A STATISTICAL STUDY OF THE HYDROLOGICAL CHARACTER OF A KARST AQUIFER IN RESPONSE TO A MAJOR PRECIPITATION EVENT

The Edwards Aquifer of south-central Texas is one of the most permeable and productive artesian limestone aquifers in the United States. It is a sole-source aquifer that supplies water to more than 1.5 million people in San Antonio and the Austin-San Antonio corridor. Severe flooding occurred in portions of south-central Texas between October 17 and 19, 1998. At least 30 in. of rainfall was documented in southern Hays County. The USGS indicated that about 2,300 mi2 in 12 counties received at least 12 in. of rain. Response to the October 1998 rainfall included flooding of major rivers and streams, increases in water levels in monitor wells, and increased flows from discharge springs, including such major discharge locations as San Marcos and Comal Springs. Historical flood peaks were recorded by 27 streamflow-gaging stations; one gage on the Guadalupe River measured a peak flow that was 2.6 times the previously recorded high flow. A statistical study of the hydrological data (river flows, spring flows, and water levels in monitor wells) associated with the storm event indicated the following: the response of river and stream flows was very rapid with pronounced peaks in flow and rapid recession curves, typical of a karst environment; responses of flows from discharge springs were rapid, indicating a strong connection with surface water; a more rapid rise to peak flows in San Marcos Springs relative to Comal Springs may correspond with a smaller watershed or a higher transmissivity aquifer; the time needed for flows from discharge springs to return to their pre-event values were long and depended on the size of the watershed and the aquifer transmissivity; periods of nearly constant discharge from springs indicated the presence of substantial storage in the Edwards Aquifer at a depth less than that achieved for peak discharges; cross-correlation between flows in San Marcos and Comal Springs was high for a zero-day lag time, consistent with a conceptual model in which there is strong communication with surface waters; and water levels in wells completed in similar recharge and fresh-water zones showed high cross-correlation.