Over the last 10 years, I have been studying the effects of excess nitrogen from urban/suburban areas on eelgrass ecosystems in coastal embayments. Together with my students and research assistants, I have examined how nitrogen loading changes the physical structure and food web relationships in eelgrass beds, and thus changes fish production (Fig. 1). We have documented declines in fish community abundance and diversity, and growth rates of individual species in eelgrass systems that have been degraded by nutrient loading.
Currently we are trying to bring all this information together in a model that will show the links between changing land use in the uplands and fish production in estuaries. This model will be most useful if it is used by local decision-makers. Thurs, we are also running some experiments designed to discover how to provide this model to decision-makers so that it is useful. This project is described below.
Social and Ecological Transferability of Integrated Ecological Assessment Models
Deegan, Linda (Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA), James Kremer (University of Connecticut), and Thomas Webler (Social and Environmental Research Institute, Leverett, MA).
Project Goals and Objectives: 1) To develop an empirically-based numerical simulation model of broad generality that links land use patterns and nitrogen loading to ecologically important and socially relevant endpoints of water quality, and fish diversity and abundance; 2) To measure the effectiveness to town planners of different model presentation formats and approaches, with the objective of determining the best and most efficient way to encourage scientifically-aware decisions at the crucial, local level of land use debate.
Approach: Our approach is to link a nitrogen loading model developed for Waquoit Bay, Cape Cod and its watershed, to an empirically-driven ecological response model that predicts phytoplankton and eelgrass productivity, system metabolism, probability of anoxic events, and fish abundance and diversity. Empirical data to build the model derives from studies in Waquoit Bay and in 15 sub-estuaries of Buzzards Bay (Massachusetts) sampled from 1993-1996. During summer 1998 we collected new data from these and other sites that are being used to validate the model. The model is the centerpiece in our study of presentation formats and approaches. Three presentation formats will be assessed for their effectiveness in encouraging willingness to use the model in planning decisions: 1) a lecture (‘sage-on-the-stage’) format, 2) a tutorial format, and 3) a deliberative workshop format, representing a gradient of increasing participant-model interaction. One of the three formats will be presented to the planning community in four separate towns (n = 4 for the three ‘treatments’), for a total of 12 towns participating in the presentations. These towns have stakeholders concerned with land use and estuarine resources, and will be drawn from coastal communities of southeastern Massachusetts, Rhode Island, and Connecticut.
Preliminary Findings: We have found that juvenile fish community characteristics and fish abundance in the studied estuaries are strongly influenced by the integrity and sustainability of eelgrass (Zostera marina) habitat (e.g., Fig. 2). This important and once common seagrass in southeastern Massachusetts estuaries is in serious decline, in part because of eutrophication effects. Eelgrass habitat quality and quantity will be an important component in the model. Initial responses to the social applicability of the model among the towns contacted indicate a high degree of enthusiasm and anticipation, and an eagerness to participate in the study.
Significance of Findings: Healthy eelgrass beds are necessary to sustain estuarine carrying capacity and biodiversity. In those communities bordering estuaries with extant eelgrass beds, land use decisions should minimize nitrogen loading that could jeopardize the integrity of this essential fish habitat.
Next Steps: The ecological generality (‘transferability’) of the model will be tested by comparing predicted and observed (1998) results, and will result in a refinement of the model prior to its presentation to planners as a potential management tool. To further test the model, and in preparation for model presentations in new social and geographic settings, we will sample water quality, eelgrass habitat, and fish communities in several estuaries of southern Rhode Island during summer 1999. The first model presentations will be conducted in early spring 1999. The model will soon be accessible as a user-friendly web site with an interactive interface allowing the input of locally-derived data on watershed use and the exploration of various scenarios involving historical land use or projected changes in watershed development. The model in its final form will provide an estimate of eelgrass habitat extent and integrity under pristine conditions, against which current conditions can be compared. In addition, the likelihood of anoxic events will be computed based on estimated system-wide, biological oxygen demand. Users will be able to view ‘their’ system in a variety of eutrophication contexts and scenarios.