4
The Use of Science in Decision Making

A key tenet of the Everglades restoration effort is that reliable scientific information will guide critical ecosystem management decisions. This principle is written as background for the Programmatic Regulations, the legal document that guides the implementation of the Comprehensive Everglades Restoration Plan (CERP): “The definition of restoration recognizes implicitly that science will be the foundation of restoration, but it also assumes … that in all phases of implementation of the Plan both restoration and the other goals and purposes of the Plan should be achieved” (33 CFR §385). Given the enormous scope and complexity of the restoration effort and the extensive research conducted in the Everglades, both effective science coordination and synthesis of scientific results are essential for science to be the foundation of the restoration.

Science and research have a long and rich history in South Florida, beginning in the mid-1800s with land surveys and collection of information on pre-drainage wildlife and vegetation conditions. Volumes of scientific studies were available by the late 1980s to support efforts to restore the Everglades. The first Everglades Research Conference, held in 1989, documented the history of the Everglades, its condition, and restoration alternatives. The results of this conference were published by Davis and Ogden (1994) and, along with a 1994 report identifying key scientific uncertainties (SSG, 1994), provided the scientific framework for the current plan to restore the Everglades.

This chapter describes the way that scientific information helps achieve the goals of the CERP. It emphasizes the importance of effective science coordination, synthesis of monitoring data, the development of useful models, and application of adaptive management to support restoration. The chapter reviews three major program documents to fulfill item 4 of the committee’s charge to provide “independent review of monitoring and assessment protocols to be used for evaluation of CERP progress (e.g., CERP



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 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 4 The Use of Science in Decision Making A key tenet of the Everglades restoration effort is that reliable scientific information will guide critical ecosystem management decisions. This principle is written as background for the Programmatic Regulations, the legal document that guides the implementation of the Comprehensive Everglades Restoration Plan (CERP): “The definition of restoration recognizes implicitly that science will be the foundation of restoration, but it also assumes … that in all phases of implementation of the Plan both restoration and the other goals and purposes of the Plan should be achieved” (33 CFR §385). Given the enormous scope and complexity of the restoration effort and the extensive research conducted in the Everglades, both effective science coordination and synthesis of scientific results are essential for science to be the foundation of the restoration. Science and research have a long and rich history in South Florida, beginning in the mid-1800s with land surveys and collection of information on pre-drainage wildlife and vegetation conditions. Volumes of scientific studies were available by the late 1980s to support efforts to restore the Everglades. The first Everglades Research Conference, held in 1989, documented the history of the Everglades, its condition, and restoration alternatives. The results of this conference were published by Davis and Ogden (1994) and, along with a 1994 report identifying key scientific uncertainties (SSG, 1994), provided the scientific framework for the current plan to restore the Everglades. This chapter describes the way that scientific information helps achieve the goals of the CERP. It emphasizes the importance of effective science coordination, synthesis of monitoring data, the development of useful models, and application of adaptive management to support restoration. The chapter reviews three major program documents to fulfill item 4 of the committee’s charge to provide “independent review of monitoring and assessment protocols to be used for evaluation of CERP progress (e.g., CERP

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 TABLE 4-1 Summary of RECOVER Monitoring and Assessment Plan Products Document Acronym Status/Date Overview Monitoring and Supporting Research (RECOVER, 2004) MAP I Final/January 2004 currently being implemented. See Appendix C for the status of monitoring components as of January 2006. Describes monitoring plan and research justifying plan. Draft performance measures are identified. Focus is the natural system. Implementation plan is included. 2005 Assessment Strategy for the Monitoring and Assessment Plan (RECOVER, 2005a) MAP II Final draft/September 2005 Provides a framework for analyzing relevant monitoring data and assessing progress toward the CERP goals and objectives. See also Box 4-1. Comprehensive Everglades Restoration Plan System-wide Performance Measures (RECOVER, 2006b) PM report Revised review draft/March 2006 Justifies selection of each performance measure. The scope, development, application, and associated uncertainty of each performance measure are discussed. Quality Assurance Systems Requirements (RECOVER, 2006c) QASR Peer-review draft/June 2006 Provides quality assurance protocols for all performance measures. Also includes information on data validation, management, and data archiving. performance measures, annual assessment reports, assessment strategies, etc.).” The reviewed documents include the CERP Monitoring and Assessment Plan: Part 1 Monitoring and Supporting Research (MAP I; RECOVER, 2004), 2005 Assessment Strategy for the Monitoring and Assessment Plan (MAP II; RECOVER, 2005a) (see Table 4-1), and the Comprehensive Everglades Restoration Plan Adaptive Management Strategy (RECOVER, 2005c; superseded by RECOVER, 2006a). Accomplishments of the science program and issues that will require further efforts are highlighted throughout the chapter. The chapter begins by assessing the monitoring and assessment programs developed by the Restoration Coordination and Verification (RECOVER) program, including the progress and challenges faced in the implementation of the programs. The chapter then describes the importance of science coordination and synthesis to support the restoration effort and

