5

Looking Forward

Many of the emerging challenges to implementation due to newly available knowledge and information are described in Chapter 4. Advancements in scientific and engineering knowledge related to the understanding of pre-drainage hydrology, climate change and sea level rise, and the feasibility of storage alternatives each are likely to have significant, systemwide impacts on the outcomes of restoration efforts. Climate change, now nationally and internationally recognized as crucial in ecosystem restoration planning, was less of a prominent issue during the development of the Central Everglades Restoration Plan (CERP; USACE and SFWMD, 1999, known as the Yellow Book). In this chapter, the committee revisits how the CERP was originally framed to accept such challenges during implementation and considers what needs to be accomplished, in parallel with ongoing project implementation, in order to set a sound course for long-term implementation and system-level ecosystem restoration.

Accordingly, in this chapter the committee describes the importance of a programmatic adaptive management approach to ensure systemwide goals are achieved. The goals and objectives at an ecosystem-scale, which have not been modified since 1999, need to be revised in light of new information that is available. A new systemwide analysis of the future of the Everglades ecosystem based on changes anticipated or likely in the CERP is needed to inform programmatic adaptive management and regional and local planning efforts, and to provide insight into the establishment of revised long-term goals and objectives. Many new tools/models are available to support this work, and they need to be used much more fully than they have been in project-level planning as well as part of the systemwide analysis.

CERP VISION FOR ADAPTING TO NEW INFORMATION

When the U.S. Congress approved the CERP in the Water Resources Development Act of 2000 (WRDA 2000), there was clear recognition that the Yellow Book (USACE and SFWMD, 1999) provided only a general outline for restoration of the Everglades and not a detailed restoration plan. To facilitate restoration actions in the face of uncertainty, an adaptive management approach (Holling, 1978) was embraced as a mechanism to incorporate emerging scientific and engineering information into the plan and to address unforeseen issues related to the restoration project. Congress approved funding for an adaptive management and monitoring program in WRDA 2000, and the Programmatic Regulations (33 CFR §385.31) directed the U.S. Army Corps of Engineers (USACE) to adopt an adaptive management approach.

The Programmatic Regulations define adaptive management as “a means for analyzing the performance of the Plan and assessing progress toward meeting the goals and purposes of the Plan as well as a basis for improving the performance of the Plan.” Specifically, the Regulations (33 CFR §385.31) require the CERP adaptive management program:

Since the launch of the CERP, substantial progress has been made in developing principles and frameworks for CERP adaptive management, along with detailed guidance for implementing adaptive management at the project level (RECOVER, 2006a, 2011b; USACE and SFWMD, 2011b). Recent attention has turned toward adaptive management at the program level (RECOVER, 2015). Nine major activities identified for CERP adaptive management at the project or program level are summarized in Table 5-1. Two major activities—establishing goals and assessment—are discussed in more detail below because of their important roles in program-level adaptive management.

TABLE 5-1 CERP Adaptive Management Activities and Associated Program or Project Activities

^a This activity is largely for the project level.

SOURCE: USACE and SFWMD (2011b).

Establishing Goals and Objectives

What is the CERP trying to achieve for the ecosystem? As discussed in Chapters 2 and 4, at a high level, this is relatively easy to articulate and generally agreed upon. The stated goal of the CERP is “restoration, preservation, and protection of the South Florida ecosystem while providing for other water-related needs of the region, including water supply and flood protection” (WRDA 2000). The Programmatic Regulations (33 CFR § 385.3) that guide implementation of the CERP further clarify this goal by defining restoration as “the recovery and protection of the South Florida ecosystem so that it once again achieves and sustains the essential hydrological and biological characteristics that defined the undisturbed South Florida ecosystem.” The CERP goal is frequently also stated as “get the water right,” although getting the water right is a means to ecological ends, not an end itself (see also Chapter 2).

To establish objectives that can be used to guide project- and program-level restoration investments, these broad goals need to be interpreted in the context of the complex Everglades ecosystem, which is naturally highly variable in space and time and has been substantially altered over the past century. At the time of authorization, the CERP laid out some ambitious albeit generalized expectations for the ecosystem. For example, the Yellow Book (USACE and SFWMD, 1999) stated: “At all levels in the aquatic food chains, the numbers of such animals as crayfish, minnows, sunfish, frogs, alligators, herons, ibis, and otters, will markedly increase.” The expectations for CERP ecosystem outcomes were founded on hydrologic outputs from the NSM and the River of Grass Evaluation Methodology (ROGEM),¹ although the Yellow Book notes that the NSM and ROGEM are useful tools for comparing alternatives but not for predicting specific responses.

Objectives provide a “means by which the restoration success of the Plan may be evaluated at specific points by agency managers, the State, and Congress throughout the overall planning and implementation process” (33 CFR § 385.3). Development of measurable objectives is crucial to effective planning, implementation, and assessment at both the project and programmatic levels and requires consideration of the inherent tradeoffs that must be made in any complex ecosystem restoration program (Reed, 2006).

