Making ecosystem reality checks the status quo
2012; Wiley; Volume: 31; Issue: 3 Linguagem: Inglês
10.1002/etc.1747
ISSN1552-8618
AutoresG.A. Burton, Dick de Zwart, Jerry Diamond, Scott D. Dyer, Katherine E. Kapo, Matthias Liess, Leo Posthuma,
Tópico(s)Heavy metals in environment
ResumoEnvironmental Toxicology and ChemistryVolume 31, Issue 3 p. 459-468 Focus ArticleFree Access Making ecosystem reality checks the status quo G. Allen Burton, G. Allen Burton burtonal@umich.edu Search for more papers by this authorDick De Zwart, Dick De ZwartSearch for more papers by this authorJerry Diamond, Jerry DiamondSearch for more papers by this authorScott Dyer, Scott DyerSearch for more papers by this authorKatherine E. Kapo, Katherine E. KapoSearch for more papers by this authorMatthias Liess, Matthias LiessSearch for more papers by this authorLeo Posthuma, Leo PosthumaSearch for more papers by this author G. Allen Burton, G. Allen Burton burtonal@umich.edu Search for more papers by this authorDick De Zwart, Dick De ZwartSearch for more papers by this authorJerry Diamond, Jerry DiamondSearch for more papers by this authorScott Dyer, Scott DyerSearch for more papers by this authorKatherine E. Kapo, Katherine E. KapoSearch for more papers by this authorMatthias Liess, Matthias LiessSearch for more papers by this authorLeo Posthuma, Leo PosthumaSearch for more papers by this author First published: 24 February 2012 https://doi.org/10.1002/etc.1747Citations: 22AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Holistic approaches to assessing stressors and managing aquatic ecosystems should be the rule; instead, they are the exception. Disjointed, overlapping, and competing environmental regulatory actions—all with the noble mission of protecting and restoring the environment—can no longer be justified. For at least 60 years, environmental regulatory programs in the United States, Europe, and other developed countries have relied heavily on various forms of assessing chemical risk to manage and protect ecosystems. Water quality, primarily focused on chemical regulation, emerged from the need to control water pollution problems caused by poor or nonexistent wastewater treatment. The result has been largely a single-chemical approach to environmental management and regulatory programs. To protect aquatic life uses for example, many countries developed water quality criteria for selected priority compounds. Legally enforcing these criteria (such as the Clean Water Act in the U.S.) has undoubtedly reduced chemical pollution, and many aquatic systems have benefited. Abundant information demonstrates, however, that single-chemical standards are just one approach to assess, manage, and regulate aquatic systems. For example, toxicity testing (e.g., the U.S. Environmental Protection Agency's [U.S. EPA] whole effluent toxicity program [WET]) has been used successfully to help assess effects of chemical interactions and the effects of unknown chemicals that may be present. Such testing, however, addresses only direct toxicity effects. Many aquatic systems are impaired by non-chemical stressors, including invasive species, habitat degradation from agriculture and urbanization, and flow modifications or are influenced by complex interactions among chemicals and other stressors (e.g., nutrients) that are not addressed using either a single-chemical approach or mixture toxicity testing. The Current Conundrum Using a suite of indicators that address many types of stressors or sources of ecological impairment is critical for improving environmental management. Approaches that integrate information obtained from different indicators that lead to a sensible and efficient plan resulting in improved environmental conditions are needed. Numerous weight-of-evidence (WoE) approaches have been suggested but rarely are these used consistently in regulatory programs. For example, the U.S. EPA's Causal Analysis/Diagnosis Decision Information System (CADDIS) approach provides a framework for linking stressors with observed biological impairment (www.epa.gov/caddis). Environment Canada's Environmental Effects Monitoring Program also considers multiple stressors using several types of endpoints. The European Union's Water Framework Directive (WFD) describes an integrative approach that considers other stressors and legislation, yet new directives and regulatory enforcements still focus on a single-chemical approach. Recently, the U.S. EPA released a draft document—"Identifying and Protecting Healthy Watersheds: Concepts, Assessments, and Management Approaches" (www.epa.gov/healthywatersheds)—to address the need for integrative efforts. This is a step forward, but reinforces that wide-ranging approaches to holistic management have not significantly impacted the regulatory process—at least to date. The norm across North America and Europe remains a single-chemical approach to regulate aquatic ecosystems. Indeed, the