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 examines how science should and does feed back into decision making within an adaptive management framework. The chapter concludes with a discussion of progress in modeling and its importance as a core science component supporting CERP planning and implementation. These topics are all elements of the adaptive management approach expected by Congress. The Senate Committee on Environment and Public Works (Senate Report No. 106-362) wrote: “the Committee expects that the agencies responsible for project implementation report formulation and Plan implementation will seek continuous improvement of the Plan based on new information, improved modeling, new technology and changed circumstances.” The success of the CERP depends on strategic, high-quality, responsive, and sustained science. THE MONITORING AND ASSESSMENT PLAN The Programmatic Regulations for the CERP (33 CFR §385) recognize the central role of monitoring and assessment to provide a scientific basis for restoration planning and implementation by mandating the development of a Monitoring and Assessment Plan (MAP). The MAP provides the framework that the RECOVER teams will use to measure and understand the ecosystem’s responses to the CERP and to help determine how well the CERP is meeting its goals and objectives. Specifically, the MAP will “(i.) establish a pre-CERP reference state including variability for each of the performance measures, (ii.) provide the assessment of the system-wide responses of the CERP implementation, (iii.) detect unexpected responses of the ecosystem to changes in the stressors resulting from CERP activities, and (iv.) support scientific investigations designed to increase ecosystem understanding, establish cause-and-effect relationships, and interpret unanticipated results” (RECOVER, 2004). The information generated from the MAP also can support CERP project planning, design, implementation, and operation and provide information needed to make informed decisions about the need to alter restoration plans through the adaptive management process. The monitoring plan is based on conceptual models of 11 physiographic regions (Figure 4-1) and of the entire South Florida ecosystem (i.e., Total System Conceptual Model). These conceptual models are an assembly of well-informed hypotheses that describe the relationship between societal actions, environmental stressors, and ecosystem characteristics and the linkages among the physical, chemical, and biological elements within the natural system (Figure 4-2). In all cases, the CERP conceptual ecological

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 FIGURE 4-1 Boundaries of the 11 conceptual ecological models. SOURCE: RECOVER (2004).

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 models are based on extensive research and the best professional judgment of scientists working in the South Florida ecosystem. The conceptual ecological models link physical stressors with ecological effects and identify the ecosystem attributes most likely to respond to CERP projects and their operations (Ogden et al., 2005b). The conceptual ecological models also provide a planning tool for translating the CERP goals into specific performance measures (or ecosystem indicators) that will be used to assess the success of the CERP. Although the relationships used to construct the conceptual models are based on extensive observation and experimentation, uncertainties remain about how the ecosystem as a whole will respond to the CERP. Hypotheses about individual system-response relationships will be examined further through the adaptive management process as CERP pilot projects are completed and projects are designed, implemented, and ultimately operated. The conceptual models and their associated causal hypotheses have been subjected to independent scientific peer review and were recently published in a peer-reviewed journal (e.g., Davis et al., 2005a; Ogden et al., 2005a; Rudnick et al., 2005). Development, peer review, and publication of the monitoring plan conceptual models are major accomplishments of RECOVER. Components of the Monitoring and Assessment Plan Collectively, four documents constitute the CERP MAP (Table 4-1). The CERP Monitoring and Assessment Plan: Part 1 Monitoring and Supporting Research (MAP I; RECOVER, 2004) describes the monitoring components for measuring the system responses to CERP implementation that will inform the assessment process. MAP I presents early drafts of the conceptual ecological models as part of the rationale supporting the selection of the monitoring components and identifies draft performance measures (further developed in RECOVER, 2006b). Examples of performance measures include the number and duration of dry events in Shark River Slough, sulfate concentrations in surface waters of the Everglades ecosystem, mangrove forest production and soil accretion, and wading bird nesting patterns. MAP I also lays out a plan for collecting data on the variables required to assess the status of performance measures. Performance measures typically do not rely on a single variable (e.g., number of wood stork nests) but instead include measures of a variety of variables. MAP I lays out the variables to be monitored, spatial sampling networks, and monitoring frequency necessary to assess the status of each performance measure.

