AUTOMATED METHODS FOR ESTIMATING BASEFLOW AND GROUND WATER RECHARGE FROM STREAMFLOW RECORDS¹

ABSTRACT: To quantify and model the natural ground water recharge process, six sites located in the midwest and eastern United States where previous water balance observations had been made were compared to computerized techniques to estimate: (1) base flow and (2) ground water recharge. Results from an existing automated digital filter technique for separating baseflow from daily streamflow records were compared to baseflow estimates made in the six water balance studies. Previous validation of automated baseflow separation techniques consisted only of comparisons with manual techniques. In this study, the automated digital filter technique was found to compare well with measured field estimates yielding a monthly coefficient of determination of 0.86. The recharge algorithm developed in this study is an automated derivation of the Rorabaugh hydrograph recession curve displacement method that utilizes daily streamflow. Comparison of annual recharge from field water balance measurements to those computed with the automated recession curve displacement method had coefficients of determination of 0.76 and predictive efficiencies of 71 percent. Monthly estimates showed more variation and are not advocated for use with this method. These techniques appear to be fast, reproducible methods for estimating baseflow and annual recharge and should be useful in regional modeling efforts and as a quick check on mass balance techniques for shallow water table aquifers.