, Washington, D.C.)

OCR for page 97
Opportunities in Applied Environmental Research and Development Still, recommendations on ecosystems such as Mono Basin and Glen Canyon lie outside the purview of EPA. Let me suggest, that there are circumstances where these systems do fall within EPA's directive. What would happen if underground gasoline tanks at Lee Vining, a town just above Mono Lake, leaked into the groundwater feeding Mono Lake? Is our information base on dosed basin lakes like Mono Lake sufficient to project the consequences and take some action? Contamination of bodies of water is always possible. Should EPA limit itself to contaminants? Is the water quality of Mono Lake or the Colorado River of importance? Dissolved salts and temperature are characteristics of water quality. Variation in these factors that affect the natural biota should be a concern of EPA. Management of our natural ecosystems is controlled by many agencies but the ultimate watchdog of ecosystem health must be the responsibility of one agency and EPA is the only agency whose sole objective is ensuring a healthy environment for humans who are inextricably dependent on natural ecosystems.

OCR for page 97
Opportunities in Applied Environmental Research and Development SELECTED DEVELOPMENT NEEDS FOR ASSESSING ECOLOGICAL RISK AT THE COMMUNITY AND ECOSYSTEM LEVEL Terence P. Boyle Water Resources Division National Park Service A Background Paper For Ecosystem Risk Assessment and Monitoring, March 2-3, 1989, Airlie House, Virginia. This paper is intended to provide some background information and suggestions for improvements in the art and science of ecological risk assessment and environmental monitoring. Emphasis is focused on research needs in the development of methods for assessments for ecosystem risk using natural ecosystems. It is not intended to be representative of all the needs in this field of endeavor. To date ecosystem risk assessment has been commonly effected by the use of surrogates in the laboratory with batteries of single species toxicity tests, and microcosm tests of various complexities which assess for the effects of chemicals at the population, community and ecosystem level of organization. Microcosm tests performed outdoors containing natural communities or portions thereof have recently come to be known as mesocosm tests and include some of the variability inherent in natural ecosystems. Several generations of computer models have also attempted to assess ecosystem risk by various types of environmental stress mainly considering ecosystem aspects of transport and the fate of chemical contaminants. There are several levels of consideration in selecting news research needs in ecosystem level risk and monitoring. These fall into three categories: 1) Development of procedures to include the administrative or management aspects of the problem, 2) The use of previous and existing monitoring data and management questions to formulate comprehensive risk assessment and monitoring programs, and 3) Selection or development of technical tools necessary to assess changes at the community or ecosystem level that will answer questions implicit in ecosystem risk assessment and monitoring or ecosystems. Administrative/Management The first set of considerations that needs to be addressed in collecting environmental data is, "What will be done with ecological data when it is generated?". These are the social, economic, political, and legal aspects of the risk assessment and monitoring that directly affect the scope and detail of any program. For example, the purpose of environmental monitoring can be divided into detecting trends, surveillance, impact assessment, or a legal regulatory aspects. These considerations predicate to a large extent the technical plans for the assessment or program such as what variables are included, which methods are used, the intensity of data collection, statistical resolution, etc. Since ecosystem risk assessment and ecologically based monitoring are predicated on the assumption of anthropogenic activities, the technical information should be planned and formatted for use in the social, political, and legal arena. Scientists tend to design risk assessment schemes and monitoring networks that are technically defensible with other scientists. While technical integrity is an important prerequisite to any scientific program, the development of procedures addressing the social, economic, legal, and political dimension is perhaps the least considered aspect in the design of environmental monitoring programs. Moreover, explicit treatment of management/science problem has received inadequate treatment in the technical literature (Shaeffer et al. 1985 and Perry et al. 1985). This problem is further compounded by the paradox of various agencies with environmental programs at all levels of government seeking to simultaneously cooperate and insure continuity of overall program goals and to differentiate their programs from one another to avoid the appearance of redundancy.