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 FIGURE 4-2 Example of 1 of the 11 conceptual ecological models that serve as the basis of the Monitoring and Assessment Plan. SOURCE: Adapted from Davis et al. (2005b).

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 BOX 4-1 Assessment Strategy for the Monitoring and Assessment Plan The 2005 Assessment Strategy for the Monitoring and Assessment Plan (MAP II; RECOVER, 2005a) provides general guidance for RECOVER assessment activities and a framework for analyzing the data generated from the CERP MAP and other relevant monitoring data to address the overall status of the South Florida ecosystem relative to CERP goals. The MAP technical assessment process (Figure 4-3) offers guidance at three levels: (1) the MAP component level, (2) the module level (i.e., Greater Everglades, Southern Estuaries, Northern Estuaries, Lake Okeechobee, South Florida Hydrology Monitoring Module, and South Florida Mercury Bioaccumulation Module), and (3) the system level (see Appendix C for a detailed list of MAP modules and their individual components). Guidance at the MAP component level is directed toward principal investigators and focuses on detecting change, establishing reference conditions, and measuring changes from those reference conditions. At the module level, guidance focuses on the integration of multiple performance measures in the evaluation of specific hypotheses. Module-level assessments are designed to determine the direction and magnitude of change in the integrated performance measures and to help evaluate whether those changes are consistent with the expected responses described in the CERP hypotheses. The module level helps evaluate progress toward interim goals and targets, identify unexpected or surprising results and episodic events, and integrate relevant project-level monitoring. Finally, the system-level guidance addresses possible decision alternatives resulting from the assessment of individual or multiple performance measures and MAP hypotheses within and across the modules. Figure 4-4 illustrates the three alternatives for interpreting assessments at the systems level. The findings that result from the various assessment activities at the MAP component, module, and system levels are presented in several annual reports that are eventually compiled and synthesized to produce the RECOVER Technical Report. The Technical Report assesses whether the goals and purposes of the CERP are being achieved and is released at least every 5 years—more frequently if deemed necessary. The annual assessment reports and the Technical Report fulfill reporting requirements to Congress, the U.S. Army Corps of Engineers and the South Florida Water Management District, and the public. The MAP II process is currently under way and the release of the first 5-year Technical Report is anticipated in 2010, although one could be released sooner under special circumstances (RECOVER, 2005a).

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 FIGURE 4-3 MAP technical assessment process. SOURCE: RECOVER (2005a).

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 FIGURE 4-4 Decision framework for interpreting systemwide assessments. SOURCE: RECOVER (2005a). The framework for analyzing the monitoring data collected as part of the MAP and assessing the effects of CERP project implementation is described in the 2005 Assessment Strategy for the Monitoring and Assessment Plan (MAP II; RECOVER, 2005a). MAP II also provides guidelines for assessing the progress toward achieving the restoration goals. Box 4-1 describes in more detail the technical assessment process outlined in MAP II. MAP I and II are discussed in more detail below. The Comprehensive Everglades Restoration Plan System-wide Performance Measures report (PM report; RECOVER, 2006b) describes in detail the development of each performance measure, how it relates to the conceptual ecological models, and how it can be used to evaluate the predicted performance of the CERP or to assess the status of the ecosystem before and after CERP project implementation. The PM report also describes restoration

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 goals or targets for each performance measure. As mentioned in Chapter 1, the PM report was not completed in time to be considered in detail in this report. Nevertheless, the committee utilized information in the most recent PM report draft to support its findings. The Quality Assurance Systems Requirement Manual for the Comprehensive Everglades Restoration Plan (QASR; RECOVER, 2006c) is a rigorous, detailed compilation of quality assurance protocols. QASR is a manual that includes the quality assurance and quality control requirements for all CERP monitoring data regardless of the data format (e.g., field observations, laboratory analyses, imagery, model output). Such a plan has enormous value. As noted by the National Research Council (NRC, 1996), “Currently, a great deal of monitoring data is collected in the United States. However, the data are incomplete … of varied quality, and non-standardized in collection protocol.” These shortcomings are potential risks to CERP success because of the massive amount of data needed to support the CERP, the varied format of the data, the number of institutions and individuals involved in data collection, and the life expectancy of the restoration. A peerreview draft of the QASR manual (RECOVER, 2006c) was released in June 2006 and the manual is nearing completion, but it will be periodically reviewed and updated as needed. The committee did not review the QASR manual in detail for this report. Completion of the full MAP is anticipated in the near future and will be a major accomplishment of RECOVER because all pieces of the MAP are essential to the assessment phase of the CERP. RECOVER has made good progress toward developing and implementing a statistically defensible monitoring plan and an ambitious assessment strategy. This conclusion is based on multiple briefings to this committee, evaluation of final drafts of MAP I and MAP II, and a review of the concerns expressed in a previous NRC review of adaptive monitoring and assessment for CERP, which focused in particular on MAP I (NRC, 2003b). Significant progress has been made since that review, and the authors of the MAP have at least partly addressed many of the previously expressed concerns about the program and its use in adaptive management (Box 4-2). The MAP documents emphasize that the monitoring and assessment plans will separate the effects of hydrologic management from other impacts (e.g., climate variation, land use). Yet, RECOVER recognizes that linking hydrologic changes to specific ecosystem responses is difficult because of the time lags involved in ecosystem responses, natural fluctuations in ecological variables that may obscure trends, and difficulties in separating system responses to hydrology from other drivers. RECOVER’s current

