Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 A Framework for the Analysis of Monitoring The previous chapter documented the wide range of monitoring pro- grams being carried out in the Southern California Bight. Because these programs can be evaluated from many different perspectives, it is important to clarify the criteria the panel used in its analysis of monitoring efforts. These criteria summarize the conceptual framework developed by the par- ent committee. They provide the basis for determining whether individual programs, as well as the monitoring system as a whole, are effective or not, and can be expressed as six questions: 1. Does monitoring address clearly stated management and societal objectives? 2. Does monitoring address the major environmental problems facing the bight? 3. Do the spatial and temporal scales of monitoring reflect those of the major environmental problems? 4. Are the technical design and implementation of monitoring of high quality? This includes proper statistical design of sampling and analysis, use of state-of-the art field and laboratory techniques, and adequate links to relevant research programs. 5. Do monitoring programs respond in a timely way to changing conditions and needs? 6. Are monitoring resources allocated effectively both within and among monitoring programs? These criteria reflect the literature on monitoring (e.g., Holling, 1978; 97
98 Green, 1979; Beanlands and Duinker, 1983; Fritz et al., 1980, National Research Council, 1986; Isom, 1986; Rosenberg et al., 1981; and Bernstein and Zalinski, 1983, 1986) and the experience of the panel members. It is important to recognize that issues addressed by the evaluation criteria are not strictly technical. This is because monitoring is defined by and carried out within a complex context that includes the interests and information needs of the public and the regulatory agencies and the requirements (procedural and otherwise) of relevant laws and permits, as well as strictly scientific and technical concerns. The analysis of monitoring must therefore look as much at the interface between policy and technical issues as at the technical issues themselves. The following sections address three areas that are especially relevant to the analysis of monitoring and that underlie the evaluation criteria: . the importance of clear objectives, · the role of technical design and its statistical component, and · the necessity for identifying, evaluating, and prioritizing environ- mental problems. THE IMPORTANCE OF OBJECTIVES Monitoring programs are intended to produce information for quan- tifying and evaluating the effects of human activity on the marine en- vironment. Monitoring is intended to provide decision makers with the information they need to make appropriate management decisions about how to protect the marine environment and its resources. Ideally, these information needs should be expressed as objectives that guide the design and implementation of monitoring programs. The objectives that currently motivate monitoring programs in the bight can be loosely structured as a hierarchy. At the highest level are broad concerns about human health and the status of the ecosystem. Beanlands and Duinker (1983) make the point that objectives at this level often reflect sociopolitical values that cannot always be quantified or supported scientifically. This, however, does not necessarily lessen their importance or relevance as the basis of management and monitoring efforts. At the next level are the laws and regulations that embody these concerns as more specific objectives or requirements. At the next level are permits for individual discharges or other activities, which in some cases contain numerical monitoring criteria. Finally, the monitoring design itself is based on decisions about what, specifically, to measure, when, where, and how often to measure it, and about what degree of uncertainty in the final answer is acceptable. Ideally, each level should incorporate the content and intent of the preceding level. Westman (1985) has described an analogous
99 hierarchy in terms of successively more specific and detailed goals, policies, strategies, and tactics. As the foregoing discussion implies, clear objectives are crucial for both the monitoring and decision-making aspects of environmental management. For monitoring practitioners, they direct monitoring efforts toward the measurement of specific parameters and of specific amounts and rates of change. Without such clear objectives, it is impossible to effectively use such technical design tools as conceptual, numencal, and statistical models, and power and optimization analyses. For managers and regulators, they provide a standard against which environmental change can be measured in order to determine if corrective action is required. It is therefore necessary to completely specie objectives at each level of the hierarchy, from broad public concerns to specific, numerical criteria. THE ROLE OF TECHNICAL DESIGN Technical design involves making decisions about what to monitor; how, when, and where to take measurements; and how to analyze and in- terpret the resulting data. The parent Committee on a Systems Assessment of Marine Environmental Monitoring developed a design methodology that the panel used to structure its evaluation of this aspect of monitoring in the Southern California Bight (Figures 5-1 to 5-4) (National Research Coun- cil, 1990~. Figure 5-1 shows that technical design must be considered in relation to the initial definition of goals and objectives and the ultimate effective dissemination of monitoring information. Figures 5-2 to 5-4 pro- vide additional detail about the relationships among specific elements of the methodology. The methodology summarized in Figures 5-1 to 5-4 reflects definite concepts about effective monitoring design and its benefits. These concepts are not the only ones that could have been used to structure an evaluation of the technical design of monitoring programs. They do, however, reflect many of the important themes that recur in the literature on monitoring design. The following is a summary of these concepts: . Appropriate technical design ensures that data collection, analysis, and interpretation will address management needs and objectives. oTech- nical design can be performed adequately only when objectives, problems, questions, or hypotheses are stated explicitly. · Sampling, measurement, and analysis designs should be developed with the goal of detecting specific kinds and amounts of change. · Predictions about the kinds and amounts of change expected should be derived from conceptual models that specify how particular human activities (causes) will lead to environmental impacts (effects).
