ESTIMATING THE MAGNITUDE AND FREQUENCY OF FLOODS IN URBAN BASINS IN MISSOURI

Streamgage frequency analyses were done on 35 streamgages on urban streams in and adjacent to Missouri. Results of ordinary least-squares regression analyses were evaluated on the basis of Mallow’s Cp statistic, the adjusted coefficient of determination, and the statistical significance of the independent variables. The independent variables of drainage area and impervious area were found to be statistically significant and readily determined from existing digital data sets. The impervious area variable was computed from the National Land Cover Dataset 2001 impervious area dataset. The National Land Cover Dataset 2001 impervious area data for each basin was compared to historical imagery and 7.5-minute topographic maps to verify the national dataset represented the urbanization of the basin at the time streamgage data were collected. Weighted least-squares regression techniques were used to determine the final regression equations for the statewide urban flood frequency equations. The final flood frequency equations for the 50-, 20-, 10-, 4-, 2-, 1-, and 0.2- percent annual exceedance probability floods for Missouri provide a technique for estimating peak flows on urban streams at gaged and ungaged sites. The range in drainage area is 0.28 to 189 square miles and the range in impervious area is 2.3 to 46.0 percent. Six of the 35 selected streamgages were used to compare the results of the existing rural and urban equations to the urban equation presented in this report for the 1 percent annual exceedance probability. Results of the comparison indicate that the estimated peak flows for the urban equation in this report were higher than the results from the rural equations. A comparison of urban equations indicates the estimated urban peak flows from this report generally result in lower frequency values than equations previously published in 1986.