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 BOX 4-2 Conclusions from the NRC (2003c) Review of MAP and Actions Taken Although the MAP was grounded on the practice of adaptive management, the least developed aspects of planned adaptive management were feedback mechanisms to connect monitoring to planning and management. Action taken: Comprehensive Everglades Restoration Plan Adaptive Management Strategy (RECOVER, 2006a) developed to articulate roles and relationships. Restoration goals, objectives, and targets were inadequately defined and not reconciled with the large-scale forces of change in South Florida. Action taken: Interim Goals-Interim Targets developed. Primary reliance on passive adaptive management limited the ability to make inferences regarding cause and effect and to distinguish policy effects from other human forces or natural processes. Action taken: None specifically; this remains an impediment. The MAP needed a rigorous quality assurance/quality control program to ensure that monitoring data are of high quality and utility. Action taken: The QASR manual is nearly completed. Including combinations of ecological performance measures and environmental variables hypothesized to impact those measures is critical for the adaptive management approach. Action taken: Performance measures revised to reflect ecological response to environmental variables. Assessment strategy recognizes this need. Region-wide monitoring of ecosystem drivers is essential to reducing uncertainties associated with the restoration plan, but had received comparatively little attention. Action taken: Developed Total System Conceptual Ecological Model (Ogden et al., 2005a), but performance measures for the total system were not identified. Scientists developing the MAP needed an explicit understanding of the information management needs and how monitoring results will be used. Action taken: Information management specialists are being recruited. Monitoring must also serve compliance monitoring and report card functions in addition to adaptive management. Action taken: No action taken. Strategies for reducing the number of performance measures and prioritizing monitoring needs were needed. Action taken: Number of total performance measures reduced to 73 in MAP I (RECOVER, 2004) from 156 (RECOVER, 2001); however, the PM report (RECOVER, 2006b) lists 83 total performance measures. The number of total performance measures (73-83) still remains a potential problem. Key uncertainties have been identified and are documented in MAP I (RECOVER, 2004) and the PM report (RECOVER, 2006b). plan is to try to distinguish natural system variation and responses to nonhydrologic drivers from hydrologic changes resulting from the CERP by using modeling tools to estimate ecosystem response in the absence of the CERP. The need for early recognition of surprising ecosystem responses

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 Scale (Spatial Extent; Resolution) Status Developers/Sources Regional; 500 × 500 m More models are being developed and calibrated http://www.atlss.org/ Local, at the scale of STA Plans to improve user interface, better represent hydraulic features W. Walker and R. Kadlec http://wwwalker.net/ Regional; 1 × 1 km Version 2.5 (in review) SFWMD http://www.sfwmd.gov/elm/ Subregional, can be used for essentially any spatial resolution Version 2005 Danish Hydraulic Institute, http://www.mikeshe.com/mikeshe/index.htm Regional; 2 × 2 mile Version 4.6.2 SFWMD http://www.sfwmd.gov/org/pld/hsm/models/nsm

