Improved Methods of Parameterization for Estimating the Magnitude and Frequency of Peak Discharges in Rural Ungaged Streams

Abstract

For the past several decades, the United States Geological Survey (USGS) and others have been tasked with development of regional predictor equations to determine flood peaks in unregulated and ungaged rural locations. This study tests several new watershed parameters not commonly used in past studies that have been extracted from available geomorphologic and climatic data using both manual and programmatic methods with the goal of improving the predictive power of the regional equations. Parameters include a derived basin slope corresponding to surface flow direction, discrete travel time for each location in a watershed to the outlet, and development of a watershed subarea correlated with the location of the maximum length isochrone line within each watershed. A case study is presented that tests these parameters in the Valley and Ridge physiographic region of Virginia. Results indicate that the proposed parameters tend to reduce standard model error in predictor equations ranging between 0.97 and 18.08% when compared with equations developed using drainage area as the lone predictor variable.