Water circulation in karst systems: comparing physicochemical and environmental isotopic data interpretation

Abstract

Many studies have used physicochemical response observed at karst springs during storm events to investigate the concept of diffuse flow in karst systems, but the complete description of the karst systems including the interactions of all components is not fully understood. Studies have shown that physicochemical response measured at karst springs can decrease or increase in response to storm events. It is clear from these studies that the use of only physicochemical response in interpreting water circulation in karst systems can be ambiguous. In this study, both physicochemical and isotope (stable and radiogenic) responses at City Spring, Cookeville, Tennessee, during snowmelt were used to interpret the hydrological cycle of a karst system. At the beginning of snow melting, all physicochemical parameters dropped, and thereafter, temperature, Ca, Mg, and HCO3 increased steadily with increased discharge and total dissolved solids. In contrast, generally Na, Cl, and SO4 remained constant after dropping; however, very high values were observed in the spring water mixed with stormwater runoff. δ2H, δ18O, and 87Sr/86Sr of spring water were stable and contrastingly different from the snow, rain, and spring water mixed with stormwater runoff samples. CO2 pressure was higher than atmospheric and appeared to be degassing. The data suggest that spring discharge is a mix of pre-event water stored in the soil, epikarst zone, and underlying bedrock. The circulation depth is unknown, but is likely controlled by the amount of storm event recharge that occurs through the piston effect. There is little evidence to support dilution by snowmelt. This study shows that combining the techniques and data can provide greater insight into water circulation and monitoring of karst aquifer systems than applying the techniques and data individually.