Use of discharge variability, aquifer memory effect, and recession curve analysis in several springs to characterize karst aquifers in the tropical area

This research was conducted on three adjacent karst springs in one of the tropical karst areas in Indonesia. This study aims to apply a simple approach using parameters of discharge variability, aquifer memory effect, and recession curve analysis to characterize the karst aquifer in the three springs. The discharge hydrograph during the study period (January 2018–April 2020) was obtained from recording water level data using HOBO U20L-02, which was converted to discharge data after several direct discharge measurements at the three springs. Furthermore, discharge variability is determined statistically by calculating the spring coefficient of variation parameter (SCVP) and flow duration curve (FDC), while the aquifer memory effect is approximated by autocorrelation calculation. Meanwhile, the degree of void development was determined by reconstructing the master recession curve (MRC) to determine the presence of laminar and turbulent flow sub-regimes. The results of the analysis show that applying these three parameters can distinguish the characteristics of the aquifer in the three springs. Kiskendo spring which offers the highest void development (Dk = 8), also has a high value of discharge variability with a short aquifer memory effect with the dominance of diffuse-conduit flow. On the other hand, the Mudal spring with the lowest Dk value has the most stable discharge variability and the longest aquifer memory effect with the dominance of diffuse-fissure flow. However, because the calculation of discharge variability and aquifer memory effect is based on statistical data, which may not always reflect the real physical function of the system in karst aquifers, it is necessary to confirm with other approaches to reduce bias in conclusions.