100 Refine Objectives l Step 1 _ Define Expectations and Goals Step 2 Define l Study Strategy Reframe Questions i ~ No Rethink Monitoring Approach Step 4 Develop Sampling Design Be Det~:t~? ~( Yes , Step 5 Implement Study 1 1 ~ 1 1 Step 6 Produce Information No ~ Is Information \ Adequate? Yes Step 7 Disseminate Information Make Decisions Burr N;;~ \ Conduct Exploratory § Studies if Needed FIGURE 5-1 The elements of designing and implementing a monitoring program. · Sampling and measurement designs should account for important sources of natural variability. · Sampling and measurement designs should be evaluated before- hand to determine their ability to detect predicted changes. · Analysis approaches should be selected before data collection to correspond to the statistical assumptions of the sampling design. · Data base systems should make authorized versions of the data readily available to analysts and managers, and should provide easy access to a wide range of analysis, graphics, and reporting tools.
101 Identify Public Identify Relevant Concerns and Laws, Regulations, and Expectations Permits ' ~' ' 1 ' . _ ,_ Focus Scientific Understanding 1 1 1t Establish Environmental and Human Health Objectives FIGURE 5-2 Step 1: Defining expectations and goals of monitonng. The technical design process illustrated in Figures 5-1 to 5-4 furnishes a framework for translating broad questions and objectives into specific decisions about what to measure, where to measure it, and how many measurements to take. Using this framework as an evaluation tool enabled the panel to use a common set of standards in considering the technical design of monitoring programs in the bight.
102 Identify Resources at Risk | Modify Resources No ~ Yes | Determine l | Appropriate Boundaries l I' Boundaries Adequate? / ~ Yes Predict Responses and/or Changes Are \ No - Predictions ~ ~ Yes Develop Testable Questions FIGURE 5-3 Step 2: Defining study strategy. ~ Develop Con eptual Model ~ 1 /Have Appropriate \ | Adjust l Resources Been ~Boundaries Selected? ~ A FRAMEWORK FOR PRIORITIZING PROBLEMS Refine Model As stated previously, this case study is oriented toward examining the monitoring system in the bight as a whole. In addition to evaluating whether individual programs meet their objectives, this necessitates de- termining whether the entire collection of monitoring programs produces
103 Develop Testable Questions Identify Meaningful Levels of Changes i ~' 1 Select What to Measure Reframe I . Questions ~ _ _ ~I Develop Monitoring Design ~I | Specify Statistical Models l No /n Predict Responses Be Seen' ~ Yes Rig fine ~ Technical Define Data Design Quality Objectives T Identify Logistical Constraints , Conduct Power Tests ' and Optimizations r .1 Develop Sampling Design No /s Design ~ Adequate? - ~~~ Yes FIGURE 54 Step 3: Developing sampling and measurement design. the information needed to satisfy both management and the public. The technical design methodology described above provided a means of struc- turing the analysis of individual programs. However, there was no similar framework available for the overall analysis of the monitoring system. The panel therefore adapted existing methods in order to perform a summary assessment of environmental impacts on resources in the bight. This as- sessment was intended to summarize the nature and severity of impacts on
104 a range of important resources in the bight and was designed to help the panel address specific questions. · Does monitoring address clearly stated management and societal objectives? · Does monitoring address the major environmental problems facing the bight? Do the spatial and temporal scales of monitoring reflect those of the major environmental problems? · Are monitoring resources allocated effectively both within and among monitoring programs? The Assessment Framework While many useful frameworks have been proposed for environmental assessments (see examples in Beanlands and Duinker, 1983; Westman, 1985; NRC, 1986), constructing one for monitoring in the bight in the context of the case study presented special difficulties. First, the goal of the assessment was to produce a synthetic overview that would aid in drawing conclusions about the entire