OCR for page 97
Opportunities in Applied Environmental Research and Development Examples of Existing Programs The assessment of ecosystem risk has two components: 1) delineation of environmental stress or threats, and 2) determination of ecosystem vulnerability and response. Table 1 lists examples of existing monitoring programs at the global, multinational, national, regional, and local levels. In general, the global and multinational programs address aerial or other transboundary transport and effects of pollutants, national programs address the environmental mandates of their respective agencies, and the regional programs address interagency needs within a recognized physical geographical boundary. If the data from these systems is to be used in any concerted fashion to aid in the formulation or selection of ecologically based risk analysis, there is a need for synthesis of data from a variety of sources. Especially at the regional level, there is usually data available from a multitude of existing monitoring programs that need to be collated and synthesized to understand the potential cumulative impacts on the ecosystems in question. Some of the existing programs listed in Table 1 and elsewhere have areas of geographical overlap and would need to be addressed in some sort of overlay fashion. Technical The second component of ecological risk assessment, that is determination of ecological vulnerability and response, has two areas of technical development that are emphasized here. First, there should be some method of determining what ecosystem processes and what types of ecosystems are most vulnerable to outside stress. While this has been done in a relatively simple way for lentic systems exposed to acid deposition by determining the Acid Neutralizing Capacity (ANC), determining what factors contribute to ecosystem 'robustness' needs to be considered in an environment with a multitude of stressors. Some of the properties of ecosystem stability discussed by Westman (1978) may be applicable for ranking the sensitivity to outside stress. Ecosystem characteristics such as total gross productivity, total biomass, redundancy of structural and functional elements, resistance of key species to environmental stress, and resilience as measured by the K or r reproductive strategies of key species, may be the explanation to classifying and ranking ecosystems as to their ability to absorb stress or their ability to recover stress. The EPA's Ecoregions program currently underway is a logical framework for answering these questions (Omernik 1987). Implicit in the Ecoregions approach, especially for aquatic systems (Larson and Hughs 1987), is the designation of 'Reference sites' where the characteristics for 'ideal' aquatic ecosystems within a given ecoregion are determined. I would like to add the concept of a natural resource inventory as elaborated in the chapters of Kim and Knutson (1986) to be selectively conducted within ecoregions considering both the potential stress acting on that ecoregion, as well as the social, economic, political, and legal considerations of the natural resources found there. With additional specifications the properties of ecosystem robustness or sensitivity could also be addressed. Moreover, the problems of temporal and spatial ecosystem variability can be addressed by a resource inventory in the context of the Ecoregions framework. With this resource inventory by ecoregion, a geographical information systems approach could assess cumulative impacts and link land-water ecosystems. Secondly, the field of ecosystem mensuration is overdue for a review and reevaluation with approaches towards ecosystem risk assessment and quality assurance and control aspects. Ecosystem assessments and responses of ecosystems to stress are most often made at the community level of organization. Community may be defined as the living portion of the ecosystem but in practice communities are operationally defined by the ecologist by the methods of data collection and the statistics that define the data. For example, in aquatic systems there are phytoplankton, periphyston, zooplankton, benthic macroinvertebrate, fish communities etc., each of which is specifically defined by methods of collection, degree of taxonomic resolution, and the statistics used to reduce and interpret the data. Community level indices have been one common method of attempting to reduce the large amount of data inherent in community level studies. Table 2 lists some of the current community level indices in use by ecologists in interpreting community level data. While these indices have been popular among both theoretical and

OCR for page 97
Opportunities in Applied Environmental Research and Development Table 1. Selected list of monitoring programs of different geographical scope. I Global Monitoring System   * International Geosphere/Biosphere Program (IGBP)   * Global Environmental Monitoring Systems (GEMS)   * World Meterological Organization (WMO)   * Man and the Biosphere Reserves (MAB) II. Multinational Monitoring Systems   * European Monitoring and Evaluation Program (EMER)   * Integrated Background Monitoring Stations (IBMS) III. National Monitoring Programs   * National Acid Precipation Assessment (NAPAP)   * National Water Quality Assessment Program (NAWQA)   * Environmental Monitoring and Assessment Program (EMAP)   * National Surface Water Survey (NSWS)   * National Contarninant Biomonitoring Program IV. Regional Monitoring Programs   * Mussel Watch   * Multidecade Monitoring Program of Chesapeake Bay   * Long-Term Resource Monitoring Program, Upper Mississippi   System (LTRMP) V. Local   * National Park Service Inventory and Monitoring Program   * National Forest Service   * Other Federal, State, and Local Governmental Agencies

