Estimation of base flow on ungaged, periodically measured streams in small watersheds in western Pennsylvania

Abstract

A 2.5-year data collection program was undertaken by the U.S. Geological Survey, in cooperation with the Pennsylvania Department of Environmental Protection (PADEP), to quantify and estimate base flow in small watersheds in western Pennsylvania where only periodic streamflow measurements had been obtained. Twelve streamgages with watershed areas of less than 10 square miles were established in western Pennsylvania for this study, with most established within Greene and Washington Counties (an area where a type of underground coal mining known as longwall mining occurs). Data from five previously established streamgages with watershed areas ranging from 48.9 to 281 square miles were also used in the analyses for this study. The index-gage method was used to relate streamflow at one streamgage referred to as the “index streamgage” to streamflow at another site of interest (usually an ungaged site, but for this study another streamgage) using a regression technique.Streamflow regressions were developed for all newly established streamgages by using the Maintenance of Variance Extension, Type 1 (MOVE.1) method. Not all streamflow data from the newly established streamgages were used for MOVE.1 regression development; only data that have little to no influence from runoff were considered. Runoff-influenced streamflow for this study was defined as streamflow on a day that precipitation occurs plus streamflow on the following 2 days. One streamflow value per day selected from a specified schedule that captures numerous non-runoff periods was used to develop a MOVE.1 regression.Prediction limits were calculated from the regression to provide the upper and lower bounds for the regression-produced streamflow estimates. Using these data, base flow at a site can be estimated with the index-gage method. The log10-tranformed prediction interval width and other regression diagnostics were used as indicators of regression quality when comparing streamgage relations to determine the best index streamgage among the streamgages established for this study. It was determined that index streamgages within about 10 miles of the site of interest provided the best estimated base flow and could, in the future, be used by mine operators and the PADEP to quantify base flow and to evaluate the effects of mining on streamflow.