Progress on articulating objectives that are sufficiently quantitative to enable such an evaluation has been limited under the CERP. In 2005, RECOVER produced a set of recommendations for quantitative “interim goals” by subregion using available models and data, assuming that the original CERP schedule as documented in USACE and SFWMD (1999) would be realized. These interim goals were determined for 2010, 2015, and 2050 with full CERP implementation, compared to 1995 base conditions. Whether these quantitative goals were appropriate and well supported by validated models can be questioned, but the objective quantification of expected outcomes in this way showed an ability to consider the future state of the system and the influence of restoration actions on that state. Examples of the expected outcomes for the Everglades with full CERP implementation (2050) include the following:

• Increase by 103,709 acres the total spatial extent of natural areas;
• 1,871,000 acre-feet (AF) of “new” water captured by the CERP;
• Increase by 31 percent fish abundance in Northeast Shark River Slough, and
• 80,000 nesting pairs of wading birds (see Figure 5-1).

___________________

¹ ROGEM was developed and used during the development of the CERP to quantitatively describe potential habitat quality responses to alternative plans. Most of the variables in the tool are driven by hydrology and equations were developed for each of the subregions based on linkages between hydrologic conditions and habitat quality in natural areas.

However, these quantitative interim goals were not adopted. Instead, the goals listed in the 2007 Interim Goals agreement (USACE et al., 2007) are largely qualitative, indicating a desired direction of change in each indicator, without a quantitative objective. For example, for the Everglades indicator of “systemwide spatial extent of natural habitat,” the goal is simply stated as “increase spatial extent of natural habitat.” The signatories noted that the assumptions made regarding implementation in the RECOVER (2005b) report are substantial and instead reiterated the “high priority [for] continued development and refinement of the recommended indicators and interim goals contained within the RECOVER Recommendations consistent with the requirements of the programmatic regulations” (USACE et al., 2007).

To the committee’s knowledge, no further work on quantitative goals to guide restoration or to evaluate past or future success has been conducted, and no adjusted assumptions regarding implementation have been incorporated into such modeling. In 2010, an effort was launched to summarize new science since the advent of the CERP and use this information to develop a “shared definition of restoration” that could be used to develop quantitative and measurable

restoration goals (Working Group, 2010). The first phase of that process resulted in the Scientific Knowledge Gained report released in August 2011 (RECOVER, 2011a). Planned subsequent phases in which this synthesis of new information was to be used to revisit and update restoration goals and targets at a systemwide scale were never initiated, although this new information has been used to inform project planning.

As noted in Chapter 4, substantial new information on pre-drainage hydrology, climate change, and sea level rise has been obtained since restoration goals were developed. The implications of this new knowledge for restoration goals and objectives are varied. Climate change scenarios suggest that under both reduced and increased precipitation scenarios, more storage than originally envisioned will be required to meet performance targets. Sea level rise projections indicate the original hydrologic and ecological objectives for some areas cannot be achieved. New knowledge of the pre-drainage system indicates that restoring pre-drainage hydrology will not result in the ecological outcomes originally envisioned for a particular area, as exemplified for marl prairie habitat in Chapter 4.

Even if the broad goals of the restoration remain unchanged, the details of specific hydrologic and ecological objectives in space and time, especially quantitative objectives, need to be revisited. The Shared Definition initiative (Working Group, 2010) may have been an appropriate vehicle for this task, but it would need to be reinitiated on the basis of current knowledge. The capacity to identify achievable goals and objectives is much improved since CERP authorization due to advances in modeling, especially in the development of systemwide ecological models (discussed later in this chapter). The capacity now exists to identify an ecological goal and then determine the hydrology necessary to achieve it. The establishment of clearly defined goals and quantitative objectives will involve evaluation of tradeoffs among various hydrologic and ecological objectives, and perhaps some rethinking of priorities, especially with respect to expectations for particular species. Establishing quantitative restoration objectives may be especially challenging in situations that involve the restoration of pre-drainage features together with the preservation of post-drainage features that are deemed desirable (e.g., the littoral zone community in Lake Okeechobee; marl prairie habitat inhabited by Cape Sable seaside sparrows, see Chapter 4). The development of these objectives should involve reassessment of the essential characteristics of successful restoration, what is desirable but not essential, what can be achieved, and what cannot.

Whether such goals for the future system should be as specific as that presented in Figure 5-1 merits reconsideration when efforts to develop quantitative goals are reinitiated. It is important for adaptive management that goals are expressed in quantities that can be both predicted with confidence in the models

and measured in the field. When quantifying goals, RECOVER scientists and other experts need to consider both modeling skill and monitoring resources.

Assessment and Evaluation

The Programmatic Regulations define assessment as “the process whereby the actual performance of implemented projects is measured and interpreted based on analyses of information obtained from research, monitoring, modeling, or other relevant sources.” This step is critical to the adaptive management process, so that monitoring data can be compared to quantitative objectives to gauge restoration progress. As noted in Table 5-1, this process is also used to identify performance issues that may need to be addressed through project- or program-level modifications. However, the definition is perhaps too narrow for what is needed for the incorporation of new information into the CERP. Assessment could also reasonably include an evaluation of modeling results based on new scientific and engineering information to determine how well on-the-ground ecosystem achievements are aligned with CERP objectives and how these might change in the future. For instance, such an assessment could be used to evaluate how various levels of storage or climate change impact the ultimate attainment of CERP goals, thereby informing future planning.

The Programmatic Regulations call for RECOVER to prepare a report every 5 years that presents an assessment of whether the goals and purposes of the CERP are being achieved, including whether the interim goals are being achieved or are likely to be achieved. To date no such reports have been produced. There is an established system status report process (RECOVER, 2007a, 2010, 2012, 2014) but to date those reports have been largely based on assessment of trends in monitoring data and are not responsive to the need stated in the Programmatic Regulations to be forward looking and identify whether goals are likely to be achieved. For a long-term program like the CERP, continually looking forward to consider the state of the system and how it will change in the future is as important as looking back to document progress.

