TRACKING ANNUAL, SEASONAL, AND STORM-TO-STORM CHANGES IN THREE PENNSYLVANIA KARST SPRINGS TO DETERMINE FLOW PATHS AND RECHARGE PATTERNS

Chemical and hydrological parameters were collected during storm events over four years at three karst springs in central Pennsylvania. Specific conductance (SpC), concentration of CO₂, concentration of NO₃^-, and depth measured over the rising and falling limbs of storm hydrographs during different times of the year and under different antecedent conditions point to varying speeds and flow paths through the karst systems.

CO₂ and NO₃^- concentrations increased from spring through summer and decreased during fall and winter, likely associated with plant growth and land use. CO₂ and NO₃^- increases occurred together at each spring, but timing of arrival was different storm-to-storm. Storm-driven changes in concentrations of chemical parameters were greatest in wet antecedent conditions and during the largest storms which indicated activation of fast flow paths under these conditions.

Weaver Spring had a discharge of 85-340 L/s, the highest [NO₃^-] and [CO₂], and the greatest differences in [NO₃^-] seasonally and during individual storms (22 to 30 ppm), which may be the result of significant agricultural land use within its catchment. During wet antecedent conditions, [CO₂], [NO₃^-], and SpC followed similar trends through each storm event at Weaver Spring. Relative timing of pulses in discharge and chemical parameters pointed toward multiple flow paths with significant recharge and storage of water and contaminants in soil. Smullton Sinks had the largest discharge (140-5,660 L/s). Parameter patterns were similar during individual storms, but storm-to-storm, peak [CO₂] arrival times were not the same (coincident with the peak for some storms and on the tail end of the hydrograph for others). This variation may indicate a changing contribution of soil water seasonally and storm-to-storm. Springhouse Spring had the smallest discharge (14 -85 L/s) and the least seasonal chemical variation, but [NO₃^-] varied from 6 to 13 ppm year to year. The smaller capture area likely played a role in less seasonal variation, and the chemographs for [CO₂] and [NO₃^-] corresponded to annual trends in discharge. This variation may indicate the dominance of slow flow under most conditions, while fast flow paths activate depending on the recharge pattern and wet versus dry conditions.