Tuesday, June 5, 2007
490

Using GIS-based Modeling of Habitat in Ungauaged River Systems: Watershed Topographic Resolution Affects Hydrograph and Habitat availability Independent of Watershed and In-stream Characteristics

Andrew F. Casper, Barnali Dixon, and James A. Gore. Environmental Science & Policy Program, University of South Florida St. Petersburg, 140 6th Ave. South, DAV 258, St. Petersburg, FL 33701

A wide variety of hydrologic models use discharge to estimate changes availability critical in-stream habitat. However a large proportion of rivers and watersheds are ungauged. One widely promoted solution for these situations uses GIS-based models of watershed topography and runoff to estimate the hydrograph. However the elevation/topographic grid resolution supplied by the most commonly available software has been in flux since the advent of the technology. Thus a largely untested assumption is that grid resolution used to model watershed runoff does not influence hydrograph. We have tested this assumption with by linking the Soil Water Assessment Tool (SWAT) model to Physical Habitat Simulation Model (PHABSIM) for in-stream habitat. Both models were calibrated for a low-gradient SE coastal system, the Hillsborough River Watershed (FL). There is a large discrepancy between modeled and measured hydrograph.  We show that the coarser the topographic grid, the more artificially smooth the watershed is which leads to unrealistically elevated discharge. Subsequently this causes a PHABSIM overestimation of the seasonal effect on habitat. Thus while GIS-based modeling of ungauged in-stream habitats on the desktop is possible, but the efficacy will depend more on the raw topographic GIS data sets than on the mathematical and computer model used.