Water Resources-Investigations Report 00-4080
By Jeffrey B. Wiley, John T. Atkins, Jr., and Gary D. Tasker
The full report is available in pdf. Links to the pdf.
Multiple and simple least-squares regression models for the log10-transformed 100-year discharge with independent variables describing the basin characteristics (log10-transformed and untransformed) for 267 streamflow-gaging stations were evaluated, and the regression residuals were plotted as areal distributions that defined three regions of the State, designated East, North, and South. Exploratory data analysis procedures identified 31 gaging stations at which discharges are different than would be expected for West Virginia. Regional equations for the 2-, 5-, 10-, 25-, 50-, 100-, 200-, and 500-year peak discharges were determined by generalized least-squares regression using data from 236 gaging stations. Log10-transformed drainage area was the most significant independent variable for all regions.
Equations developed in this study are applicable only to rural, unregulated, streams within the boundaries of West Virginia. The accuracy of estimating equations is quantified by measuring the average prediction error (from 27.7 to 44.7 percent) and equivalent years of record (from 1.6 to 20.0 years).
Abstract
Introduction
Description of study area
Physiographic provinces
Climate
Regional historical floods
Previous studies
Development of estimating equations
Peak discharge data
Basin characteristics data
Magnitude and frequency analysis
Data correlation
Regional regression analysis
Estimating procedure
Accuracy of flood-estimating equations
Limitations of flood-estimating equations
Summary
References cited
Appendix 1: Accuracy of estimating equations
The text and graphics are presented here in pdf format (print quality):
To view and print report you need to use Adobe Acrobat Reader, available as freeware.
| AccessibilityFOIAPrivacyPolicies and Notices | |
  |
|