WORKSHOP ON THE DESIGN AND SPECIFICATIONS
FOR THE FLORIDA BAY WATER QUALITY MODEL:
REPORT OF THE MODEL EVALUATION GROUP
Workshop Moderator, Dominic M. DiToro
22-24 October 1996
Key Largo, Florida
Christopher F. D'Elia, co-chair
CBL, Center for Environmental and Estuarine Studies
University of Maryland System
U.S. Geological Survey
Department of Civil Engineering
University of Virginia
Steven C. McCutcheon, co-chair
Hydrologic and Environmental Engineering
Submitted to the Program Management Committee
Florida Bay Research Program
SUMMARY AND RECOMMENDATIONS
The Florida Bay Program Management Committee (PMC) convened a workshop in Key Largo, Florida from October 22-24, 1996 to focus on the identification of model components and activities critical to the management of Florida Bay. A "Model Evaluation Group" (MEG) was impaneled to observe the workshop proceedings and make recommendations about future directions of the Florida Bay modeling effort. The MEG's major recommendations are given below.
1. Practical management alternatives and issues must motivate the modeling effort. The model must be able to:
2. Important issues to be explicitly addressed by the water quality model include:
3. Development and early implementation of a hydrodynamic model to link to the water quality model is a high priority. The CH3D model has been linked to water quality and sediment transport models elsewhere; however, it will be very difficult to develop a linkage ab initio for the RMA2 model (Berger, workshop presentation). Accordingly, the MEG believes that in the long run, the CH3D model offers an important advantage over RMA2.
4. Expectations and goals should be explicitly agreed to by key players at the outset and be kept realistic by:
It is vital that a simple, large box model be formulated prior to linkage of the hydrodynamics model as recommended by COE/WES (Dortch, workshop presentation). Preliminary models are crucial to confirm and further develop the initial mass balance calculations presented by Walker (workshop presentation); to gain an early appreciation of the behavior of the Bay; to facilitate development of detailed models; and to guide future data collection.
5. An elaborate modeling system cannot address all resource management issues. Some processes are not sufficiently understood at the present time, requiring special research investigations to define parameters and kinetic rates.
6. The members of the PMC and their agencies need to set priorities, coordinate effectively, and communicate goals and objectives clearly. Recommendations made by previous review panels with regard to organizational improvements for program management should be pursued without further delay.
As part of the process of designing and providing specifications for the modeling effort on Florida Bay, a modeling workshop was convened by the Program Management Committee (PMC) from October 22-24, 1996 in Key Largo, Florida. A "Modeling Evaluation Group" (MEG) was impaneled to evaluate and guide the PMC's modeling efforts. This panel included Ted Callender, Chris D'Elia, Winston Lung, and Steve McCutcheon. The following report summarizes the meeting and gives the panel's recommendations for future directions.
Discussion at the outset of the workshop set the stage for the ensuing discussion by reviewing progress of the past several years. Accordingly, David Rudnick, who presented introductory comments, urged the group to frame modeling efforts in the context of the central questions that have been addressed by the previous PMC efforts, review panels and workshops (Table 1).
Table 1. The "central questions" of the Florida Bay interagency science program.
1. How and at what rates do storms, changing freshwater flows, sea level rise and local evaporation/precipitation influence circulation and salinity patterns within Florida Bay and influence the outflow from the Bay to adjacent waters?
2. What is the relative importance of the input of external nutrients and internal nutrient cycling in determining the nutrient budget of Florida Bay? What mechanisms control the sources and sinks of the Bay's nutrients?
3. What regulates the onset, persistence and fate of planktonic algal blooms in Florida Bay?
4. What are the causes and mechanisms for the observed changes in seagrasses and the hard bottom community of Florida Bay? What is the effect of changing salinity, light and nutrient regimes on these communities?
5. What is the relationship between environmental and habitat change and the recruitment, growth and survivorship of higher trophic level species?
