Relations are provided to estimate the magnitude and frequency of floods on Washington streams. Annual-peak-flow data from stream gaging stations on unregulated streams having 10 years or more of record were used to determine a log-Pearson Type III frequency curve for each station. Flood magnitudes having recurrence intervals of 2, 5, 10, 25, 50, and 100 years were then related to physical and climatic indices of the drainage basins by multiple-regression analysis using the Biomedical Computer Program BMDO2R. These regression relations are useful for estimating flood magnitudes of the specified recurrence intervals at ungaged or short-record sites.Separate sets of regression equations were defined for western and eastern parts of the State, and the State was further subdivided into 12 regions in which the annual floods exhibit similar flood characteristics. Peak flows are related most significantly in western Washington to drainage-area size and mean annual precipitation. In eastern Washington they are related most significantly to drainage-area size, mean annual precipitation, and percentage of forast cover. Standard errors of estimate of the estimating relations range from 25 to 129 percent, and the smallest errors are generally associated with the more humid regions.

No abstract