Scientific Investigations Report 2008–5084
Abstract
The U.S. Geological Survey, in cooperation with the Texas Department of Transportation and in partnership with Texas Tech University, investigated a refinement of the regional regression method and developed alternative equations for estimation of peak-streamflow frequency for undeveloped watersheds in Texas. A common model for estimation of peak-streamflow frequency is based on the regional regression method. The current (2008) regional regression equations for 11 regions of Texas are based on log10 transformations of all regression variables (drainage area, main-channel slope, and watershed shape). Exclusive use of log10-transformation does not fully linearize the relations between the variables. As a result, some systematic bias remains in the current equations. The bias results in overestimation of peak streamflow for both the smallest and largest watersheds. The bias increases with increasing recurrence interval. The primary source of the bias is the discernible curvilinear relation in log10 space between peak streamflow and drainage area. Bias is demonstrated by selected residual plots with superimposed LOWESS trend lines. To address the bias, a statistical framework based on minimization of the PRESS statistic through power transformation of drainage area is described and implemented, and the resulting regression equations are reported. Compared to log10-exclusive equations, the equations derived from PRESS minimization have PRESS statistics and residual standard errors less than the log10 exclusive equations. Selected residual plots for the PRESS-minimized equations are presented to demonstrate that systematic bias in regional regression equations for peak-streamflow frequency estimation in Texas can be reduced. Because the overall error is similar to the error associated with previous equations and because the bias is reduced, the PRESS-minimized equations reported here provide alternative equations for peak-streamflow frequency estimation. |
Version 1.0 Posted August 2008 |
Asquith, W.H., and Thompson, D.B., 2008, Alternative regression equations for estimation of annual peak-streamflow frequency for undeveloped watersheds in Texas using PRESS minimization: U.S. Geological Survey Scientific Investigations Report 2008–5084, 40 p.
Abstract
Introduction
Purpose and Scope
Acknowledgments
Current (2008) Regional Regression Equations for Peak-Streamflow Frequency Estimation in Texas
Inconsistent Peak-Streamflow Frequency Curves by Regional Regression
Regional Regression Equation Applicability and Implementation
Biased Peak-Streamflow Frequency Values
Alternative Regression Equations for Estimation of Peak-Streamflow Frequency for Watersheds in Texas
Regression Equations Based on Logarithmic Transformation of Drainage Area
Regression Equations Basedon PRESS Minimization and Power Transformation of Drainage Area
Summary
References Cited
Appendix 1—Computational Script Using PRESS Minimization and Drainage Area, Mean Annual Precipitation, and Main-Channel Slope
Appendix 2—Computational Script Using PRESS Minimization and Drainage Area and Mean Annual Precipitation
Appendix 3—Computational Script Using PRESS Minimization and Drainage Area
For additional information contact: Director, Texas Water Science Center U.S. Geological Survey 8027 Exchange Drive Austin, Texas 78754-4733 World Wide Web: http://tx.usgs.gov/ |
Part or all of this report is presented in Portable Document Format (PDF); the latest version of Adobe Reader or similar software is required to view it. Download the latest version of Adobe Reader, free of charge. |