While these organizing questions provide the framework for the entire Florida Bay effort, the present task is more constrained and pertains to issues discussed at workshops directed at hydrodynamics (April, 1996) and nutrients (June, 1996). The review panels at the previous workshops have made a number of substantive recommendations that should also be considered (Armstrong et al., 1996; Boesch et al., 1996; Boesch et al., 1995).
In its report (Armstrong et al., 1996), the review panel at the April, 1996 Hydrodynamics Workshop considered the following major topics: model appropriateness, system representation, boundary conditions, interfacing with water quality models, field measurements, and model and data collection evaluations groups. The panel's major recommendations are summarized in Table 2.
Table 2. Key recommendations of the April, 1996 Hydrodynamics Workshop.
1. The 2-D RMA2 finite element model is appropriate for an initial exploration of Florida Bay hydrodynamics. The number of elements currently in the RMA2 model probably can be reduced
2. A 3-D model may be required to deal with western regions.
3. Boundaries should be expanded northward and westward to include Shark River influence, as well as circulation along the west coast of Florida, and southward to include offshore of the Keys
4. The review panel was concerned regarding linkage of RMA2 with sediment transport and water quality models.
5. Water quality modeling, including seagrasses and benthic exchanges, is an essential tool in the development of the restoration plan for Florida Bay.
6. A central repository for data and vigorous coordination of field measurement programs should be initiated.
In its report (Boesch et al., 1996), the review panel at the Nutrient Workshop considered critical issues related to modeling that will be essential for the restoration of the ecosystem (Table 3). Major topics considered by the review team were the adequacy of databases, research and monitoring programs, modeling needs, restoration objectives and strategies, and the Florida Bay science program management.
Table 3. Recommendations related to modeling from June, 1996 Nutrient Workshop.
1. Pursue the development of a coupled circulation-ecosystem model of Florida Bay as a tool to systematize data, pose hypotheses, and anticipate the effects of different water management scenarios.
2. Include as key features in the model: (1) coupled hydrodynamic-nutrient-phytoplankton-water quality variabilty [a "water quality" model], (2) suspended sediments and their influence on turbidity, and (3) seagrass populations and influence on sediments, nutrients and geochemistry.
3. Critical questions to address in the model include: (1) What is the fate of nutrients discharged from the Shark River? (2) What is the biological availability of particle-bound P? (3) Why are the features of nitrogen cycling in Florida Bay apparently distinct from those in other coastal water bodies? (4) What are the mechanisms that trigger and sustain algal blooms in Florida Bay? And (5) What are the origins of turbid water?
The ad hoc review panel at the nutrient workshop further emphasized that certain essential state variables must be measured continually in a monitoring program and that fundamental processes, such as mineralization of TOP and TON, must be measured via experimentation. Furthermore, the panel observed that "the temptation to produce a model which answers all possible questions (e.g. regarding food webs and living resource production) should be avoided." The panel asserted emphatically that the restoration program should have more clearly stated goals, and that the Florida Bay Interagency Science Program program should make improvements in management in a number of areas.
CONSENSUS ON UNDERSTANDING AND MODELING FLORIDA BAY
Overall Progress on Understanding and Modeling Florida Bay
The MEG was impressed with the level of interest, commitment and activity that is now being shown by federal and state agencies, academic and government scientists, non-government organizations (NGO's), and the public for Florida Bay. Considerable progress has been made in understanding the structure and function of this unique resource and ecosystem, and several useful monitoring programs have been instituted at different times to provide important information on a regular basis about the Bay and its water quality.
The modeling workshop that this report addresses was highly successful and relevant to issues confronting the Bay. Attendees came well prepared and anxious to enter into a constructive dialogue. The workshop moderator, Dominic DiToro, did an exemplary job in keeping the focus of the group on achievable goals, while at the same time allowing sufficient latitude for fruitful scientific discussion at a more detailed level.