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 Model Name Full Name and Main Function Example Applications REMER Regional Engineering Model for Ecosystem Restoration encompasses most of the South Florida area and includes diverse hydrologic attributes and processes such as canal control structures, surface-subsurface interaction, pumping wells, retention ponds, lakes, levees, culverts, roads, bridges, one-dimensional canal flow, two-dimensional overland flow, three-dimensional subsurface flow, etc. Model is not yet complete SFRSM South Florida Regional Simulation Model is a finite-volume-based model capable of simulating multidimensional and fully integrated groundwater and surface-water flow. Regional long-term (decades) simulations of complex hydrology with management (e.g., southwest Florida) SFWMM South Florida Water Management Model simulates hydrology and water systems and is widely accepted as the best available tool for analyzing structural and/or operational changes to the complex water management system in South Florida at the regional scale. Regional Modeling for the Everglades Agriculture Area Storage Reservoir SICS Southern Inland and Coastal Systems numerical model simulates hydrologic conditions for the transition zone between the wetlands of Taylor Slough and C-111 canal and nearshore embayments of Florida Bay. It is a useful tool for understanding the effects of coastal hydrology and for defining boundary conditions of other models. Linking with SFWMM and the Florida Bay hydrodynamic model to project coastal flows to Florida Bay and coastal wetland salinities under restoration conditions in the future WAMVIEW WAMVIEW is an ESRI ArcView version of an earlier model of WAM (Watershed Assessment Model) for simulating water quality as well as physical and chemical processes. Used to assist SFWMD to develop pollutant load reduction goals NOTE: The list is not intended to be comprehensive. Numerous other models describe water circulation, water quality, and aspects of system ecology, especially in the estuaries and Lake Okeechobee.

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 Scale (Spatial Extent; Resolution) Status Developers/Sources Regional, adaptable to subregional and local Activities related to REMER terminated in 2006 USACE (Cheng et al., 2005) Regional; 0.1-2 mile triangular elements Calibration and verification under way as of December 2005 SFWMD http://gwmftp.jacobs.com/Peer_Review/web_page/peer_review_sfwmd.htm Regional; 2 × 2 mile Version 5.5 SFWMD http://www.sfwmd.gov/org/pld/hsm/models/sfwmm/ Local/Subregional; 500 × 500 m   Swain et al., 2004 Regional; 0.1 ha Version 1.1 http://www.epa.gov/ATHENS/wwqtsc/html/wamview.htm; http://www.stormwaterauthority.org/assets/073PLWAMModel.pdf

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 tance to these features of extreme events, such as hurricanes, versus average annual flow characteristics), the development of regional-scale, predictive flow models to support adaptive management will be essential. All existing models have the potential to serve the adaptive management process because they predict different aspects of hydrology at various scales. However, most of the models were developed for purposes other than the CERP and, therefore, lack the linkages to other models necessary to support adaptive management during the restoration. The inability to link various models inhibits integrating knowledge about different features of hydrology and extrapolating new information from one scale to another. The most important limitation is that the resolution (2 × 2 mile grid cells) of the SFWMM and NSM is too coarse to be useful in projecting the ecological effects of the hydrology they depict. The limited ability to link ecological models to hydrologic models at the high-resolutions relevant to ecological concerns has been a critical deficiency in planning efforts and will limit the effectiveness of the adaptive management process if not remedied. For limited areas, a high-resolution multi-data source topography (HMDT) has been created using light detection and ranging (LiDAR) and USGS high-accuracy elevation data (HAED) where available (USGS, 2004; see Box 4-5). In most of the area where such LiDAR and HAED data are not available, however, estimates of elevations within each SFWMM 2 × 2 mile cell are developed at a resolution of 500 × 500 m.5 Although HMDT has been used for Across Trophic Level System Simulation (ATLSS, Table 4-2) to reduce the inaccuracies associated with the 2 × 2 mile resolution of the SFWMM (Duke-Sylvester et al., 2004), additional high-resolution models are essential to link hydrology with ecology. An important recent development is the effort to address this problem by constructing new models with higher spatial resolutions. One such model under development is the South Florida Regional Simulation Model (SFRSM), being developed by the SFWMD. The SFWMD also is developing a Natural System Regional Simulation Model (NSRSM) as an alternative to the NSM. The primary goal of the NSRSM is the same as that of the NSM, but it is based on the same governing equations, object-oriented design, and numerical methods of the SFRSM, which can simulate multidimensional groundwater and surface-water flow. Additionally, the spatial extent of the NSRSM is larger than that of the NSM, and a number of datasets used to set initial model conditions have also been improved (e.g., land cover, topogra- 5 http://www.atlss.org/~sylv/HTML/Everglades/HMDT-ShortReport/main.html.