monitoring system in the bight, both technical and institutional. This is in contrast to more typical assessments that focus only on identifying and quantifying the environmental impacts of individual projects. Second, the time available for developing this overview was necessarily short and the technical and financial resources available were limited. Third, there are extensive and diverse human and natural sources of perturbation in the bight and methods for characterizing multiple and cumulative impacts are not well developed. For example, effects on fish populations may derive from: · coastal power plants~ntrainment of larvae, impingement of adults; municipal wastewater outfalls-habitat alteration, changes in food supply, contamination; · dredged material disposal-habitat alteration, contamination; · storm runoff-contamination; and · sport and commercial fishing-increased mortality. · E1 Ninos~hanges in distribution and community structure' habitat alterations; and · major storms-habitat alteration. Such effects act on different spatial and temporal scales, and this adds to the challenge of understanding and portraying impacts. ~ accommodate these constraints and difficulties, the panel used a combination of matrix and ad hoc assessment methods (Westman, 1985. ~ The matrix approach was adapted from a framework developed by Clark (1986) for identifying
105 The assessment produced synoptic overviews that were useful in evaluating the overall pattern of monitoring in the bight. However, before reviewing the assessment products and explaining the supporting detail, it is important to understand the limitations of the matrix and ad hoc methods used. In most cases, the limitations of each method were somewhat balanced by the strengths of the other. The procedure described by Clark (1986) proceeds through a series of steps that specify: valued ecosystem components (VECs), · marine constituents (both natural ecosystem parameters and an- thropogenic contaminants) that cause changes in the VECs, and · sources of natural and human-induced perturbation that create or cause changes in these constituents, which are linked in a matrix with specific VECs to show how they along with contamination in the bight affect marine resources (Figure 5-5~. The selection of perturbations, constituents, and VECs is necessarily somewhat arbitrary. Given the size of the bight and the multiplicity of resources and sources of impact, some selection among these was unavoid- able. This selection reflects the values and biases of the panel, but the critical reviews by experts and scientists outside the panel were designed to balance competing points of view. However, there is no denying that other reviewers might have generated parameters that would have led to a different assessment. The matrices do not specifically identify primary, secondary, and higher order interactions among perturbations, constituents, and VECs. This would be a severe shortcoming if the matrices were used as a stand-alone assessment method. In this case, however, the matrices were used as a cross check for the conclusions derived from the ad hoc approach and to enforce a degree of systematic thinking. While the matrices themselves do not specify interactions, they were discussed at length during preparation of the matrices and as part of the ad hoc approach. The matrix products do not quantity effects and impacts. Rather Figure 5-5 scales two impact attributes, the potential influence of each source of perturbation and the degree of scientific certainty associated with this conclusion. This is similar to the scaling of impact magnitude and importance proposed by Leopold et al. (1971) in a similar matrix. This subjective scaling would be a major shortcoming if the panel's intent was to perform a damage assessment, a detailed project assessment, or a comparison of two or more alternative development scenanos. However, in cumulative impacts. The ad hoc portion of the assessment (Rau and Woolen, 1980) consisted of brainstorming sessions with experts and critical review of the matrix products by individual scientists. The matrix products were modified a number of times to incorporate feedback from brainstorming sessions and individuals' reviews.
