Skip to main content
U.S. flag

An official website of the United States government

Results of generalized least-squares (GLS) regression for 136 USGS streamgages in urban areas in Tennessee and parts of Alabama, Georgia, Mississippi, North Carolina, and South Carolina

October 7, 2024
This dataset contains results of generalized least-squares (GLS) regression for 136 streamflow gaging stations (streamgages) operated by the U.S. Geological Survey (USGS) on urban streams in Tennessee and parts of Alabama, Georgia, Mississippi, North Carolina, and South Carolina. GLS regression was used to relate drainage area, percentages of the drainage basins of the streamgages in developed land use (the sum of classes 21-24 from the 2011 NLCD; Homer and others, 2015), and the percentages of the streamgage basins in the Piedmont and Ridge and Valley level 3 (L3) ecoregions (U.S. Environmental Protection Agency, 2010) to previously published estimates of streamflows corresponding to the 0.5, 0.2, 0.1, 0.04, 0.02, 0.01, 0.005, and 0.002 annual exceedance probabilities (AEPs; Wagner and Ladd, 2024). From this regression, equations were developed that can be used to predict streamflows corresponding to the eight selected AEPs at ungaged locations on urban streams (percent developed imperviousness greater than 10%; Homer and others, 2015) in the study area. Standard errors of prediction ranged from 35.8 percent for the 0.2 AEP (the 5-year flood) to 55.4 percent for the 0.002 AEP (the 500-year flood). GLS regression was conducted using version 3.0 of USGS weighted regression (WREG) software (Farmer, 2019; Eng and others, 2009); input and output files from WREG are provided. USGS site information, values of the five basin characteristics, at-site estimates of the selected AEPs and their variances (from Wagner and Ladd, 2024), GLS estimates of the selected AEPs and their variances, weighted estimates of the selected AEPs and their variances and associated 95-percent confidence limits (see Appendix 9, "Weighting of Independent Estimates", in England and others, 2019), and the residual, leverage, and influence values from GLS regression are provided in CSV format ("GLSresults.csv"). A table of the final GLS regression equations for each AEP and their associated performance metrics is also provided in CSV format ("RegressionEquations.csv").

The following input files for USGS WREG software are located in the "WREG_input" directory within the "WREG_input.zip" folder: a text file of basin characteristics for all 136 selected streamgages in the study area ("SiteInfo.txt") and a PeakFQ output (*.PRT) and export (*.EXP) file for each streamgage, located in the "PeakFQ" subdirectory. The following eight WREG output files, located in the "GLS_results" folder within the "GLS_results.zip" directory, are in sub-directories for each of the 8 selected AEPs: "coef.txt," "CoVarMat.txt," "FitandResid.txt," "modelInput.txt," "outputRaw.rda," "PerformanceMetrics.txt," "ResLevInf.txt," and "Weighting.txt."
Publication Year 2024
Title Results of generalized least-squares (GLS) regression for 136 USGS streamgages in urban areas in Tennessee and parts of Alabama, Georgia, Mississippi, North Carolina, and South Carolina
DOI 10.5066/P1TE9KTH
Authors Daniel M Wagner, David E Ladd
Product Type Data Release
Record Source USGS Asset Identifier Service (AIS)
USGS Organization Lower Mississippi-Gulf Water Science Center - Nashville, TN Office
Rights This work is marked with CC0 1.0 Universal
Was this page helpful?