National Association of Clean Water Agencies (NACWA) noted that a "meaningful, functional watershed approach remains elusive" and identified several obstacles to implementing such an approach, including addressing agricultural nonpoint sources, adaptive management based on science, and breaking down programmatic regulatory and enforcement silos 1. In general, the U.S. EPA and individual states have implemented Independent Application, which requires that each of the three measures of integrity—chemical criteria, WET, and biocriteria—be evaluated separately. Each of these measures alone, however, fails to answer questions about biological health effectively. By combining physical habitat, chemical conditions, WET, and biological assessment data, a more realistic assessment of biological community health and impacts will result. This integrated, multiple-stressor approach could help ensure that resources are directed toward mitigating the most limiting stressors, whether they are physical or chemical, to improve biological communities. Several environmental lawyers were surveyed in preparing this article. All agreed that water quality regulatory programs are driven by broad-based comparisons of chemical concentrations with established water quality standards. Further, regulations tend to emphasize assessing and controlling point source discharges and disregard other stressors in bodies of water being examined (G.A. Burton, University of Michigan, unpublished data). Such a policy approach raises a crucial question: Are our regulations effective at managing ecosystems exposed to mixtures of chemicals, let alone a wide range of chemical, physical, and biological stressors? In our view, focusing on chemical water quality, based on complying with single-chemical standards, results in oversimplifying complex factors that cause undesirable ecological conditions. What results is a distorted view of the importance of point sources and chemical factors. Examples abound in which regulatory management actions have been driven based solely on numeric chemical thresholds being exceeded. These actions result in questionable environmental protection or limited restoration success. Certain risk-driven criteria based on single-species tests, for example, may be much lower than natural background concentrations, such as for naturally enriched constituents such as aluminum, copper, and other metals. According to risk-based criteria principles, this could lead to policies designed to reduce national emissions, and, indeed, this has been the case. An alternative would be to elevate the standards so they do not fall below background concentrations and not use risk-based criteria for the aforementioned constituents. For example, the Alaska Department of Environmental Conservation has developed an approach to address certain naturally elevated metals. The Louisiana Department of Environmental Quality has developed a similar approach to address naturally depressed dissolved oxygen regimes in habitats such as bayous. From a broader perspective, using additional regional ecological information regarding the origins of certain compounds (an approach used to develop nutrient criteria or thresholds in some U.S. ecoregions, for example) would be a more realistic risk management approach that considers local natural background concentrations. Solely using a risk-based method, applied nationwide, would not help reach environmental goals in these cases. Another example is the U.S. water quality assessments such as the National Water Quality Inventory Report to Congress, which characterizes information as reported by each state, routinely includes more stream miles impaired by sedimentation and other habitat-related stressors than any other cause, including chemicals. The causes of increased sedimentation include changes in hydrology, increased impervious surfaces, improper storm water management, and farming practices that promote excessive erosion. A single-stressor focus to regulating aquatic ecosystems (i.e., a chemical standards focus) undermines introducing approaches that would prioritize stressors or considering cause-effect relationships. As a result, the bureaucratic process—such as measuring which chemical standards are exceeded—drives aquatic ecosystem management and restoration rather than ecological and toxicological principles. Yet doesn't it make sense to consider an integrated policy cycle for water management that takes into account diverse factors that affect ecosystem quality? Might this be more effective than continuing to regulate and manage ecosystems via separate programs? If so, what would be needed? Our Solution: Ecosystem Reality Checks Ecosystem Reality Checks (ERC) should be a required step between assessing risk (the science) and management actions (society). They should be holistic, as promoted in CADDIS and other eco-epidemiological approaches that rely on WoE. This