106 VALUED ECOSYSTEM COMPONENTS SOURCES \ OF PERTURBATION \ 00 a) AC ~CO 1/ 0) ~= Storms ~ ~0 ~ El Ninos ~0~ ? ~ ~ ~ Upwelling ~ O ~ ? ~ Basin Flushing Mass Sediment Flows O O Elm Blooms/lnvasions Diseases of ? Ecological Interactions Q~ ~ ? ~ Power Plants HE ~ ~ ? Wastewater Outfalls ~ O Dredging River Flow and Storm water Runoff ~ ? O ~ ~ O 0~ Commercial Fishing Sport Fishing o To loo To ? ? 0 Marine Commerce and Boating Habitat Loss and Modification Oil Spills Oil Seeps Atmospheric Input POTENTIAL INFLUENCE E| Controlling ~ Major E] Moderate ~: Some ? - Some evidence for impact but further study needed Blank - no impact ASSESSMENT RELIABILITY FIGURE 5-5 Impacts on the marine environment of the Southern California Bight. Individual cells of the matrix illustrate the presumed relative impact of each source on each component, along with the associated scientific certainty. Each column represents cumulative impacts on individual components; each row shows the effects of individual perturbations on all components. This figure was used to summarize and investigate ways of identifying and ranking impacts in the bight. SOURCE: After Clark, 1986.
107 this case the panel's goal was to produce a high-level overview that would assist in comparing the overall pattern of impacts with the overall pattern and structure of monitoring programs. In addition, much of the background information used in both the matrix and ad hoc efforts was derived from extensive and quantitative research, monitoring, and modeling programs. The overviews that resulted from the assessment lack detail about the nature of the effects they represent. Again, this is less of a problem given the panel's task In fact, the high-level, summary character of the overviews was actually helpful in elucidating the weaknesses of the existing monitoring structure. The ad hoc method depends on the collected experience and insights of the participants. As a result, conclusions are dependent not only on the selection of participants but also on their values and biases. Under the circumstances, the panel believed that enlisting the participation of a cross section of scientists from the bight region was the most efficient means of integrating the wealth of scientific and technical information available. Involving scientists of differing affiliations helped to balance individual values and biases. In addition, the matrix method helped to focus, systematize, and cross check each person's opinions and judgments. No assessment method is perfectly objective. While quantitative mod- els are increasingly valuable, even they depend on certain simplifying as- sumptions and often are challenged. Similarly, even a moderately sized monitoring program must make judgments about which aspects of the envi- ronment to measure or ignore, since it is impossible to measure everything. The panel used the assessment products to derive conclusions about the structure and focus of the monitoring system in the bight. The conclusions were judged to be robust enough to form the basis for conclusions and recommendations, even in light of the acknowledged limitations of the assessment methods used. A Synoptic Overview The matrix in Figure 5-5 is a useful heuristic tool. It shows that all ecosystem components are impacted by more than one kind of perturbation. It also shows that perturbations typically affect more than one ecosystem component. For example, storms affect soft benthos, kelp beds, and human health; wastewater outfalls affect soft benthos, microheterotrophs, and demersal fish populations. Figure 5-5 helps categorize the types of monitoring programs in the bight. Some programs examine the effects of one perturbation on a single resource. These programs focus on one cell of Figure 5-5 and are called single-cell assessments. For example, the impingement sampling program carried out by the Southern California Edison Company is intended to