As part of the Adaptive Management Program, the CERP Programmatic Regulations also call for the agencies to “conduct an evaluation of the Plan using new or updated modeling that includes the latest scientific, technical, and planning information.” These evaluations, termed “CERP Updates,” are to be conducted “whenever necessary to ensure that the goals and objectives of the Plan are achieved but not any less often than every 5 years.” These evaluations can result in consideration of adjustments in operations, CERP components (removing, adding or changing), or any combination of these. The Initial CERP Update (RECOVER, 2005b) was conducted and published in 2005, which included revisions to performance measures, model updates, and changes in modeling

assumptions regarding existing and future conditions. No such evaluation of the CERP has been conducted since 2005 despite substantial changes in scientific, technical, and planning information, as described in Chapter 4.

In 2000, when the CERP was authorized with its strong emphasis on adaptive management, it was seen across the United States as a leader in approaches to incorporating new and developing information into the restoration plan (e.g., Kallis et al., 2009; Linkov et al., 2006). The frameworks and processes developed still provide a model for others, but what were the leading concepts of the time in ecosystem restoration have not been realized, particularly at the system scale. Tools are now available to support follow-up on key components, such as quantitative goals for the system and forward-looking assessment and evaluation. Given the new information and knowledge outlined in Chapter 4, it is even more important to look forward and anticipate issues, rather than only looking to the past and present to monitor restoration progress. Whether storage limitations really do compromise the overall success of the CERP is now a question that can be answered. Some critics say climate change will fundamentally limit the success of restoration (Holthaus, 2015), but the implications of climate change and sea level rise to CERP goals under various scenarios can be quantified, and other benefits not envisioned in the original restoration plan can be explored. Climate change is now widely recognized as critical to planning any new large scale coastal ecosystem restoration. By using new tools (discussed later in this chapter) and continuing to think about the future, even while in the midst of massive project implementation, the CERP can continue to lead in adaptive management and adaptation to changing circumstances and new science.

PROGRAM-LEVEL ADAPTIVE MANAGEMENT FOR CERP

The CERP Program-Level Adaptive Management Plan (RECOVER, 2015) was recently released and outlines a structured approach to implement systemwide adaptive management. Project-level adaptive management has been addressed in a previous report (RECOVER, 2011b). The RECOVER (2015) report is an important step in the evolution of the Monitoring and Assessment Plan (MAP) that has been developed over the past decade (RECOVER 2006b, c, 2007b, 2009b). The latest report addresses three program-level components missing from the MAP, specifically to

1. identify and prioritize (rank) programmatic uncertainties that might limit meeting CERP goals and identify strategies to address them,
2. develop an adaptive management approach to address the prioritized uncertainties, and

3. develop management option matrices containing options of how to improve restoration performance if goals are not being met.

Uncertainties were grouped into three tiers based on the level of knowledge, relevance to improving the design or operation of CERP projects, and the risks of not meeting CERP goals if the uncertainty is not addressed (see Box 5-1). Uncertainties designated with a combination of high risk, low knowledge, and high relevance were considered Priority 1 programmatic uncertainties and have been designated as “decision critical.” They are referred to as “showstoppers.” RECOVER (2015) states that if these are not addressed then a component of the restoration could be “paralyzed” and progress toward meeting CERP goals will be effectively stopped. A total of 13 Priority 1 programmatic uncertainties are identified that reflect uncertainties related to both planning and scientific questions (Box 5-1). The Priority 1 uncertainties identified by RECOVER (2015) include system-scale strategic issues as well as issues that are of concern in only some parts of the ecosystem. Some address large-scale sequencing and feasibility while others address more tactical issues related to project design. An additional 22 Priority 2 uncertainties of medium relevance and knowledge and medium or high risk are identified.

The Program-Level Adaptive Management Plan provides details on the identified uncertainties, including an explanation of how CERP progress would benefit from addressing each, and it identifies existing and potential strategies for dealing with the uncertainty. The report also specifies the information needed to complete the adaptive management feedback loop and presents management options matrices (MOMs), to summarize actions that could be taken if restoration efforts are not meeting performance targets. The MOMs describe the indicators used to assess performance, the thresholds for those indicators that define when corrective action should be taken, and options that could improve performance. Both the uncertainty tables and the management options matrices summarize a wealth of information and are intended as a quick-reference guide for managers.

A key element of the Program-Level Adaptive Management Plan is the summary of the project-specific goals, interim goals from RECOVER (2005b), “full restoration targets” that indicate good CERP performance relevant to a specific indicator based on RECOVER’s documentation of performance measures (RECOVER, 2007b), and triggers for management action. This information can be used to assess when options for adaptive management actions, specified in the document, are needed. However, for many of the identified uncertainties, the goals and targets have not been defined and are shown in the tables simply as “TBD” (to be determined). As discussed earlier in this chapter, quantitative objectives are critical to successful adaptive management. The lack of specific, program-level quantitative restoration objectives is a long-standing issue in

building the adaptive management program (NRC, 2008). Without the articulation of relevant, quantitative restoration objectives for the program, progress in evaluating the CERP cannot be achieved.