Consensus Reached by Workshop Participants
The open scientific dialog was intrinsically useful, but also meshed well with preparations by the Corps of Engineers, William Walker, and others in formal presentations (see the attached schedule of presentations). The result was a solid consensus on all of the important elements of an integrated water quality-sedimentsea grass model. At the end of the workshop participants explicity agreed that:
The workshop consensus was based on important points made during formal presentations first by David Rudnick on the management issues (covered in the Introduction above), and then by Mark Dortch of the COE, who stated that the purpose of the workshop was to develop a report that he would write and present at the December 1996 Annual Science Conference. The report is intended to be a general assessment that will define general modeling requirements to address a range of management issues. The report is not intended to be a work plan for a particular organization, nor to bias any peer-review and management selection of any group. However, Mark Dortch and the COE will clearly have an inside track based on experience in water quality modeling and preferred funding mechanisms through the Jacksonville COE. Nevertheless, it is important for team building and managing the Florida Bay Program that any proposals from WES/COE, or anyone else invited to propose an integrated model, be thoroughly peer reviewed and assessed by the Florida Bay PMC.
Based on the background of the meeting, the MEG generally agreed with the consensus, reaching essentially the same conclusions on the modeling framework in a separate meeting the night before, with only two exceptions. First, the unique system, consisting of a mantle of sediment over bedrock and largely discrete particles, seems ideally suited for mass balance modeling of sediment transport and dispersion. The key is whether sediment transport should be simulatecFor whether local resuspension and deposition occurs as was assumed in the workshop consensus. Second, the MEG lacked the experience and adequate information possessed by local scientists to reach specific decisions on the classes of SAV and phytoplankton to include in the model. Regardless, it is remarkable that an outside panel anticipated the wider consensus of experts knowledgeable of this unique system working in concert with modeling experts experienced with many systems. This is a tribute to the many investigators present who shared concisely, freely, and openly of their own experience in Florida Bay and of the advancing state of the art of modeling now conducted for many water bodies.
KEY ISSUES OF CONCERN TO THE MODEEING EVALUATION GROUP
In the same constructive spirit that guided this workshop and the April 1996 workshop on hydrodynamics, the present report both confirms the MEG findings that the workshop consensus is generally sound, and addresses potential problems and gaps in knowledge. In general, the Florida Bay investigation is well conceived and sensibly executed, but improvements are possible.
Sediment Resuspension and the Effect on the Light Field
The MEG believes that good modeling codes are presently available for calibration to simulate resuspension and deposition, whereas the workshop consensus was that light attenuation as influenced by local resuspension and settling, should merely be simulated descriptively. The MEG further believes that if phosphorus desorption kinetics are important in nutrient cycling, the model calculations should portray accurately different particle transport and residence times in the water column. The MEG is concerned that a simple conceptual model of resuspension, nutrient release and deposition on an unspecified time scale may not adequately address Florida Bay conditions.
Sediment resuspension is potentially important to P release from the sediments and to the subsequent resorption on the surface of resuspended material. Accordingly, sediment resuspension represents a potential source/sink of P in the water column and thus relates in a different way to light attenuation incorporated in the phytoplankton growth model. The conceptual model proposed by Reddy at the workshop related to P uptake by benthic periphyton and possible transfer to "leaf" organic matter, suggests that wind-generated resuspension events might translocate the CaCO3-sorbed P to other parts of the pelagic system. This P, in turn, might be released to solution and available for rapid incorporation into organic matter. On the other hand, diagenetic release of P to interstitial water and the enrichment of sediment particles that reside in the benthos may provide a source of P to the water column when resuspension of these particles occurs, since sorbed P in resuspended sediment may subsequently become desorbed under the different physico-chemical conditions of the water column. One can only speculate at present, and there is need for much research on these topics.
Submerged Aquatic Vegetation (SAV) and Relationship to Water Quality
The whole area of growth, die-off, and decay of SAV species (Thalassia and Ha/odu/e) appears to be fundamental to the understanding of nutrient cycling in Florida Bay. Seagrasses are the most dominant biological system of Florida Bay; the canopy strongly affects the water column and 80% of the biomass resides in benthic sediment. It seems clear that there is limited understanding of the role of seagrasses in nutrient cycling.
