Flood frequencies and sedimentation from forest watersheds

Abstract

A multi‐watershed, multi‐storm, multi‐variable approach was used to develop a hydrologic basis for evaluating flood control in Southern California. Flood peak discharge was shown to be related to six variables: watershed area, two storm precipitation variables, a temperature index of snow, a watershed wetness variable, and forest cover density. Since peak discharges have been measured under widely varying forest cover conditions, the partial regression coefficient for forest cover density was used to convert measured discharges to estimated discharges for a single comparable cover condition (a 40‐year‐old cover). Then, the estimated discharges were used in frequency analyses. The relation of discharges to storm precipitation and the wetness variables provided a means of, extending the short‐time discharge record to a long‐time record on the basis of a 69‐year record of precipitation. The relation of sedimentation of reservoirs to maximum yearly peak discharge, area of main stream channel, and forest cover density was determined. Together with the discharge frequency this provided a means of determining sedimentation frequency and mean annual sedimentation. From the determinations of the forest cover effectiveness, the effect of forest fires on peak discharges and sedimentation were evaluated for 41 watersheds.