Overall, the Program-Level Adaptive Management Plan asks highly relevant questions about the CERP (see Box 5-1), touching on many of the systemwide issues highlighted in Chapter 4, such as storage, defining goals, and climate

change. The document also helps to outline some of the steps that need to be taken (and when) to address these questions and inform future decision making. Project-level and systemwide monitoring, while informative, is not sufficient to inform the challenges and tradeoffs in decision making and management at the program level. Strategies for addressing uncertainties, for example, to determine how sea level rise will affect restoration efforts, also require continuing research

and modeling with specific application to the South Florida ecosystem. The CERP programmatic adaptive management strategies identified to address the Priority 1 uncertainties include research, modeling, and synthesis, in addition to monitoring, and where specified, these strategies provide concrete actions to inform future decisions. In general, the strategies are only described briefly, and some need to be expanded to be fully responsive to the uncertainty from a systemwide perspective (for example, including downstream performance measures in an analysis of the sufficiency of storage). These research, modeling, and synthesis efforts should be forward-looking, and consider a range of future conditions under climate change, even for questions that are not specifically focused on climate change. The management options matrix is presented only for assessing monitoring results of “actual performance” but those options could just as easily apply to the results of forward-looking modeling analyses or research. Adaptive management should not be viewed as an activity that is only needed 5-10 years after several projects are fully constructed and monitoring data is collected—instead, RECOVER (2015) recommends that adaptive management at the program level is needed “now” or “immediately” for nearly every Priority 1 uncertainty with an action time identified.

Implementing Program-Level Adaptive Management

Adaptive management is a required and essential component of the CERP designed to ensure that new knowledge is linked to decision making so that restoration goals can be achieved in the most effective way possible, but little progress has been made in implementing program-level adaptive management to date. Successfully addressing the mission critical uncertainties identified in the CERP Program-Level Adaptive Management Plan (RECOVER, 2015) requires a structured implementation strategy that is currently lacking. An implementation strategy would identify tasks that require immediate action, a timeline for accomplishing this work, and a budget to accomplish each of the tasks. Additionally, an implementation strategy should outline the entities qualified to conduct each of the tasks (i.e., RECOVER, the Interagency Modeling Center, CERP agencies, universities, consultants), so that staffing and other resource needs can be better understood. Addressing these program-level uncertainties will require additional agency resources. RECOVER funding and staffing have been cut substantially over the past decade. Additional dedicated resources may be necessary to make sufficient progress and ensure systemwide and forward-looking perspectives on each of the questions, and effective communication of the results to CERP decision makers and the public. If resources are not sufficient to address all the priority uncertainties for which action is needed “now” or “immediately,” additional prioritization will be needed, with the potential risks of such delays clearly articulated.

Rapid implementation of a fully developed Program-Level Adaptive Management Plan is critical and long overdue given the timelines identified in the Programmatic Regulations. As a means to track implementation progress, the CERP should develop plans for periodic reporting on the extent of progress made toward restoration goals, whether thresholds for action have been crossed, what decisions have been made or modified, and what the outcome of the management response has been. Although the Systems Status Report (RECOVER, 2014) provides an existing structure for such reporting, it focuses on recent trends and current status rather than considering expectations of future performance. Communication strategies that focus on providing detailed information on the progress made in addressing the highest priority uncertainties is needed. The CERP can learn from approaches used in other systems. Reports with clear graphics that quickly convey progress, summarize the overall conclusions, and indicate the trends in recovery are valuable to communicate progress. To complete the adaptive management feedback cycle, the CERP needs to go further and specify how decisions are being made and will be made to adjust program management if uncertainties are not resolved. If the CERP adopted this strategy, annual reports would be a useful and timely way to indicate progress towards the resolution of the most important scientific questions related to restoration. Such reports should also indicate what changes to CERP goals are necessary if uncertainties cannot be resolved within the specified timelines.

SYSTEMWIDE ANALYSES NEEDED TO ENSURE CERP ACHEIVEMENT

A program as extensive and complex as Everglades restoration must by necessity be implemented a few projects at a time, but this fact makes the need for a periodic holistic look at system-scale response ever more critical. Renewed attention is needed toward the future of the ecosystem and how society can shape it through the CERP and other non-CERP restoration efforts, considering the new information that has developed since the CERP was launched. Current system status reports focus primarily on documenting the character of the current system; they do not look forward or stimulate thinking about the future state of the system. A periodic holistic assessment would consider new scientific and technical information, changing conditions in both natural and human system, and the very real constraints imposed by funding and regulatory processes, and it provides a means of assuring that each project can be planned, designed, and implemented to work within the system context. As outlined above, the programmatic regulations for CERP envisage this need by requiring the development of goals against which progress can be measured, as well as periodic assessment and evaluation resulting in CERP Updates. Such assessment and evaluation goes above and beyond the activities necessary to address the program-level adaptive

management uncertainties, although, if structured correctly, could help address some of these uncertainties (e.g., storage). CERP Updates were considered a key component of programmatic adaptive management in the Programmatic Regulations, and the CERP agencies need to renew their attention to these key forward-looking evaluations of progress and performance at the program scale.

