Multivariate regression models are often applied to hydrologic regions toestimate peak flows at ungauged basins in an area. Such regression models can bederived using a region of influence (RoI), which is a selected set of basins that arehydrologically similar to the ungauged basin for which peak flow estimates arerequired. These regions can be poorly defined resulting in unstable parameterestimates because observations at a few basins may disproportionately influence theparameters. Conventional treatment is to drop the basin, if the problem is evenrecognized. We propose a leverage-guided RoI regression approach that redefinesthe region of influence. This new procedure uses two newly defined RoI leverage andinfluence metrics. The proposed approach is applied to 996 streamflow gaugingstations in the southeast United States to estimate the 50-year peak flow. The newleverage-guided RoI regression approach resulted in lower root-mean-squareestimation errors, produced fewer observations with large leverage, and eliminated allinfluential observations.
展开▼