The MEG suggests that the Program Management Committee (PMC) convene a group of experts with respect to seagrass ecology, seagrass dynamics, carbonate geochemistry, benthic processes, and water-quality modelers for the purpose of defining a conceptual model of nutrient cycling between the water column and benthic systems. The outcome of this exercise would be to identify critical research components that would yield quantitative information needed by the water-quality modelers. Then, the PMC could foster a few interdisciplinary studies to provide the necessary data.
Based on the discussions in the workshop, a consensus has been reached to develop a integrated modeling framework to address management questions for Florida Bay. The modeling framework should consist of the following components: a seagrass model, a water quality model, and a hydrodynamic model.
To keep the modeling effort to a manageable level and yet still robust enough to accurately quantify the seagrass effect on turbidity and nutrient recycling, two generic species of seagrass (i.e., fast growing and slow growing species) are recommended for the model. Model coefficient values should be derived from specific studies by local scientists. The water quality model will include a sediment benthic diagenesis module directly interacting with the water column kinetics. The pH-carbonate equilibria will be incorporated due to the unique geological characteristics of Florida Bay and the pH effect on phosphorus precipitation, adsorption, and desorption. Two phytoplankton species will be simulated in the water column. Another key element in the water quality model is the quantification of light attenuation in the water column, which is closely related to the suspended solids concentration. In lieu of a sediment transport modeling effort, a simplified approach is recommended: i.e., quantify bottom shearing stresses based on the wind speed and wind-generated waves on the water surface. The magnitude of the shearing stress therefore determines the resuspension flux of the solids. In areas covered with seagrass, resuspension rates are reduced, depending on the density of coverage.
While the fine, structured grid model, RMA2, is to be used to compute detailed circulation patterns and mixing in the Florida Bay, the total number of cells should be significantly reduced to make the computational effort reasonable.
Reduced numbers of grid points will insure that the code is transferable to the COE Jacksonville District and the South Florida Water Management District (SFWMD). In the meantime, the structured grid hydrodynamic model, CH3D, should be configured for the study area. The advantage of the CH3D model is that it can be directly linked with the water quality model CE-QUAL-ICM, a technolagy that has been successfully demonstrated in previous studies by the COE/WES modeling staff. The disadvantage is that local morphological and topographic features of Florida Bay are less resolved than with the RMA2. The relative importance of morphological resolution on general water quality patterns needs further assessment.
Modeling Strategy and Steps
The workshop participants have identified and discussed three primary technical issues for this modeling study:
1. Linkage between the hydrodynamic and water quality model. The model linkage issue can be approached in the following manner. First, the fine, unstructured grid model, RMA2 will be run to match the measured salinity distributions in the Bay. Once the model is calibrated and verified, its results will be saved and serve as a data set for calibrating the subsequent mass transport modeling effort. In a dual-track effort, the structured grid hydrodynamic model, CH3D, will be configured and run to match the salinity results of the RMA2 model. Matching the salinity field is a key test for the hydrodynamic modeling effort. The calibrated mass transport can then be used to drive the water quality model. It is important that both the CH3D and water quality models use the same spatial grid system for a direct linkage without any spatial averaging.
2. Effect of seagrass die-off on turbidity and nutrient recycling in the water column. Seagrass die-off affects the turbidity and nutrient recycling in the water column. Accordingly, model sensitivity runs are recommended to quantify seagrasses' effects on the hydrodynamics and water quality.
3. Nutrient loads to the Florida Bay. Since data on nutrient loads from the mainland are limited, it is necessary to use a watershed model or other methodologies to develop the loads from the mainland to Florida Bay.
The MEG agrees, in priniciple, with the COE that CH3D and CE-QUAL-ICM are adequate models for Florida Bay. However, any work plan or proposal for water quality modeling should be peer-reviewed prior to initiation of funding. The PMC should be open to any alternative proposals applying similar models.