108 assess the potential impacts of coastal power plants on pelagic fish popula- tions. Other monitoring programs examine the effects of one perturbation on a range of resources. These programs focus on an entire row of Figure 5-5 and are called row assessments. For example, the 301(h) monitoring program around the Orange County wastewater outfall is designed to doc- ument the effects of the outfall on a range of resources, including soft benthos, water quality, and demersal fish populations. Monitoring pro- grams that consider how several perturbations, acting together, affect a single resource would focus on an entire column of Figure 5-5 and are called column assessments. There are no examples of such programs in the bight, a fact which will be addressed in more detail in Chapter 6. Further, there are no coordinated monitoring programs in the bight that focus on the effects of two or more sources of perturbation on a range of related resources. Such a program, for example, might document the combined effects of fishing, power plants, and wastewater outfalls on demersal and pelagic fish populations. Figure 5-5 also presents subjective judgments about the relative im- portance and degree of scientific certainty associated with each impact. For example, wastewater outfall impacts on soft benthos are more severe and extensive than those from dredging. As another example, it also shows that conclusions about kelp bed impacts are probably more reliable than those about effects on fish eggs and larvae. Such comparisons aid in analyzing existing monitoring programs by suggesting where further research would be more appropriate and useful than routine monitoring. As Chapter 6 makes clear, available financial and technical resources in the bight are not systematically allocated to research and monitoring on the basis of a comprehensive overview like the one in Figure 5-5. As with Figure 5-5, Figure 5-6 is a useful heuristic tool that supplies insights about the structure of existing monitoring programs in the bight. It shows quite clearly that the impacts that are relatively well understood (e.g., coastal power plant plumes, disposal of marse dredged material, nutrients, fine particles) are those whose scales are either less than or of the same order of magnitude as those of monitoring programs. It also demonstrates that, with the exception of the CalCOFI program, the temporal and spatial scales of individual monitoring programs are insufficient to resolve patterns of effects on larger scales. While the effects of scale are becoming a matter of concern to ecologists (Wiens, 1989), Figure 5-6 demonstrates that monitoring programs in the bight are not consistently designed with such scale effects in mind. As Wiens (1989) points out, these effects can be complex, and-if not considered carefully ". . . we may think we understand the system when we have not even observed it correctly."
109 Supporting Detail As the first step in the matrix assessment procedure, the effects of the constituents on the VECs are identified (Figure 5-7), and the ways in which sources of perturbation cause changes in these constituents are then specified (Figure 5-8~. This permits sources of perturbation to be linked (through changes in the constituents) directly to effects on VECs in a matrix (Figure 5-5~. This in turn allowed the panel to summarize the effects of various human and natural processes on the VECs. Finally, the temporal and spatial scales of constituents and perturbations (Figure 5-6) are compared to the spatial and temporal scales of relevant monitoring programs. Figure 5-7 qualitatively shows the effects of changes in marine con- stituents on valued marine ecosystem components. VECs include important ecosystem components and major fisheries, as well as demersal and pelagic fish life stages that occupy distinct habitats and might be affected differ- entially. Constituents are divided into physical oceanographic parameters (e.g., waves or temperature), and into Boating, dissolved, suspended, and settleable categories. Figure 5-7 shows that specific constituents impact more than one VEC and that some VECs are affected by more than one constituent. The constituents shown in Figure 5-7 were selected because they are typically measured in monitoring programs. Their division into floating, dissolved, suspended, and settleable categories reflects the fact that their association with particles of different sizes significantly influences the fates and effects of most contaminants. However, the selection and arrangement of these constituents is certainly not the only one possible. For example, rather than focusing on physical and chemical parameters, the constituents could include important dynamic processes, such as production, nutrient regeneration, the flux of organic matter, and recruitment and mortality. Figure 5-8 furnishes the next link in the matrix-based assessment by showing which sources of pertur~afion affect which constituents. This then permits connecting sources of