Components of a Forward-Looking CERP Assessment and Evaluation

What is envisioned in such a holistic assessment? In Chapter 4 the committee identified a number of new developments and issues that potentially constrain the achievement of the originally envisaged restoration (e.g., feasibility of planned CERP storage, climate change). Coincidentally, many of these issues have also been identified as Priority 1 uncertainties in the Program-Level Adaptive Management Plan (RECOVER, 2015), reiterating their importance to the future of the program. The committee recommends that existing and developing models (discussed later in this chapter), together with any new tools needed to provide a complete assessment, be used to address important questions about the future of the ecosystem and the role of the CERP. These questions include but are not limited to the following:

• What is the effect of reductions in storage in the degree of change that can be achieved relative to CERP goals?
• How does presence vs. absence of particular future CERP components affect future ecosystem conditions systemwide?
• How sensitive is this future system state to key assumptions about important but currently unknown externalities such as future climate or sea-level rise rates?
• Does the knowledge gained since the late 1990s (e.g., pre-drainage hydrology, sea level rise) require refinement of the broad directional goals laid out by the CERP agencies?

The analyses appropriate to address these questions at the system scale need not be as detailed as has been conducted for project-level analysis. That does not mean this system-level analysis is straightforward. The committee recognizes that consideration of the entire system limits the complexity that can be considered, and some simplifications or generalizations will be needed. Limitations in analyses conducted in the near-term should be acknowledged while continued investments are made to improve system-level predictions of ecosystem condition (as discussed later in this chapter). This new effort will not be simple or easy—but it is essential to the long-term success of Everglades restoration.

Experience with other large-scale ecosystem restorations indicates that conducting assessments of future states at the system scale highlights many of the

challenges of implementation (e.g., Lund et al. 2010; Peyronnin et al., 2013). For example, the lack of an optimal solution for all desirable facets of the system, the role of uncertainty about future conditions (e.g., rainfall amounts and timing), and changed circumstances over decades (e.g., sea level rise, encroachment on the natural system by development) can limit the ability to achieve outcomes that were once considered possible. Although some might consider that illuminating such issues makes a complex stakeholder interaction even more challenging, not confronting these issues in a science-based, objective manner can lead to even less desirable circumstances, such as unrealistic expectations, litigation, and waning public or congressional support.

Continued long-term restoration support demands that a clear vision of the future is articulated and that the program is responsive to new information. Ultimately, such analyses can help all engaged and interested in Everglades restoration see the future that those on the ground are so diligently striving for. Such analyses stretch the bounds of our science, and they may also show the bounds of our ability to change the system. But they can also demonstrate the benefits of continued investment, even under alternate futures and the severe consequences of not following through with the restoration vision.

Shortcomings in restoration outcomes identified in this assessment will illuminate the need for modifications, either in future project planning efforts or in the restoration goals and objectives themselves. The Programmatic Regulations identify a process for those modifications to occur (i.e., a CERP Modification Report) that can be initiated if appropriate. From the perspective of national discussion of large-scale ecosystem restoration, this is a true application of the adaptive management promise of the CERP. The CERP can again be an example to other large-scale programs with multi-decadal implementation that considering new information and changing circumstances can lead to better long-term outcomes.

Effecting a Systemwide CERP Assessment/Update

This forward-thinking analysis, assessment, and evaluation require a focused effort and a dedicated team. The Science Coordination Group could provide important leadership and serve as a forum for public input, as was done in the Central Everglades Planning Project. RECOVER staff may be available to contribute to this effort, but CERP agencies need to ensure that the right team is in place to execute this system-level analysis and not rely solely on the skill sets and experience they use for other existing tasks. This systemwide assessment should not take effort away from ongoing and anticipated project planning efforts as different skill sets and tools are required; neither should it slow down or delay the current implementation of projects identified in the Integrated

Delivery Schedule (IDS). Continued implementation of projects already planned, authorized, and funded should continue. However, proceeding with continued project-scale planning without a systemwide understanding of the implications of major changes could lead to poorly informed decisions. If the system-level analysis does detract from project-level planning, the committee suggests that the system-level analysis be prioritized to provide an improved context for project-level decisions. Any delays in ongoing or near-term project-level planning should not delay overall restoration progress, because there are enough authorized (or soon to be authorized) projects that are expected to fully use available funds for at least the next 15 years, as noted in the IDS.

The goal is to develop within a limited time frame (i.e., 18-24 months) a clear vision of what successful CERP implementation might achieve under anticipated or possible future conditions. CERP agencies should commit the necessary resources to meet such a timeline. This analysis can utilize many of the modeling tools and approaches discussed in the next section.

Once this forward-looking assessment is conducted (including but not limited to a CERP Update, as described in the Programmatic Regulations), it should be used by CERP agencies to consider their path forward and whether adjustments to the CERP are needed. This process—the realization of program-level adaptive management, as originally envisioned in WRDA 2000 and the Programmatic Regulations—requires clear communication of the findings of such technical assessments to decision makers and to stakeholders. Building clear and credible linkages between science and decision making is challenging. Adaptive management at the program scale may illuminate the need for further refinement of CERP governance structures (including the linkages between the CERP decision-making agencies [i.e., USACE and SFWMD] and RECOVER, the Task Force, and the Science Coordination Group) to ensure that new information generated is used appropriately to guide restoration decisions to best support long-term restoration objectives.