The Roles of Research and Monitoring
Both research and monitoring play crucial roles in support of the modeling and management effort for Florida Bay. As the original Florida Bay science report (Boesch et al., 1993) observed, "we cannot overemphasize the need to provide sustained support for research, monitoring and modeling activities to provide managers proper information." Estuaries and coastal areas are notoriously "individualistic," and the modeling of estuaries has only in the last decade begun to develop adequate sophistication and technical reliability to be effective management tools. Although there are now excellent examples of coupled hydrodynamic-water quality models for estuaries, these have generally benefited by having identifiable support for research targeted to key needs, such as understanding nutrient and oxygen fluxes at the sediment-water interface. They also have intensive monitoring programs to provide baseline information about hydrographies, water quality, and natural variability.
There can be no doubt that Florida Bay possesses unique features that have yet to be included in any water quality model for an estuary. For example, the carbonate sediments in the Bay contrast markedly with the clastic sediments encountered in temperate estuaries such as Chesapeake Bay, Long Island Sound, and San Francisco Bay. The relationship between water column productivity and sediment oxygen demand/nutrient flux has never been sufficiently examined for a carbonate environment to provide information needed in the water quality model. Accordingly, we recommend that in addition to providing continued support for the Florida Bay monitoring program, that federal and state agencies join forces to establish an independent research fund for activities that provide information essential for development of the model. As the Florida Bay Science Review Panel has observed, "The Program Management Committee should commit to external peer review of proposals . . ."
Agency Coordination and Leadership
The members of the PMC and their agencies need to set priorities, coordinate effectively, and communicate goals and objectives clearly. Recommendations made by previous review panels with regard to organizational improvements for program management should be pursued as soon as possible.
We recognize that in most cases, individuals on the PMC are limited in their ability to commit their own agency to changes necessary to improve the management and coordination function. We also recognize that federal agencies do not have consistent and equivalent policies for managing research, funding extramural activities, and conducting peer review of scientific proposals. The recent comments of the Ad Hoc Committee on Nutrients (Boesch et al., 1996) and Peer Review Panel on Hydrodynamic Modeling about ways to strengthen the management of the Florida Bay Science Program remain relevant. Particularly important will be to "hire" or designate a full time program manager whose responsibility it will be to coordinate and track Florida Bay research and make appropriate recommendations to managers." We would supplement that statement with the further recommendation that the individual should also have a budget to service coordination functions (such as organizing and conducting workshops, paying travel expenses for panelists and advisors, preparing reports and communications to the public, etc.).
For necessary changes in management structure and function to be made, it may be wise to convene separately a meeting of responsible agency officials above the level of PMC principals. The PMC has wisely coordinated agency efforts, but it seems that more is possible. Given the dedication of the PMC, the responsible agencies owe it the effort to achieve even more effective cooperation.
Armstrong, N.E., DiToro, D.M., Hansen, D., Jenter, H, King, I.P., McCutcheon, S., Raney, D.C., and Signell, R. 1996. Report of the Florida Bay Model Review Panel on Florida Bay Modeling Workshop, April 17-18, 1996.
Boesch, D.F., Armstrong, N.E., Cloern, J.E., Deegan, L.E., Perkins, R.D., and Williams, S.L. Report of the Florida Bay Science Review Panel on Florida Bay Science Conference: A Report by Principal Investigators, October 17 & 18, 1995.
Boesch, D.F., Armstrong, N.E., D'Elia, C.F., Maynard, N.G., Paerl, H.W., and Williams, S.L. 1993. Deterioration of the Florida Bay Ecosystem: An Evaluation of the Scientific Evidence. Report to the Interagency Working Group on Florida Bay.
Boesch, D.F., Caffrey, J.M., Cloern, J.E., D'Elia, C.F., DiToro, D.M., and Walker, W.W. Jr. 1996. Florida Bay Nutrients: Perspectives on the July 1-2, 1996 Workshop.