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 phy, predevelopment river network). Most importantly, like the SFRSM, the NSRSM employs a finer spatial resolution than the NSM in some regions, with grid sizes ranging from 0.1 to 2 miles on a side, making the model useful at scales more relevant to many ecological processes.6 The development of subregional and local hydrologic models to provide a linkage between the large-scale, regional hydrologic models and small-scale, local ecological models has increased predictive ability for some critical ecosystem attributes. For example, modelers specifically responded to the need to link ecological and hydrologic models and to better model the southern portion of the Everglades ecosystem (NRC, 2002b) by developing the Southern Inland and Coastal Systems (SICS) and the Tides and Inflows in the Mangrove Ecotone (TIME) subregional models. SICS has been linked with parts of the ATLSS models (e.g., ALFISHES) to provide hydrologic information for determining fish population dynamics (Langevin et al., 2004); TIME has a spatial scale (500 m × 500 m, or 0.31 mi × 0.31 mi) conducive to linkage with other ecological models. The ability to link regional hydrology models to subregional and local hydrology models and to ecological models is essential to the CERP adaptive management strategy. Without such linkages it will be difficult to provide the information required to make management decisions based on observations of ecological performance measures. The CERP remains deficient in this regard, and more efforts to improve the linkages are needed. Ecological Models Ecological models are essential tools for assessing ecological effects of the CERP and for adaptive management. Possible ecological effects, including changes in primary productivity of ecosystems and changes in population dynamics of plants and animals, are among the most important criteria for evaluating the performance of the CERP. However, ecological models are less “mature” (DOI, 2005) than hydrologic and water quality models. The conceptual ecological models (see Figures 4-1 and 4-2) describe the current understanding of the critical relationships that affect ecosystem functioning in South Florida (Ogden et al., 2005b). The models have considerable heuristic value and, most importantly for purposes of the current discussion, the intended role of the models in adaptive management is 6 http://my.sfwmd.gov/pls/portal/url/page/PG_GRP_SFWMD_HESM/PG_SFWMD_HESM_RSM?navpage=rem.

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 BOX 4-5 Advances in High-Resolution Topography The topography of South Florida provides a subtle but complex platform for the Everglades ecosystem. The land surface in the entire watershed has no more than about 65 feet of vertical relief, and water falls only 20 feet in the 100-mile reach from Lake Okeechobee to the ocean at Florida Bay. Despite this shallow gradient, the Everglades watershed has substantial microtopographic landscape complexity. Variations in terrain height as small as 6 inches create a variety of ecological conditions related to the frequency, timing, and duration of inundation. For example, the ridge-and-slough landscape, one of the major habitat types in the Everglades ecosystem, consists of parallel and usually submerged ridges and sloughs with height differentials of less than 3 feet (Figure 4-7). An understanding of the subtle variations in terrain is critical to explaining the hydrol FIGURE 4-7 Ridge-and-slough landscape of Water Conservation Area 3A (2005). SOURCE: Photo courtesy of Christopher McVoy, SFWMD, 2006.

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 ogy and ecology of the region and to predicting the potential outcomes of management decisions using numerical models. Digital information about Everglades terrain initially came from topographic contour maps that were developed through surface surveys. When it became clear that the restoration of the Everglades through the CERP would require highly detailed topographic data, investigators sought a new approach to defining the terrain that would provide highly accurate, high-resolution topographic data for the emerging hydrology and ecology models in digital form. Everglades scientists have worked to develop high-resolution topographic data to a standard of plus or minus 6 inches, first through an unsuccessful experiment with LiDAR in the late 1990s and subsequently with an Airborne Height Finder (AHF) system. Despite these efforts to improve the resolution of Everglades topography, hydrologic models remain limited by topographic data, and they are unlikely to be able to produce improved predictions unless they use terrain data with greater resolution. Inaccuracies in predictions are directly related to inaccurate elevation data with a resolution that is too coarse to characterize significant topographic controls on hydrology. Further application of the USGS’s AHF system is unlikely to improve the present situation because the resolution required by hydrologic and ecologic modelers is finer than the AHF system produces. A resolution at the scale of a meter or so is needed in some crucial areas where the microtopography is complex, where habitat houses endangered species, and where switching points occur for the flow of surface water. The only current method for creating a digital elevation model (DEM) with plus or minus 6-inch-height accuracy, broad regional coverage that is the same throughout the watershed, and at a horizontal resolution down to about 3 feet in crucial areas is the use of new versions of LiDAR (ASPRS, 2001; Baltsavias, 1999; Fowler, 2001). This technology, and the contractors who use it, have undergone substantial improvements in the past few years. Modern postmission processing of the data can differentiate between the tops of vegetation and the ground surface. Instrumentation and techniques have also improved dramatically. Where the older instruments issued 10,000 light pulses per second, the newer instruments emit up to 50,000 pulses per second, and the large number of emissions also means that it is possible to penetrate relatively dense vegetation such as that found in the Everglades. Ongoing experiences near Hilton Head, South Carolina, show that LiDAR is effective in an environment similar to the Everglades (Greene, 2004; Tullis, 2003). The USACE has also successfully connected topography of the surface with bathymetry from underwater surfaces in coastal waters using LiDAR in their Scanning Hydrographic Operational Airborne LiDAR Survey (SHOALS), so similar success should be expected in the Everglades case (Irish and Lillycrop, 1999). Finally, LiDAR has become increasingly inexpensive, especially when it is combined with other airborne approaches and over large areas where unit costs are low. It is not clear whether LiDAR can produce a region-wide, inexpensive, high-resolution grid for the CERP, but it is clear that this technology is superior to others. It is almost certain that the newer LiDAR systems can meet the plus or minus 6 inches vertical accuracy requirements of the CERP. A horizontal grid of 1 m is easily within the capability of the system, and its potential for wide coverage indicates that it is possible to create a regional hydrologic model with the exceptional resolution needed to include microtopography.