TOOLS TO SUPPORT FORWARD-LOOKING ANALYSES

Since the CERP was developed, much has been learned and new tools have been developed based on advances in knowledge and data availability, supported by advances in computation efficiency. Such advances enable thinking about the future of the Everglades, how the forces of people and nature shape that change, and which actions can best promote a vibrant and sustainable Everglades ecosystem. In this section, the committee discusses how these tools could be leveraged to address some of the program-level uncertainties and improve the restoration to maximize benefits and avoid unacceptable impacts, even under changing climate conditions. Substantial advances have been made

in the capabilities of CERP hydrologic and hydraulic models, which have been the primary means by which alternatives have been evaluated, particularly in the early years of CERP planning. Those advances are not documented in this review, but interested readers could consult NRC (2007) and Obeysekera et al. (2011c) for more information. Instead, this section focuses on models that are linked to hydrologic models to evaluate the ecological and water quality outcomes of restoration alternatives. The committee also discusses strategies to use hydrologic models to assess future climate scenarios. These tools are available or can be readily developed, as demonstrated by other large-scale ecosystem restoration programs (e.g., Peyronnin et al., 2013), to take advantage of the extensive available monitoring data to show what the future might hold for the Everglades based on options for project implementation and operation, and climate change.

Ecological Modeling

The use of ecological models was limited in the development of the CERP (USACE and SFWMD, 1999) and the planning and implementation of the restoration that followed. In contrast, restoration planners had more experience and confidence in hydrologic models and have relied on these tools for assessment and planning (see NRC, 2007). As a result, hydrologic restoration goals have been emphasized over ecological restoration goals in the restoration effort (see Chapter 4). Hydrology is not viewed as more important than ecology. However, as hydrology is manipulated through restoration and an important ecological driver, hydrologic goals became primary by default, with ecological outcomes projected by assumption rather than analysis.

During the development of the CERP, this approach was necessary as only hydrologic models could be applied for systemwide analysis necessary to evaluate alternative management scenarios. Past reports of this committee have been critical of the lack of progress in developing and integrating linked hydrologic, water quality, and ecological modeling tools (NRC, 2007, 2008, 2010, 2012). Fortunately, ecological modeling capacity has advanced considerably since the committee’s last assessment of ecological modeling (NRC, 2012). This advancement has been stimulated by the activities of the Joint Ecosystem Modeling (JEM) effort, a partnership among state and federal agencies, universities, and other organizations dedicated to research and development of ecological modeling in support of the restoration.² A number of useful ecological models are now available that can link to hydrologic models to simulate effects of restoration activities on particular species or habitats, at local to systemwide scales (Appendix C, Table C-1). The capability now exists to evaluate restoration alternatives based on

___________________

² See https://www.jem.gov/.

predicted ecological outcomes, coupled with the predicted hydrologic outcomes that have traditionally been used. Indeed, these ecological models are now being regularly employed in evaluating restoration alternatives, some on a systemwide scale. Thirteen ecological models were used in the development of Central Everglades Planning Project (USACE and SFWMD, 2014a); ecological models of marl prairie habitat and wood stork foraging conditions were heavily employed in evaluating water management alternatives in the biological opinion issued for the Everglades Restoration Transition Plan (ERTP; FWS, 2016) (see Chapter 3).

The marl prairie index model, employed in analyses associated with both the Central Everglades Planning Project and the ERTP, is one example of an ecological model with capability to make systemwide projections. It does so through linkages with the Regional Simulation Model (RSM) and Natural System Regional Simulation Model (NSRSM)—models capable of simulating the hydrologic conditions under pre-drainage, current, and future (with or without restoration) scenarios. This marl prairie index model enables analysis of one of the most difficult aspects of the restoration, continuing to provide the habitat on which the endangered Cape Sable seaside sparrow depends while restoring other habitats in the southern Everglades (see Chapters 3 and 4). Another example is the suite of wading bird models (Wader Distribution Evaluation Modeling, WADEM) that forecast potential wildlife responses to changes in water management and climate (Beerens et al., 2015). Using those models, investigators have the ability to forecast the changes in water levels that drive the spatial patterns of suitable foraging habitat for wood storks and other wading birds, enabling managers to predict the likely effects of changes in water management options.

Given the capability of these and other ecological models, such tools should be routinely employed during restoration planning, at both project and systemwide scales, as they were in evaluating alternative restoration scenarios related to the Central Everglades Planning Project and the ERTP. The development of EverVIEW tools that enable visualization of some ecological model outputs add to the utility of employing ecological models. Ecological models will be useful in evaluating not only impacts of alternative restoration scenarios on individual species and habitats but also tradeoffs between the needs of different species and restoration goals. Efforts are under way to integrate models to enable multi-species assessments that could be used to evaluate ecological tradeoffs among alternative management plans. Inverse ecological modeling tools are also being developed that could be used to optimize restoration features to reach multiple ecological targets. Ecological models also can be used to explore the implications of climate change and sea level rise, changes in CERP project feasibility, and the improved understanding of the pre-drainage system (see Chapter 4). The incorporation of ecological models linked to hydrologic models into evaluation of restoration alternatives is an important recent advance in restoration planning.

Biogeochemical Modeling

Improvement of both water quality and hydrology are needed to reverse the decline of key Everglades attributes (NRC, 2012). Although water quality is an important component of “getting the water right,” water quality modeling continues to lag behind the development and application of hydrologic and ecological modeling as part of the CERP toolbox. The reasons for this lack of emphasis are several. Water quality modeling is a challenging undertaking, particularly for a large, complex landscape like the Everglades. Also, under the consent decree, water will not be redistributed within the WCAs and Everglades National Park unless the total phosphorus concentrations are below established limits, and under these conditions, the risks of adverse water quality may be diminished. However, given the potential for water quality criteria to limit new water inputs or the redistribution of existing flows, water quality modeling becomes even a more important tool to examine the implications of CERP projects on total phosphorus throughout the ecosystem, especially when water quality is close to the established phosphorus criterion. Without such modeling tools to foster further examination of scenarios at the interface of water quality and quantity, decision makers are more likely to be risk averse when confronted with decisions pertaining to water quality possibly to the detriment of key ecosystem components driven by the system’s altered hydrology. Water quality modeling could also be used to better understand the transport and fate of contaminants within the Everglades. Environmental quality should be an important driver affecting wetland habitat (e.g., stands of cattails), but water quality is not addressed in current ecological models. An improved capacity to simulate water quality could lead to improved ecological modeling tools. The development of a regional coupled hydrologic water quality model would provide an important tool for quantitative evaluation of a range of alternative restoration scenarios and their potential short- and long-term effects on biotic and abiotic attributes.

