Estimating the magnitude and frequency of floods at ungaged locations on streams in Tennessee through the 2013 water year

To improve estimates of the frequency of annual peak flows for ungaged locations on non-urban, unregulated streams in Tennessee, generalized least-squares multiple linear-regression techniques were used to relate annual peak flows from streamgages operated by the U.S. Geological Survey to physical, climatic, and land-use characteristics of their drainage basins. Geospatial data acquired since the previous study in 2003, annual peak-streamflow data through the 2013 water year, and Bulletin 17C methods for frequency analysis of annual peak-streamflow data were used in the study. Generalized least-squares regression equations were developed for four hydrologic areas with distinct hydrologic, geologic, and topographic characteristics. Drainage area was used as an explanatory variable in equations developed for each hydrologic area. In addition to drainage area, a 2-year recurrence-interval climate factor was used for hydrologic area 1, a 10–85 channel slope was used for hydrologic area 2, and percent imperviousness was used for hydrologic area 4. The regression equations can be used to estimate annual exceedance probability streamflows for ungaged locations on non-urban, unregulated streams in Tennessee. The term “unregulated” indicates that streamflow is not appreciably influenced by regulation from reservoirs or other impoundments. Average standard errors of prediction for the regression equations ranged from 44.4 to 51.4 percent for hydrologic area 1; 30.4 to 42.9 percent for hydrologic area 2; 35.7 to 42.6 percent for hydrologic area 3; and 32.5 to 47.4 percent for hydrologic area 4.