8
Marine Module
The Marine Module consists of a dispersal model that predicts near-shore concentrations of nutrients, metals, and pollutants. It uses nutrient and pollutant loadings generated by the Integrated Water loads. Components of the model include patterns of circulation, water depth and loading.
From the beginning, the NRC committee hoped that the Marine Module would be the most developed and integrated component of the entire CCAM. This feeling was based, in part, on comments expressed by both URS Corporation, Inc., staff and professional biologists during the 2-day workshop in Key Largo in January 2001. When it received the Draft CCAM in November 2001, the Committee was therefore surprised that the only input to the marine ecosystem came from the Integrated Water Module and that the Smart Growth scenario showed no measurable impact on the marine ecosystem. It is inappropriate to make such an inference using the current version of the Marine Module, as will be explained below.
The contractors examined considerable literature (both peer-reviewed and gray literature) in their search for quantitative data on the impacts of population increases on marine ecosystems. Their presentation at the National Research Council meeting in January 2001 in Key Largo, Florida, was compelling, and the NRC Committee felt at that time that they were well on the way to establishing this relationship for impacts on seagrass, fishes, and coral-reef communities (National Research Council, 2001). As they could not find a direct and appropriate data set in the literature on these impacts relative to population growth, however, URS Corporation, Inc., did not consider other approaches. This is particularly disturbing because significant direct impacts of land use and population growth
have occurred in the Florida Keys, as pointed out by both the URS staff in the Terrestrial Module and by professional biologists.
MAJOR CONCERNS
The lack of data on seagrass loss from propeller scars and on the impact on fish species might still have been overcome if URS Corp. had used the quantitative data on these issues (Sogard et al., 1987; Koenig and Coleman, 1998; Thayer et al., 1999) in an innovative manner. It appears, however, that the contractors did not attempt this approach or the use of other data sets, such as creel surveys in state and national parks to assess fishing effort relative to population increases. This approach is particularly important for day or weekend visitors who are not included in land use pattern estimates.
Historical changes in marine resources were not considered in the Marine Module as they were in the Terrestrial Module. Current resource conditions could be set within a historical framework using historical maps of marine resource distribution and (GIS) technology.
The Marine Module also lacks data on boat pollution loadings, pathogens, and fecal coliforms. In addition, canal water quality was not considered because of the scale of the diffusion model used despite the fact that canal water quality is an important issue for near-shore environments and is a major public concern.
The background levels of metal concentrations in surface seawater are based on out-dated standards that should be revised (Draft CCAM Appendix D). Using the new standards would markedly increase the incremental calculations computed for stormwater and wastewater impacts, but would lower total concentrations below levels of concern. More current values (Pilson, 1998) include Cd = 0.01 µg/L, Cu = 0.25 µg/L, Pb = 0.002 µg/L, and Zn = 0.4 µg/L.6
The final CCAM report should provide the user with a more complete view of what was actually considered relative to establishing the relationships needed for input into the Marine Module. The approaches used need to be clearly stated.
Given the lack of inputs into the Marine Module, it is not possible to draw any meaningful conclusions on the impact or lack of impact of current or future development activities (see Overarching Comments). The findings from the Smart Growth scenario reported in the Draft CCAM assume complete fulfillment of all water module and terrestrial management plans outlined in the document. If those actions are not completed, the Smart Growth scenario findings are invalid, since they would be based on assumptions and conditions that are not met. Finally, the URS Corp. should address the important issue of the lag time between development and subsequent measurable impacts on marine resources. In order
to track ecosystem management, the module needs to consider the length of time needed for specific marine resources to “rebound” from the current conditions.
Diffusion Model
The diffusion model used as the basis of the Marine Module is based on the solution for a vertical line source (Fischer et al., 1979). As indicated in the Draft CCAM report, the line-source solution requires that the velocity, u, be perpendicular to the shoreline in order to ensure an “infinite” zone for lateral dispersion, (i.e., a zone with no lateral boundaries). On the basis of Figure 4.7 in the Draft CCAM, it appears that this is how the mode is used. It is not clear, however, as to how the GIS layer is interrogated to obtain this velocity. If a non-perpendicular velocity vector is used, the line-source equation cannot be applied as indicated because the vector violates the assumption of an infinite lateral extent. (Image sources are a possible solution to this problem.) By extension and more seriously, there cannot be a solely offshore velocity; a return flow must exist that would bring offshore water back onshore between the plumes (analogous to the onshore flow near a rip tide). The line-source solution method discussed in the Draft CCAM report does not account for such likelihood. Finally, the line-source solution assumes a constant depth, when in fact, the depth increases offshore reducing the velocity, u, and increasing the vertical mixing zone. These effects are not accounted for in the solution methodology.
The appealing simplicity of the line-source solution (in lieu of a complex numerical model) therefore prevents all but questionable applications of the module to areas in which velocities are not entirely offshore or in which depths vary. The type of condition for which the line-source solution is suitable is a steady-state discharge from a source of constant concentration (e.g., the discharge of a waste treatment plant through an ocean outfall into shallow water of constant depth).
Wastewater discharges for the Florida Keys are roughly constant in time but emerge, as the Draft CCAM report points out, from shallow aquifers in a continuous fashion along the shoreline, not as a point source per watershed. The loading is assumed to include stormwater discharges, but in reality these are episodic and likely distributed along the length of the shoreline rather than concentrated at one location per wasteshed. In fact, the current line-source model may well overpredict concentrations by ignoring the distributed nature of the loadings. The time scale for mixing in the near-shore zone between storms may result in a relatively uniform distribution of discharged constituents in that zone, but there is no way of determining such a distribution from the line-source mixing model. A two-dimensional, steady-state model with provisions for exchange with offshore waters might suffice to distribute conservative constituents discharged continuously, on the average. A truer representation of marine water quality awaits development of a transient, 2-D model, perhaps using the GIS system to interpolate
and extract needed velocity vectors for an x-y grid in the shallow, near-shore water. The current line-source solution, interrogated at 100-m increments and summed over adjacent “plumes,” is inadequate for inferring marine water quality.
In short, the Marine Module is not an adequate tool for generating predictions on marine water quality or other important environmental endpoints given the limitations of its diffusion model. The Draft CCAM Marine Module should not be used to make inferences about marine water quality.