Files
Abstract
This thesis presents a comprehensive model for predicting storm runoff andcontributing areas based on precipitation and topography. Independent storms are definedusing a procedure by Restrepo-Posada and Eagleson (1982) and stormflow is defined withgeochemistry. Recursive digital filters are compared and calibrated to geochemicalhydrograph separation on Panola Mountain Research Watershed enabling the digital filtersto be used to define stormflow on Coweeta Watershed 18. High resolution topography ofthe watersheds, derived from LiDAR, is used to extract dense drainage networks. Thenetworks predict surface runoff flow paths and the areas of a watershed that producerunoff. The mean lengths of streams of the networks are used to parameterize thegeomorphologic instantaneous unit hydrograph that is used to model storm hydrographs.Finally, modeled storm hydrographs are compared to observed and digitally filteredstormflow. Comparison of observed and modeled storms determines if this model issuitable for ungauged basins.