Estimating design discharges for drainage structures in western Kansas.

PDFTech ReportK-TRAN: KU-12-4C1912Drainage structuresDesign methodsRunoffFloodsWatershedsRational formula (Hydraulics)Regression analysisKansasKansas. Dept. of Transportation. Bureau of Materials & ResearchMcEnroe, Bruce M.Young, C. BryanWilliams, Aaron R.Hinshaw, MattUniversity of Kansas. Dept. of Civil, Environmental and Architectural EngineeringKansas State University. Transportation CenterUnited States. Dept. of Transportation. Research and Innovative Technology AdministrationNTL-HIGHWAY/ROAD TRANSPORTATION-Bridges and StructuresNTL-HIGHWAY/ROAD TRANSPORTATION-DesignUS Transportation CollectionKDOT engineers have expressed concern that the hydrologic methods in the current KDOT Design Manual (Volume I,Part C, 2011) may lead to over-sizing of drainage structures in Western Kansas. Some new structures designed by the currentmethods are much larger than the previous structures at these locations or existing structures directly upstream or downstream,where the older structures had no known history of overtopping. There are reasons to suspect that current methods may not be wellsuited to small watersheds in Western Kansas, particularly for areas with high soil permeability.This report examines the applicability of KDOT\ue2\u20ac\u2122s current hydrologic methods to Western Kansas and develops newRational C values and flood-frequency regression equations for this region. In addition, KDOT\ue2\u20ac\u2122s current hydrologic methods arecompared with those of nearby state DOTs.In order to develop new flood-frequency regression equations and recommendations for Rational C values for WesternKansas, we assembled a data set of all USGS gaging stations that met the following criteria: (1) at least 10 years of peak flowrecords, (2) watershed area less than 100 mi2, (3) unregulated watersheds (no major lakes or reservoirs), and (4) watersheds within100 miles of the Kansas border and west of 97.5\uc2\ub0 longitude. The resulting data set contains 156 stations, 62 of which are inKansas.Regional flood frequency analyses were performed on this data set using Generalize Least Squares regression in WREG1.0. Soil permeability was found not to be a significant predictor variable. Regression equations were developed for WesternKansas, but our comparisons show that these equations are not a substantial improvement over existing regression equations.Based on an evaluation of available methods, we recommend the Extended Rational method for watershed areas > 640 ac and < 30mi2 and the USGS four-parameter regression equation for watersheds \ue2\u2030\ua5 30 mi2 in both Western and Eastern Kansas.An analysis of Rational C values indicates that C values currently used for design in Western Kansas are too high forrecurrence intervals below 100 years. New Rational C values for Western Kansas were developed and checked against regressionmethods for consistency. Our proposed C values for Western Kansas are lower than the current values for all recurrence intervalsbelow 100 years. We also propose certain adjustments to the Rational C values for Eastern Kansas. We recommend that urban openspaces and pervious surfaces within the right-of-way be considered equivalent to pasture/range rather than cropland in bothWestern and Eastern Kansas