Advances in low streamflow statistics estimation using baseflow correlation

Low streamflow estimates are required for water quality and quantity management purposes. However, estimation of low streamflow statistics at ungauged river sites is a vexing problem. The baseflow correlation method is an information transfer technique that is used to estimate low flow statistics at an ungauged site by correlating a nominal number of measured streamflow discharges during baseflow conditions at the ungauged site with those at nearby-gauged sites. This dissertation focuses on advances in low streamflow statistics estimation using baseflow correlation. Jackknife experiments employing two high quality data sets are developed to investigate improvements in the baseflow correlation method.;The first jackknife experiment is employed to investigate the performance of baseflow correlation with a single gauged site. Numerous method and experimental parameters are examined. The second experiment is developed to investigate the performance of the six new estimators using multiple gauged site information. The new estimators are compared with baseflow correlation with a single gauged site. The United States Geology Survey (USGS) Hydro-Climatic Data Network (HCDN) data set is employed in both of the aforementioned experiments. This data set contains high quality streamflow data for nearly 1600 gauged sites. The third experiment is performed to compare baseflow correlation with regional regression for estimating low flow statistics. A subset of the U.S. Geological Survey (USGS) National Water Information System (NWIS) data set is employed in this experiment. This data set consists of 29 USGS gauged sites primarily within Tennessee and Kentucky.;The results confirm that Baseflow measurements should be obtained during low flow seasons and as far as possible from runoff events. The various method assumptions are generally reasonable. Using a log-space variance weighted average appears the best estimators among the six new multiple-site estimators. The method performance can be improved by using multiple site information, especially when less than 10 Baseflow measurements are used. Results also indicate baseflow correlation performs better than regional regression when enough Baseflow measurements are available at the ungauged site.