means they should address ecological status (endpoints) relevant to the sources of stress and incorporate outcomes that propose alternative, iterative management options. A generalized framework for an ERC process is shown in Figure 1. From the center to the end, the color ramp represents an increase in scientific inquiry, regulations, and intra- and interagency involvement to confirm ecosystem diagnostics, that is, the ERC. The beginning, at the center of the cycle, is a single regulatory program with the goal of reducing risks to a single chemical. To begin the process, fate and toxicity data are prepared to predict risks in the field. A retrospective comparison with field data yields knowledge gaps that lead to more effective predictive models, such as refinements in exposure (e.g., equilibrium partitioning, bioavailability predictors) and risk. A more holistic analysis may indicate that simply understanding one chemical may be insufficient to account for, predict, or achieve a desired ecological status. Hence, in an ERC, mixtures of chemicals are considered, which requires temporally and spatially explicit exposure assessments that can be overlaid with information on species presence based on known behaviors, typically requiring tools such as geographic information systems. A result of these analyses can be the probabilistic assessment of a suite of chemicals within a specific geographical space and time. When compared with appropriate biomonitoring data, these data determine the coherence of the "prospective" tools. It is at this point where agencies focused on chemical regulations must partner with agencies responsible for factors important for ecological status. Multiple nonchemical stressors and their relationship to species exposures and responses are integrated into the chemical risk predictions. This then allows us to determine the relative risk or hazard of each stressor in terms of their contribution to ecological impairment. Through interagency cooperation, each stressor is managed appropriately, which leads to achieving the desired ecological status and in essence confirming the diagnosis. This process represents an Ecological Reality Check. Figure 1Open in figure viewerPowerPoint How the Ecosystem Reality Check cycle has generically been implemented. From the center to the end, the color ramp represents the increase in scientific inquiry, regulations, and intra- interagency involvement to confirm ecosystem diagnostics. EqP = equilibrium partitioning; GIS = geographic information systems. The Single-Chemical Versus a Multi-Stressor Approach The importance of a multiple-stressor perspective (the Environmental Reality Check [ERC]) in assessing and managing watersheds is apparent in a case study performed for the Hocking River watershed (Ohio, USA). A general description of the Ohio data sources and the statistical method used for the second and third approaches can be found in Kapo et al. 11. The case Using several environmental data resources, three different approaches were applied to identify sites in the Hocking River watershed at risk for impacts to their fish communities based on the environmental conditions present. The results of each approach were compared with real-world observations (sites with known biological impact) to determine how well each approach captured ecosystem reality (Fig. 2). 2. A weight-of-evidence based geographical information system evaluation demonstrates how different lines of evidence yield different conclusions regarding which waters are impaired. The combination of habitat, water chemistry, and landscape yield more realistic assessments of stressor-response relationships 11. The investigations The presence of ecological impacts was based on biomonitoring surveys in the watershed conducted by the Ohio Environmental Protection Agency. Three approaches were applied to predict the real-world distribution of ecological impacts in the watershed as a function of cumulative risks. First, the traditional single-chemical approach identified sites having one or more water quality criteria exceedances for metals. This approach demonstrated the weakest agreement with ecosystem reality, because the vast majority of real-world impacted sites were left unaccounted for. Second, a biomonitoring and mixture toxicity approach, which incorporated cumulative metals (mixture) toxicity, fared significantly better at identifying impacted sites than the single-chemical approach, in part because it incorporated cumulative risks (here modeled by an ecotoxicological response addition model, applied to all metals present at the sites). This approach incorporated a retrospective component by using statistical relationships statewide between bio-monitoring data and mixture toxicity to identify sites where ecological impact was likely to occur. Third, the most successful identification of impacted sites was achieved using a multi-stressor approach