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 explicitly recognized and defined. Indeed, the conceptual models provide the foundation for the monitoring plan and were used to generate the set of causal hypotheses about how the natural systems in South Florida have been altered by water management (Gentile et al., 2001). However, quantitative ecological modeling for the CERP is limited. The emphasis of the current quantitative ecological modeling effort is, appropriately, on using simple models (HSI and Spatially Explicit Species Indices) to evaluate restoration scenarios and identify information needs, until sufficient data exist to calibrate and test more complex individual-based models. Much of the quantitative ecological modeling supporting the restoration is encompassed by the ATLSS program (see Table 4-2), which contains a suite of models for many species. These models simulate ecological patterns and processes in response to different restoration scenarios, such as water management. Model outputs include dynamics and spatial distribution of vegetation, primary productivity, nutrient cycling (e.g., total phosphorus levels), and habitat suitability and population dynamics of various animal species such as white-tailed deer, snail kite, Cape Sable seaside sparrow, and wading birds. Many of the model outputs have been linked to performance measures of hydrologic scenarios of the CERP. Another, much-anticipated, ecological model is the Everglades Landscape Model (ELM). ELM is designed to simulate landscape-level responses to management activities (Table 4-2). The latest version of ELM is under review (RECOVER, 2006b). Future Modeling Needs Although the overall modeling effort is extensive and continues to improve, several areas require special attention in future modeling efforts so that CERP projects can be designed and managed adaptively to enhance the potential for restoration success. First, in addition to the limited linkage between the primary hydrologic models and ecological models and the relatively slow development of quantitative ecological models, the lack of linkage among water quality and ecological models is a particularly important problem at the subregional level (e.g., Lake Okeechobee water quality model) as well as at the regional level. Furthermore, ecological modeling capability should be enhanced to better support restoration decision making and to improve linkages with other types of models for adaptive management. Existing models also need to be linked with socioeconomic models, such as demographic models, urban growth models, and land-use models (Liu, 2001; Walker, 2001; Walker and Solecki, 2004), because socioeconomic activities will greatly

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 shape the future of all aspects of the Everglades ecosystem and, thus, the fate of CERP projects. Second, the design of the entire modeling effort needs to be directed toward its role in adaptive management, as has already been done with the conceptual ecological models. The scales at which models are built need to match the scales at which decisions are made, management takes place, and ecosystems respond to management. A wider range of sensitivity analyses and uncertainty analyses is needed to further evaluate the performances of existing models (DOI, 2004) and to explain the sources and consequences of uncertainty. More empirical data from experiments that involve ecosystem manipulations, such as the Loxahatchee Impoundment Landscape Assessment, are needed to inform models. Third, efforts to link and focus models to fit the needs of adaptive management require a coordinated, multidisciplinary approach. While investment in a large and varied modeling effort is necessary and appropriate, coordination is necessary to avoid duplication of effort. It is important that the role of each model in the adaptive management process be well defined in terms of the processes it addresses, how it is to be modified based on feedback from monitoring, and the way it is to be used to inform decision making. In 2003, the Interagency Modeling Center was established to provide a centralized pool of resources and modeling expertise (DOI, 2004). This collaborative effort among federal and state agencies aims to improve modeling efficiency and model consistency. Pooling modeling talents into one unit may facilitate coordination of the modeling activities, foster development of better linkages among models, and give rise to new models to meet the needs for monitoring data integration and assessment and testing the conceptual understanding of ecosystem patterns and processes. A drawback to the Interagency Modeling Center is that it has the potential to isolate modelers from the scientists collecting the monitoring data if interactions among the groups are not close enough. Coordination of modeling and monitoring should be of high priority because of their intimate relationship in the adaptive management process. Fourth, models will need to incorporate data at more precise spatial or temporal scales that are compatible with model structure and that address ecological needs (see an example in Box 4-5). CONCLUSIONS AND RECOMMENDATIONS The committee reviewed three major program documents that collectively provide a foundation for ensuring that scientific information needed