Some limited water quality modeling has been used to inform Everglades restoration projects. The Everglades Landscape Model (ELM) was been used in the Decomp planning process to assess water column phosphorus concentrations and sediment accumulation rates within the WCAs. The Dynamic Model for Stormwater Treatment Areas (DMSTA) was used in the Central Everglades Planning Project (USACE and SFWMD, 2014a) to assess mean phosphorus concentrations and water quality constraints in project implementation. The Watershed Assessment Model was used to develop the basin management action plan for Lake Okeechobee (FDEP, 2014). Additional details and a summary of relevant water quality models that have been used in Everglades-related research is summarized in Childers et al. (2011).

Water quality modeling requires a hydrologic modeling framework, and the current RSM potentially provides this framework for the Everglades. The RSMWQ (the water quality engine for the RSM) is a model designed to run with the RSM. It has two components to simulate water quality—the first is for transport of soluble and dissolved constituents, and the second is a flexible biogeochemistry module that allows the model user to define the state variables and algorithms to describe biogeochemical cycling in aquatic ecosystems. Unfortunately, little progress has been made in the development and application of the RSMWQ in recent years, even though it has been under development for over a decade (James and Jawitz, 2007; Jawitz et al., 2008).

The development of a comprehensive water quality model remains an important but ambitious goal. Such a tool would allow the user to address a host of interconnected issues beyond the management and fate of inputs of total phosphorus, including loss and accretion of peat, effects on wetland habitat, and linkages to other water quality contaminants like nitrogen, sulfur and mercury, in the Everglades and in downstream coastal ecosystems.

Climate Modeling Tools and Approaches

The CERP was originally designed for the next 5 or 6 decades assuming that historic climate will continue, but this stationarity assumption is no longer appropriate for multidecadal restoration and water supply plans. As discussed in Chapter 4, it is essential to assess the performance of CERP under a changing climate—that is, to project the systemwide hydrologic and ecological outcomes from the restoration plan under different climate scenarios. Several modeling tools and strategies are available to support forward-looking evaluations of the CERP.

Climate change scenario analysis in CERP planning has been limited to project-based assessments of the loss of benefits associated with sea level rise (e.g., USACE and SFWMD, 2014a). Additionally, analyses have been performed to assess the hydrologic and ecological impacts on base conditions of a 1.5°C increase in temperature coupled with ±10 percent change in average annual precipitation (Obeysekera et al., 2015; see also Chapter 4). An important next step is to examine the hydrologic and ecological responses a broader range of feasible future scenarios, including seasonal shifts in the distribution and quantity of precipitation. These analyses should consider not only changes to base conditions but also the effect of the CERP and various feasible storage configurations on hydrologic and ecological outcomes. Such an analysis could help address the critical uncertainty of how much storage is needed to meet the goals of the CERP under alternative future conditions. These analyses are possible with existing hydrologic and ecological models, and over time improvements in regional

climate modeling tools and refined regional climate projections will narrow the uncertainty associated with current analyses. With existing and improving modeling tools, including ecological models that incorporate climate change factors, scenario analyses considering a variety of feasible future conditions can be used to evaluate potential impacts of climate change and sea level rise on ecosystems, key species, and habitats as well as how restoration efforts can be used to alter these outcomes.

It is not only possible, but likely, that optimizing restoration planning for one scenario may preclude choices that accommodate other scenarios, and decision making focused on robust outcomes can provide more assured outcomes for public investments. Thus, decision-making tools are needed that address uncertainty and vulnerability of both human and natural systems. A number of modeling and data-supported approaches have been developed to aid in planning and managing integrated natural and built infrastructure projects such as the Everglades restoration under scenarios of climate change. Robust decision making (e.g., Groves et al., 2013) is an analytic framework that helps identify potential robust strategies, characterize the vulnerabilities of such strategies, and evaluate tradeoffs. Robust decision making has been used to evaluate vulnerabilities and climate adaptation options for the Colorado River Basin and evaluate tradeoffs among different scenarios of climate change and water management strategies. Poff et al. (2015) introduced eco-engineering decision scaling, a methodology that explicitly and quantitatively explores tradeoffs in stakeholder-defined engineering and ecological performance metrics across a range of possible management actions under unknown climate and hydrologic futures. This methodology was applied in the analysis of flood management adaptation options in the Iowa River, in a USACE-built system consisting of a dam (Coralville) and floodplain levees that have been found to be increasingly vulnerable to extreme events. Ray and Brown (2015) developed a decision-tree framework to assimilate climate change information into water resources project planning and design. This approach has been used in a number of studies (e.g., California, Mexico City, Brahmaputra River basin, Bangladesh) to improve the resiliency of water resources infrastructure to climate change, sea level rise, and extreme events. These approaches explicitly incorporate the types of decisions to be made within restoration program planning, a feature that could be particularly useful to the CERP.