in which a variety of stressor variables in addition to cumulative metals toxicity were evaluated. Significantly, the multi-stressor approach demonstrated that the specific influence of metals toxicity across the impacted sites was greatly outweighed by other sources of potential stress, such as aspects of general physical and chemical factors in the watershed. The implications The three assessment approaches correspond to potentially very different management strategies. Two of the approaches (the single- and refined-chemical approaches) focus only on reducing metal emissions and exposures for a limited number of sites with observed impacts. These two approaches are characterized by incomplete recovery as a result of neglected stressors and neglected sites. One of the approaches—the multi-stressor approach—evaluates the ecosystem from a wider perspective to identify the most important stressors and handle them accordingly. This approach to assessing stressors best reflects ecosystem reality and offers the most practical guidance for subsequent management actions. Implementing a system of ERCs could condense disparate schemes from different regulatory fields, with the potential to grow it into an overarching regulatory paradigm. Such a system could explicitly consider the role of chemical contamination in light of multiple stressors in human-altered aquatic systems. An ERC is necessary because chemicals, whether singly or in mixtures, are just one class of stressors that affect aquatic biota and, in many cases, are no longer the chief cause of ecosystem impairments. Complementary Approaches to Evaluating Stressors Currently, protecting our waterways largely follows a chemical-specific approach (see sidebars, Single-Chemical Versus a Multi-Stressor Approach and Tidal Surface Waters in The Netherlands: Masking Nonchemical Stressors). Typically, water and sediment quality assessments are conducted in response to a regulatory directive, rely heavily on chemical benchmarks, and use standardized toxicity and bioaccumulation tests 2. It is also well recognized, however, that assessing ecosystem and sediment quality that integrates multiple methods into a WoE-based approach can be more accurate and more useful from a resource management perspective. Even these multiple-stressor approaches, however, are often conducted in a disjointed or inconsistent manner, such that they may fail to establish stressor causality 3. For example, site characterizations of the quality of physical and chemical factors (e.g., suspended solids, sedimentation, nutrients, toxicity), and biological status (structure, function) may change throughout the year. Sampling for each of these factors, though, may be relegated to convenient sampling periods (summer and fall, lower flows) and locations (e.g., bridges) that are not necessarily relevant to the spatial and temporal dynamics of each factor. Hence, protecting biotic communities may be unsuccessful because decisions and actions have been based on data collected in time and space that do not coincide with actual stressor events, whether they are single or multiple. Obviously, this has marked repercussions on how to improve, restore, or remediate the ecosystem. While the single-chemical management approach has resulted in many success stories since the 1970s, continued successes appear to be partial and less common today; indeed, they are perhaps more coincidental than intentional. Still, chemical pollution has been tagged recently as one of nine major drivers of global concern in an exploration of safe planetary boundaries of major stressors 4. Methods are lacking, however, regarding how to determine the reality of these drivers' significance alone or specifically when considered in conjunction with the planetary boundary on biodiversity loss 5. In other words, a chemical focus is insufficient because other factors need to be considered to manage ecological status. Tidal Surface Waters in The Netherlands: Masking Nonchemical Stressors The importance of a multiple-stressor data analyses in assessing and managing national assessments is apparent in an assessment of chemical risk of Dutch sediments. The case A biomonitoring dataset was collected for sediments in Dutch surface waters. The acute toxicity of the local chemical mixtures ranged from very low to 40% exceedence of species' EC50 values. It was expected, therefore, that toxic effects would be visible when samples were ranked according to increasing toxicity. The investigations To reveal these expected impacts, simple product-moment correlations were derived between mixture toxic pressures in sediments and taxa abundance, for nearly 100 sediment taxa (per taxon). Thereafter, these correlation values were rank ordered, starting with taxon with the most negative correlation to those with the most positive correlation. A significant correlation was found only for a low fraction of taxa (Fig. 3A). As