The vegetative resistance to flow and the effects of sheltering from wind of different plant communities in the Florida Everglades combine with land-surface elevational differences and evapotranspiration losses to control the velocity, flow direction, water depth, and hydroperiod in Taylor and Shark River Sloughs. The major objectives of the work reported here were (1) to provide detailed information on species composition, vegetative characteristics, vegetative structure, and biomass for quantification of vegetative resistance to flow and the effects of sheltering from wind, and (2) to extrapolate this information to classify the vegetation regionally and to improve existing vegetation maps for use in surface-water models.

Water management decisions derived from predictive models are critical to successful restoration of the Everglades. Without accurate and precise accounting for vegetative characteristics at scales commensurate with model discretization, flow can not be simulated accurately. Two surface-water models are being developed for Taylor Slough; one covers most of the slough, including the mangrove/Florida Bay interface, and one covers only the upper part of Taylor Slough below the park road. In order to model the surface-water flow, it is necessary to extrapolate point measurements of velocity and surface-water slope made concurrently with characterization of vegetation to the entire model area. Vegetative resistance to flow can be expressed by either Manningís n or the Darcy-Weisbach friction factor, but these two coefficients must be related to the actual field characteristics of the vegetation through which flow occurs to provide the basis for accurate predictions of flow. Several steps and numerous experiments were necessary to develop and refine measurement techniques to quantify this frictional resistance/vegetative cover relation. Measurements of critical variables were conducted in a hydraulic flume as well as in the Shark River and Taylor Sloughs to define the effects of frictional resistance on regulation of flow.

Flume experiments: The initial measurements of resistance coefficients were derived from carefully controlled flow measurements. Uniform dense stands of sawgrass were grown in the 200- foot long, 6-ft wide USGS tilting flume at Stennis Space Center, Mississippi (Lee and Carter, 1996). In several series of experiments conducted at various flow depths, vegetative resistance was calculated from velocity, flow depth, and water-surface slope. During each experimental series, the vegetation in the flume was sampled to determine, as a function of distance from the bed or the sediment/water interface, the biomass per unit area, the number of stems and leaves per unit area, leaf and stem width, and leaf area index (LAI). Sponges were added in one series of experiments in order to simulate periphyton. Following the measurements of vegetative resistance, the flume was refitted to measure wind effects on flow when sawgrass was present.

Shark River Slough: Because the transfer of laboratory results to the field is a critical part of this research, two Shark River Slough sites were selected to provide sawgrass communities of varying densities in which to make water-velocity and surface-slope measurements. A total of forty-two 0.5- m-square vegetation quadrats were sampled in April and November 1996. Vegetation including periphyton was harvested in horizontal layers, 10-cm or 20-cm thick, from the soil/water interface through the water column to the top of the plants. Species composition, density, size, dry biomass, and LAI were determined for each layer. Quadrats were sorted into density-related vegetation classes based on species composition and biomass (excluding periphyton). Composite profiles (layer by layer means) were determined for each class. The 42 quadrats fell into nine different classes; rush and mixed sawgrass/rush classes were generally sparse or medium density, whereas sawgrass fell into four density classes, and cattail was very dense.

Taylor Slough: For modeling efforts in Taylor Slough, vegetation was sampled on three east-west transects across the slough. Sampling and processing were similar to, but slightly more detailed than, that in the Shark River Slough, and the quadrats were grouped into the same classes as in Shark River Slough. Composite layer-by-layer profiles were determined for each class. Two factors complicated the relation of biomass to velocity, dead plant material and periphyton. Large amounts of dead material or litter were mixed with the plants and, as generally calculated, LAI estimated only the live or total standing surface area opposing horizontal flow. We are developing a method for incorporating the dead litter, which is nonuniform in size and shape, in the LAI calculation. Periphyton was present in horizontal layers or vertical sleeves surrounding individual leaves. Although periphyton was spatially and temporally unpredictable and generally associated with sparse to medium vegetation classes, it also was present in denser classes. It could be accounted for in the total biomass figures for individual quadrats or mean class composites, but complicated calculation of a class composite biomass by layer and could not be included in a LAI.

A geographic information system was developed to help assimilate and interpret available spatial data such as digital orthophotoquads, digital line graphs, and a 1993-94 Landsat vegetation classification map of southern Florida developed by the USGS and the University of Florida. Working with the 32-class Landsat image, color infrared aerial photographs, and other available maps, we recombined map classes to show the areal extent of vegetation types considered to have different roughness characteristics and thus different effects on flow velocity. Cross checking these map classes with actual ground vegetation and characteristics determined through sampling, we developed a final map product for use with surface-water models in Taylor Slough. With the more accurate flow simulations that these models can provide, water managers can more confidently and precisely control and manipulate water levels, flows, and hydroperiods.

REFERENCE

Lee, J.K., and Carter, Virginia, 1996, Vegetation affects water movement in the Florida Everglades: U.S. Geological Survey Fact Sheet FS-147-96.

(This abstract was taken from the Proceedings of the South Florida Restoration Science Forum Open File Report)