CONCLUSIONS AND RECOMMENDATIONS

When the CERP was launched in 2000, adaptive management was embraced as a means of incorporating new information into the plan and addressing unforeseen issues related to the plan, and the CERP was widely viewed as a

leader in adaptive management. Since that time, a framework for CERP adaptive management has been developed, and a structure for implementation at a project-level adopted, but the original vision of adaptive management at the program level remains unfulfilled. In this chapter, the committee outlines steps that need to be taken for the CERP program to continue to lead in adaptive management and, more importantly, to ensure restoration success by incorporating new knowledge and changing circumstances into the restoration plan at the systemwide scale.

The CERP has made limited progress in articulating restoration objectives that are sufficiently quantitative to support effective planning, implementation, and assessment. An effort is now needed to develop quantitative restoration goals that capture new science and address potential conflicts in restoration. When authorized, the CERP goals were broad narrative statements on restoring the South Florida ecosystem and ensuring that the water needs of the region were met. Reaching these goals requires that realistic, quantitative objectives be developed and applied to project- and program-level restoration, which in turn requires consideration of the inherent tradeoffs that must be made in any complex ecosystem restoration program (as discussed in Chapter 4). Work has stalled on improving the quantitative interim goals (RECOVER, 2005a), which were not adopted because of the substantial assumptions that were made in their development. Developing quantitative objectives is an essential component of adaptive management, and once established, these objectives should be periodically revisited to ensure they are still desirable and achievable given new knowledge and modeling capability and major changes that affect future systemwide operations under the CERP.

The CERP Program-Level Adaptive Management Plan is an important first step in identifying critical uncertainties affecting restoration progress, but it requires an implementation plan and sufficient resources to be effective. The plan asks highly relevant questions about the CERP that are related to questions of storage, design and implementation, and climate change. Many of the questions can and should be addressed now through new research and modeling in addition to ongoing monitoring. Monitoring alone cannot address the challenges and tradeoffs required for decision making and management at the program level. The RECOVER (2015) report concludes that a failure to address the Priority 1, mission-critical uncertainties will paralyze progress toward meeting CERP restoration goals and that many of these uncertainties need to be addressed immediately, but no actions have been taken to implement the plan. To expedite implementation of the Program-Level Adaptive Management Plan, an implementation strategy to address the Priority 1 uncertainties is needed that identifies tasks, timelines, resources, and staffing needed, and the highest priorities if sufficient funding is not available for the ideal implementation plan.

A systemwide analysis of the potential future state of the Everglades ecosystem, with and without CERP and other restoration projects, should be conducted in conjunction with a CERP Update, which is long overdue. The regular 5-year CERP updates called for in the Programmatic Regulations to evaluate the restoration plan considering new scientific, technical, and planning information have not been routinely conducted. A holistic, forward-looking analysis of the possible future state of the ecosystem is needed in the light of new knowledge gained over the past 16 years. This analysis should consider various scenarios for climate change and sea level rise, and explore the ecosystem implications of various options for future CERP implementation. By exploring alternative future scenarios, considering uncertainties in climate or funding to support implementation, decision makers and stakeholders will be better informed of the implications of near- and long-term decisions. The halfway point in the original CERP timetable is an appropriate time for such analysis and evaluation of the future condition of the ecosystem. Challenges identified by this analysis may illuminate the need for modifications, either in future project planning efforts or in the restoration goals and objectives themselves. Although some might consider that illuminating such issues makes a complex stakeholder interaction even more difficult, failing to confront these problems in a science-based, objective manner can lead to even less desirable circumstances, including unrealistic expectations, litigation, and reduced public or congressional support. The analysis and evaluation process conducted as part of the CERP Update will enable the CERP agencies to ensure restoration expectations are clear and can be achieved and to determine if further modifications of the CERP, as allowed for in the Programmatic Regulations, are needed.

Developed and developing tools exist that can support forward-looking analyses of the CERP for project and systemwide analyses. Tools and strategies are available to explore future climate change and sea level rise scenarios, examine the robustness of the CERP to these potential futures, and enhance decision making under uncertainty. These approaches can illuminate opportunities to adapt the restoration plan to changing precipitation, hydrology, and sea level rise and mitigate the impacts of climate change. The capability for ecological modeling has advanced in recent years, to the point that models can be used to project systemwide effects of restoration activities for a variety of ecological performance measures. Ecological models link the response of species and habitats to underlying hydrologic models at local or systemwide scales and allow alternatives to be evaluated based on projected ecological outcomes. Ecological models are now being used along with hydrologic models in planning and assessments related to restoration—a major advance. Ecological models may be especially useful in evaluating tradeoffs between restoration goals and targets. In contrast, development and application of water quality models in the CERP con-

tinues to lag behind the use of hydrologic and now ecological models. Robust and well-tested water quality models are important tools to inform restoration strategies, particularly those that involve new water flows or redistribution of existing flows, and continued attention is needed to developing these models. The development of a robust Everglades water quality model is a key need moving forward. Improved water quality modeling tools also should lead to further refinement of ecological models, since Everglades habitat, species distribution, and ecological functioning are closely linked to water quality. As modeling advances toward an integrated set of tools to evaluate hydrologic, water quality, and ecosystem response to changes, there is a need for comprehensive sensitivity and uncertainty analysis of these linked models to inform and guide assessment and planning decisions.