Estimation of 7‐day, 10‐year low‐streamflow statistics using baseflow correlation

Abstract

Estimators of low‐streamflow statistics are often required for water quality and quantity management at gauged and ungauged sites. Here the technique of baseflow correlation is investigated as an estimator of the 7‐day, 10‐year low‐streamflow statistic at ungauged sites. This method uses an information transfer technique to estimate streamflow statistics at an ungauged site by correlating a nominal number of measured streamflow discharges during baseflow conditions at the ungauged site with those at a nearby gauged site. A regional assessment of baseflow correlation estimators is made by employing daily streamflow values at more than 1300 USGS HCDN stream gauge sites. A jackknife simulation is performed in each of the 18 USGS water resource regions located within the conterminous United States. Potential gauged sites are selected using a variety of watershed, geographic, topographic, and geologic classification systems. The results of this study indicate that baseflow correlation performs well in the United States when baseflow measurements are nearly independent and potential gauged sites are located within 200 km. The method is improved as the number of baseflow measurements is increased, although some leveling off of performance was observed with more than 15 baseflow measurements. A comparison of baseflow correlation with regional regression shows that baseflow correlation is the preferred method for estimating low flows in much of the United States.