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 to support restoration planning will be available in a timely way. The committee also examined the extensive set of models that have been developed to support restoration planning and adaptive management. The MAP documents reviewed describe a well-designed, statistically defensible monitoring program and an ambitious assessment strategy. The plan provides for a continuous cycle of monitoring and experimentation, as well as regular and frequent assessment of the findings. In combination, the MAP provides an approach to reduce uncertainty associated with the conceptual ecological models that are the foundation of the monitoring plan and to create new knowledge for understanding old and emerging problems. The MAP should also lead to better simulations of the ecosystem and help identify information gaps that currently impede adaptive management. Implementation of the monitoring plan is occurring more slowly than planned, and two key elements of the MAP are still incomplete. The effectiveness of the MAP as a component of the adaptive management strategy can be determined only by implementation. Each of the components of the MAP needs to be in place and tested to enable integration of scientific information into the decision-making process. A spatially and temporally robust baseline of monitoring data tied to performance measures is essential for a rigorous assessment of the progress of restoration of the natural system. A well-planned, transparent information management system is required to facilitate effective data assessment and information sharing. Additional key staff and staff-support positions devoted to information management and implementation of the monitoring activities are needed to facilitate more rapid implementation of the MAP. Continuing to winnow the number of performance measures to an even smaller subset that includes a limited number of whole-system performance measures would help ensure that the MAP is sustainable over the lifetime of the CERP. Organizational mechanisms for coordinating and synthesizing science related to both CERP and non-CERP projects are essential to ensure that research can support informed project planning and decision making. Science coordination is occurring at multiple levels within individual organizations with a focus on individual agency missions, but it remains unclear whether science is being effectively and collaboratively coordinated across the major programs that support the restoration. There is a critical need for synthesis of CERP and non-CERP science knowledge to help identify and reduce scientific uncertainties that impede restoration. The CERP Adaptive Management Strategy provides a sound organizational model for the execution of a passive adaptive management program.

OCR for page 86
Progress Toward Restoring the Everglades: The First Biennial Review – 2006 The strategy should be implemented soon to test and refine the approach. The CERP AM Strategy proposes a process for addressing uncertainty and supporting collaborative decision making. The objectives, mechanisms, and responsibilities are well specified in the adaptive management strategy, but the all-critical linkages among the planning, assessment, integration, and update activities require further development. The adaptive management strategy should be fully implemented soon to test these all-important linkages to refine the strategy accordingly. Incorporating active adaptive management practices whenever possible will reduce the likelihood of making management mistakes and reduce the overall cost of the restoration. Active adaptive management approaches that are specifically designed to address uncertainties, such as the Decomp Physical Model (see Chapter 5), offer greater opportunities for learning than an entirely passive approach. Regardless of which adaptive management approach is used, it remains to be seen how willing decision makers will be to make significant alterations to project design and sequencing, as opposed to limiting adaptive management to making modest adjustments in the operation of the CERP projects after their construction. A coordinated, multidisciplinary approach is required to improve modeling tools and focus modeling efforts toward direct support of the CERP adaptive management process. Models are used to forecast the short- and long-term responses of the South Florida ecosystem to CERP projects and, thus, are the critical starting point for adaptive management. An impressive variety of models has been developed to support the CERP, but better linkages between models, especially between hydrologic and ecological models, are needed to better integrate scientific knowledge and to extrapolate new information to the spatial scales at which decisions are made. In addition, hydrologic models suffer from the lack of high-resolution input data describing the basic terrain, so that their predictions are sometimes in error, and their connections to other more high-resolution ecosystem models is difficult. The development of quantitative ecological models is lagging behind the development of hydrologic models, hindering the model linkages necessary to support the restoration efforts. Because models themselves must be improved through comparison with actual outcomes, coordination between modeling and monitoring efforts, within the adaptive management framework of iterative improvement, should be a high priority.