expected, some impacts were negative (chemical loads reduced abundance of these taxa, left), whereas some were positive (abundance increase at increased chemical loads, right). It could be (wrongly) concluded that no mixture impact was present for the remaining taxa (middle). Yet just the opposite is true. A statistical model (generalized linear model) that describes a taxon's abundance data in relation to a suite of potentially relevant abiotic habitat characteristics (including the toxic pressure of mixtures) indicated that the field abundance of 74% of the species appeared to be significantly influenced by chemical mixtures. Again, both increasing and decreasing impacts were found, as well as "optimum" abundance trends (bell-shaped abundance change with increased chemical loads). The combination of generalized linear models with Monte Carlo simulation showed the sensitive, opportunistic, and optimum species (Fig. 3B). 3. (A) Rank-ordered correlation between toxic pressure of mixtures (msPAF-EC50) and taxon abundance for 103 taxa (279 sites) showed an indifferent response for most taxa (blue). (B) Field-based exposure-abundance patterns for a random subset of taxa derived from generalized linear models + Monte Carlo simulation showed that 74% of all taxa exhibit significant sensitive (negative), optimum, or opportunistic (positive) abundance changes with increased toxic pressure. The implications The implication is that a predicted toxic pressure of 40% (for 40% of species the EC50 would be exceeded) apparently does not show up as clear abundance reduction for an equivalent percentage of species through simple correlation techniques; indeed, they remained masked by other stressors' effects. The ERC that was applied—through the generalized linear models and the further data analyses—showed that systematic abundance changes associated with chemical mixtures occurred in most species and were thus apparently masked by other stressors. The example suggests a high association between predicted and observed fractions of species that were negatively affected (nearly 1:1), followed by major opportunist responses. The Ecosystem Reality Check conducted here has policy relevance, such as via the European Water Framework Directive. In that directive, the policy aims are to reach Good Chemical Status as well as Good Ecological Status. Deviations from Good Ecological Status ask for evaluating the possible impacts of mixtures of chemicals on local species assemblages. In the context of ERC, an approach such as a validated toxic pressure approach can be applied as a lower-tier approach to quantify the fraction of species affected by a local mixture, which expands the classical approach of evaluations via Water Quality Criteria. What Is Blocking the ERC Process? In ecosystems for which environmental management decisions have been made to remediate and restore waterways, it is both surprising and disconcerting that few cases have demonstrated improvements in ecosystem quality in recent years. Most stream restoration or remediation efforts in urban-dominated watersheds have been unsuccessful because important stressors have not been removed or inadequate refugia exist 6. Indeed, 80% of benthic taxa decline in a wide range of watersheds when impervious areas are only 0.5 to 2% of the area 7. Routine monitoring of chemical standard exceedances does little to account for impairments due to urban runoff. In addition, chemical standards may be attained, but without source populations available, restoration efforts fail. Current management efforts are often focused on restoring physical habitats or removing contaminated sediment mass and fail to take a holistic approach to restoring the biological integrity of our waterways. If an ERC approach had been established to assess both baseline conditions and the role of other stressors and their sources, then corrective actions could have facilitated successful (and measurable) remedies and restorations 5. So why don't we apply ERCs yet? Is there a block in science, in policy, or both? Crossing the Rubicon: The Stepping Stones Before attempting to solve a problem, risk assessors commonly frame the problem. A key issue is this: What are the current regulations on which to establish an ERC? What stepping stones are in place to support or launch the ERC as a mindset or method? In addition to chemical-oriented policies, such as the Toxic Substance Control Act (TSCA) and Federal Insecticide Fungicide and Rotenticide Act (FIFRA), the U.S. EPA has divisions devoted to the safety of environmental compartments or species groups of concern, such as surface waters, ground water, and terrestrial organisms (for example, the Office of Water, Office of Solid Waste and Emergency Response, and the Office of Research and Development). Their efforts are often relatively uncoordinated, and thereby unable to iteratively assess the aggregate and potentially cumulative