perturbation to effects on VECs. For exam- ple, the amount and distribution of fine particles and nutrients are affected by wastewater outfalls (Figure 5-8), and such changes can potentially affect the soft benthos (Figure 5-7~. This suggests a potential mechanistic link between wastewater outfalls and effects on the soft benthos. Similarly, marine commerce and boating create floating debris (Figure 5-8), which af- fects marine birds (Figure 5-7~. (These admittedly simplistic examples were chosen for illustrative purposes; the reader is encouraged to investigate other links suggested by Figures 5-7 and 5-8~. These two figures can be integrated to furnish a synoptic view of the impacts of both natural and human perturbations on the VECs. Thus, one
110 Hour Day Week tAonth Year Decade Century 3 2 E y o - 1 ~: ~n o _- x _ ~NONMOBILE TEMPERATURE \ _- _ ~ CURRENTS DISSOL\/ED OXYGEN PATHOGENS* I ~ FINE PARTICLES NUTRIENTS POLAR ORGANICS ~ 1 PATHOGENS / / ~ I TEMPERATURE* I COARSE PARTICLES I ORGANOMETALLICS ~ \lOlATILE ORGANICS _ _ __ METALS HYDROPHOBIC ORGANICS COMPLEXED _ - METALS 4 3 -2 UJ LL i~ ~ LL -. 0 1 2 3 TIME (log YEAR) x Power Plants x Outfalls x Harbors/Marinas | Fishing I Dredging I Peak River Flow I I Storms I I EI Ninos I I Blooms I I Diseases I I Ecological Interactions 1 000 km 100 km 10 km 1 km 100 m
111 cO a' .- co a) a) ~5 c'd) cl7 c) ct ~ - - T°¢ 1 -! in I In In s ~ `,, Q > as I . . CD o - c) a, = c) ._ a) o o C> LL - cn i m0 _ . Z 111 Z LL tar: ~ En LLI ~ m J co 0 FIGURE 5-6 Characteristic temporal and spatial scales of important constituents and sources of perturbation. Constituents are the same as those in Figure 5-6, but have been abbreviated. "Temp *" refers to temperature changes from coastal power plants, and "Temp" to natural temperature changes (i.e., El Nino events). "Path *" refers to bacterial contamination from wastewater outfalls and storm runoff, and "Path" to pathogens from natural sources that cause diseases in urchins, fish, and other organisms. The abscissa represents a crude estimate of the half life or recurrence time of each constituent. The ordinate represents the spatial displacement likely to occur over that time, or the scale of activity. For example, nonmobile metals and hydrophobic organics are presumed to persist in the environment and spread much more widely than nutrients. The temporal and spatial boundaries of existing monitoring programs are outlined by a solid line, with the exception of the CalCOFI Program, whose parametem are indicated by an "X" in the upper right of the figure. Constituents with similar temporal and spatial scales are outlined with dotted lines. SOURCE: After Clark, 1986.
112 \ MARINE CONSTITU ENTS \ VALUED ECOSYSTEM co Q o C _ o o y _ C _ ~ ~Q o o s C o ~o C _ ca 0 4 Q o o o N cry (n o s m 0 ~ _ 13 C5 Q - a, . _ o, - U) s In {C At 13 a) C ~ V a, J`45 E E CO~ o) ~ IL0> sa: Inm: ._~ ~C c .m ~- a> <n CD Gk a' o ,_ at m I a, C C ~ - ~ I OCEANOGRAPHIC Currents Winds Waves Tem peratu re Dissolved Oxygen FLOATING Floating Debris DISSOLVED Nutrients Volatile Organics Polar Organics Complexed Metals Organometallics SUSPENDED Fine Particles Coarse Particles Nutrients Pathogens Hydrophobic Organics Organometallics Nonmobile Metals SETTLEABLE Fine Particles Coarse Particles Nutrients . Hydrophobic Organics Nonmobile Metals . · · · FIGURE 5-7 Mapr impacts of natural features of the ecosystem and anthropogenic contaminants on valued ecosystem components. The dots indicate that the listed constituent (left) is presumed to have a significant impact on that ecosystem component (top). Both direct and indirect impacts are included. Volatile organics include phenols and halomethanes; polar organics, PAH, ODT, and PCB. Nonmobile metals include lead and cadmium; complex metals-nickel and copper; organometallics mercury, tin, selenium, and arsenic. Nutrients include both dissolved nutrients and nutrients associated with particles. Pathogens include those from both anthropogenic (e.g., coliforms) and natural (e.g., urchin disease) sources. Dissolved constituents are those less than .04 him in size. Settleable constituents are defined operationally as those that settle to the bottom as a function of size and specific gravity. Valued ecosystem components include various parts of the food chain, communities associated with specific habitats (e.g., kelp beds), and important fisheries. Commercial shellfish include abalones, lobsters, and urchins. SOURCE: After Clark, 1986.
