TRANSIENT WATER BUDGETS FROM STREAM HYDROGRAPHS

MURDOCH, LARRY, lmurdoc@clemson.edu and HODGES, R., rhodges@clemson.edu, Department of Geological Sciences, Clemson Univ., Clemson, SC 29634

Variations in stream discharge recorded on hydrographs are the response of a watershed to climatic and other stresses, and for more than 100 years hydrogeologists have sought to interpret hydrographs to gain insights into watershed characteristics. A common method of interpretation is to use an algorithm to identify times on the hydrograph when the stream is supplied solely by baseflow (baseflow turning points). A variety of algorithms for baseflow separation have been proposed, but methods for identifying turning points yield similar results when implemented using accepted principles. The key step follows baseflow separation and involves how the magnitude of recharge is inferred between the times of baseflow. Several methods have been proposed and the most popular ones can be traced to an analysis in the 1960s by Rorabaugh, who derived an equation that predicts the baseflow following an instantaneous change in head over an aquifer. Later investigators have developed graphical methods, and computer programs, for inferring recharge based on the assumption of instantaneous recharge. This general approach allows valuable information about the characteristics of a watershed to be extracted from a stream hydrograph, but it can be tedious to implement and it is unable to close a water balance without including extraneous sinks (e.g. riparian evapotranspiration).

We have revisited the single linear-reservoir analysis suggested by Rorabaugh and can show that simple manipulations allow the average recharge to be calculated directly from baseflow turning points. Moreover, the analysis predicts the baseflow at any time once the recharge has been determined. This allows the baseflow to be separated from the total flow on a daily basis. The amount of water stored in the aquifer follows directly from the baseflow, and including precipitation data in the analyses allows evapotranspiration to be calculated on a daily or monthly basis (depending on the time interval of the precipitation data). This produces daily records of total stream discharge, stormflow, baseflow, recharge, aquifer storage, and ET from the hydrograph. Importantly, the analysis is based on an internally consistent model so the water balance is always closed.

Applications for this analysis include the estimation of low-flow characteristics of a stream during drought, which is a common application for hydrograph analyses. We have developed an easy-to-use computer program that implements the hydrograph analysis using the widely available EXCEL software, which facilitates data handling and graphics and provides a ready mechanism for conducting auxiliary calculations. As a result, this program can be used to quickly estimate the basic water budget in a watershed from data available on the Internet. This type of water budget should be included in any environmental site assessment, although it is often omitted because the analyses have been tedious to make. Water balance data are critical input parameters to ground water flow models, and we expect that the results of this program will help during the initial calibration of such models. The analysis provides detailed records of the transient response of components of the water budget, providing key insights into the timing and magnitude of critical events that are important to watershed management. The method is based on sound principles and can readily be expanded to include a variety of more advanced processes.