OCR for page 97
Opportunities in Applied Environmental Research and Development Table 2. Community level indices currently in use in ecological studies. A. Structural based indices 1. Diversity 2. Similarity (>20) B. Saprobian based indicies (invertebrates) 1. Chandler's 2. Chutter's 3. Hilsenhoff's C. Biological based indices 1. Index of Biological Integrity (IBI) (Fish) 2. Insect Community Index (ICI) 3. Biotic Condition Index (BCI) (Insects) applied ecologists seeking to measure the basic structure analyze the response of natural communities to various influences, the field of community metrics remains relatively poorly technically developed. One fundamental property of community structure is the taxonomic enumeration and how the number of individuals within the community is distributed among them. The structurally based diversity and similarity indices in Table 2 (Washington 1984, Cheetham and Hazel 1969) appear to be differentially sensitive to both the initial structure of the community and also the manner in which the community is changed (Boyle et al. 1989). While the principal deficiency in their application is similar to a Type H statistical error, that is; no response or paradoxical response to rather drastic changes in a community structure. The Saprobin based indices are dependent on the ranking of the relative sensitivity invertebrate taxa to high organic loading or low dissolved oxygen level—sewage impacts (Chandler 1970, Chutter 1972, Hilsenhoff 1977). These indices may give erroneous readings to community change due to environmental stress when the relative sensitivity among the taxa in the community is different than the Saprobian system. These indices also suffer from regionalism,' that is the ranking system for taxa sensitivity to the strees is not necessarily easily directly applicable from one region to another. The Index of Biological Integrity (IBI) incorporates a number of biologically interpretable metrics of the fish community into a single index (Karr et al. 1986). This index has been successfully adapted to an number of different regions within the United States. It does appear to be sensitive and able to detect a number of direct and indirect environmental stress. However, one documented potential weakness in using the IBI for assessment or monitoring is that while the fish community is sensitive to a wide variety of environmental conditions the index appears insensitive to the early stages of some type of stress (Berkman et al. 1987). The Insect Condition Index has not received wide enough application or use for evaluation here. The Biotic Condition Index (Winget and Mangum 1979) is an attempt to rank the relative sensitivity of aquatic insect taxa to several environmental parameters in streams. It has similar limitations of applicability as the Saprobian based indices. BIBLIOGRAPHY Berkman, H.E., C.F. Rabeni, and T.P. Boyle. 1987. Biomonitors of stream quality in agricultural areas: fish vs. invertebrates. Environ Management 10:413-419. Boyle, T.P., G. R. Smillie, J.A. Anderson, and D.R. Beeson. 1989. A sensitivity analysis of nine diversity and seven similarity indices. J. Water Pollution Control Fed. In Press. Chandler, J.R. 1970. A biological approach to water quality management. Water Pollution Control 69:415-421.

OCR for page 97
Opportunities in Applied Environmental Research and Development Cheetham, A.H.J. and J.E. Hazel 1969. Binary (presence/absence) similarity coefficients. Journal of Paleontology 43:1130-1136. Chutter, F.M. 1972. 1972. An empirical biotic index of water quality in South African streams and rivers. Water Res 6:19-30. Hilsenhoff, W.L. 1977. Use of arthropods to evaluate water quality in streams. Technical Bulletin No. 100 U.S. Department of Nature Research 16 pp. Hughs, R.M. and D.P. Larsen 1988. Ecoregions: An approach to surface water protection. J. Water Pollution Control Fed 60:486-493. Johnston, C.A., N.A. Detenbeck, J.P. Bonde, and G.J. Niemi. 1988. Geographical information systems for cumulative impact assessment. Photogrammetric Engineering and Remote Sensing 54:1609-1615. Karr, J.R., K.D. Fausch, P.L. Angermeir, P.R. Yant, and I.J. Schlosser. 1986. Assessing biological integrity in running waters: A method and its rationale. Special Publication No. 5. Illinios Natural History Survey. Champaign, IL. Kim, K.C. and L. Knutson. 1986. Foundations for a national biological survey. Association of Systematics Collections 215 pp. Omernik, j.m. 1987. Ecoregions of the conterminous united states. Ame. Geogr 77:118-125. Perry, J.A., D.J. Schaeffer, H.W. Kerster, and E.E. Herricks. 1985. The environmental audit ii: Application to stream network design. Environ. Management 9:199-208. Schaeffer, D.J., H.W. Kerster, J.A. Perry, and D.K. Cox. 1985. The environmental audit i: Concepts. Environ. Management 9:191-198.

OCR for page 97
Opportunities in Applied Environmental Research and Development This page in the original is blank.