113 can start with VECs such as soft benthos or demersal fish populations, identify the constituents that affect them, and then trace these constituents back through their relationships with sources of perturbation to finally determine all the kinds of perturbations that affect these ecosystem compo- nents. The result of this process can be displayed as a matrix (Figure 5-5) that summarizes the impact of each kind of perturbation on each ecosystem component. Figure 5-5 was constructed using other knowledge from the ad hoc method in addition to the mechanistic linkages shown in Figures 5-7 and 5-8. This points up shortcomings in the selection and organization of the constituents shown in Figure 5-7. For example, Figure 5-5 shows that sport and commercial fishing impact pelagic and demersal fish by directly removing individuals from the population. However, since Figure 5-8 does not include mortality as one of the marine constituents, Figures 5-7 and 5-8 do not combine to predict impacts on fish from fishing, an obvious failing. In addition, Figure 5-5 indicates that blooms, natural diseases, and especially ecological interactions have significant effects on the VECs. However, Figure 5-7 shows that none of these important sources of perturbation interact strongly with any of the constituents other than temperature and dissolved oxygen. The panel thus combined insights from both the matrix and ad hoc methods without rigidly adhering to the limitations of either. Figure 5-5 is an informative way to organize existing knowledge about impacts on marine resources. However, the spatial and temporal scales of both perturbations and ecosystem processes vary widely and this informa- tion is necessary to evaluate the effectiveness of monitoring. The overall assessment framework therefore includes a means of organizing and com- paring the temporal and spatial scales of constituents and perturbations. A preliminary approach is presented in Figure 5-6. The constituents are placed in a logarithmic time-space coordinate system based on crude esti- mates of their half-lives in the marine environment (for contaminants) or their typical scale of activity (for ecosystem features). The temporal and spatial range of existing monitoring programs is indicated, and the temporal and spatial scales of important perturbations shown along the x and y axes, respectively. SUMMARY This chapter presents the criteria and concepts used to organize the analysis of monitoring efforts reviewed in the next chapter. Six key questions made up the evaluation criteria used to assess both individual monitoring programs and the collection of monitoring programs in the bight. These questions addressed both the policy and technical aspects of monitoring,
114 \ SOURCES OF \ PERTURBATION MARINE CONSTITUENTS \ .~ c ~ _ ~ ,5 ~_ o ~ ~ ·'- ~ ~ ' ~ ~ ~ o- ~ ~ ~ ~ ~ ~ 6 ~ OCEANOGRAPHIC Currents Winds Waves Temperature Dissolved Oxygen · · · · · ~ ~· · ~ · · FLOATING Floating Debris DISSOLVED Nutrients Volatile Organic$ Polar Or$ian,cs Complexed Metals Organometallics · · · · SUSPENDED Fine Particles Coarse Particles Nutrients Pathogens Hydrophobic Org&n~cs Orgenomet~llics Nonmobile Metals _SETTLEA~6 Fine Particles Coarse Particles Nutrients Hydrophobic Organics Nonmobile Metals · · · ~ ~ - · · ~ ~ · · · ~ ~ · ~ · · · ~ ~ ~ FIGURE 5-8 Sources of major perturbations to the bight's marine ecosystem and their major impacts on marine constituents. The dots indicate that the listed perturbation (top) is presumed to have a significant effect on the listed constituent (left). Both direct and indirect impacts are included. Perturbations include both human and natural sources of change. Basin flushing refers to the turnover of near-bottom water in offshore basins; mass sediment flows to sudden, large movements of sediment on the shelf; blooms or invasions to rapid increases in population levels of otherwise rare species (e.g., the echiuran Lis~iolobus or the kelp isopod Peniidothea resecta). Multiple sources of impacts of one kind (e.g., power plants, dredging) have been lumped to provide a consistent level of generality among perturbations. SOURCE: After Clark, 1986.
115 emphasizing the panel's focus on the functioning of the monitoring system as a whole. Three areas that are especially relevant to the evaluation criteria were also discussed. Clear objectives are crucial in providing direction for monitoring design and implementation. An effective technical design then translates these objectives into decisions about what to monitor; how, when, and where to take measurements; and how to analyze and interpret the data. Finally, an overall assessment of environmental problems in the bight provides a framework for determining if all important questions are being addressed and whether monitoring resources are being allocated effectively.