effects of multiple chemicals, let alone other stressors. Although the U.S. EPA's CADDIS methodology is highly flexible and has, as mentioned, the potential to affect corrective management, its connection to chemical and other regulatory environmental policies and, more specifically, water regulatory programs in the United States, has not been delineated clearly. Thus far, CADDIS has been used mostly in special cases within a regulatory or management context, for example, certain Total Maximum Daily Loads [TMDLs]). While the U.S. EPA has jurisdiction over most chemical management policies in the United States, other agencies manage other stressors, such as the U.S. Department of Agriculture (soil till management, forest management), the U.S. Food and Drug Administration (pharmaceuticals), and National Oceanic and Atmospheric Administration (marine ecological status). In essence, there is no coherent coordination in the United States—let alone among regional, state, and local agencies—to avoid pitfalls and holistically manage relevant stressors that contribute to ecological status. The situation in Europe is similar. The European Union has enacted a suite of regulations, much like those in the United States, but there is room for improvement. A recent illustration of this is provided in a cross-regulations evaluation of how chemical mixtures are handled in risk assessment and management 8. Many regulations handle mixtures differently and are based on different underlying scientific views and arguments. Sometimes, a regulation looks at single compounds only because scientific methods for handling mixtures were deemed absent at the time the regulation was enacted. In other cases, subgroups of compounds are considered on the basis of assumed similar mechanisms of action, which is then taken as justification for partial mixture modeling and risk assessment. This evaluation concluded that for Europe, "There is sufficient know-how [in terms of dealing with mixtures, yet]…the question as to how this scientific knowledge might be best transferred into appropriate regulatory approaches is, however, not at all trivial…". This evaluation stipulated that "…consistent and clear [policy] mandates are needed to take mixture toxicity into account in the numerous pieces of legislation that contribute to the protection of…the environment from chemical risks…" 8. These statements refer to 21 relevant EU substance- or product-oriented pieces of legislation. The picture in Europe thus becomes more entangled when one looks at overarching regulations such as the Water Framework Directive, the Marine Strategy Directive, and the proposed Soil Directive. The Kortenkamp et al. report 8 suggested focusing on receptors experiencing multi-exposures rather than focusing on sources. Not all chemicals co-occur in the field, however. Using an ERC approach, mixture risk assessments are needed only for those compounds that are likely to co-occur in realistic field settings. Notably, a similar conclusion was drawn more or less independently in the recent EU project named NoMiracle (http://nomiracle.jrc.ec.europa.eu/default.aspx), a project designed to increase knowledge on the transfer of pollutants between different environmental compartments and on the impact of cumulative stressors, including chemical mixtures. In short, regulatory action—at least in the United States and Europe—is triggered along separate regulatory lines, in which scientific foundations vary and are not viewed in the broader context. Nevertheless, there is a way forward by using a growing focus on the final policy target: The overall impacts in aquatic systems. This more holistic approach to environmental regulation is possible through Registration, Evaluation, Authorisation, and Restriction of Chemical substances (REACH), the EU's chemical-focused regulation. REACH starts from the chemical perspective, whereas the Water Framework Directive (WFD) focuses on the water body. REACH aims to prevent and limit the emissions of the most toxic compounds. It refers to possible accumulation of compounds that are produced and used, such as specific mixtures that accumulate at downstream locations. The regulation allows for the idea of addressing such sites of concern and specifically mentions a link to the WFD. Hence, the crucial regulatory mandate is a very short portion of the law, but it is there to act as stepping stone. The WFD adds to REACH through the Good Chemical Status approach, which identifies selected priority and basin-specific compounds for phase out and the integral focus on the net effects of all stressors. Water authorities are required to act, for example, through River Basin Management Plans, to reach the holistic goal of "Good Ecological Status." What can and cannot be prevented in source-oriented policies is captured by the compartment-oriented policies, which are cross-linked. Even so, mandating stepping
Referência(s)