An artificial recharge experiment in the San Jacinto basin, Riverside, southern California

Abstract

During a 3 month recharge experiment related to conjunctive use of water resources, 1.5 × 10 6 T of imported water were percolated through a pond of 128 m by 128 m in the San Jacinto basin. The infiltration rates, which declined with time, averaged 1.9 m day −1, equivalent to four times the lowest laboratory-measured hydraulic conductivity of the fluvial deposits. Ponding altered the unimodal grain-size distribution at the ground surface to types without a dominant mode, but this redistribution did not always lead to reduction in conductivities, which varied over at least three orders of magnitude. The water table 80 m downstream from the ponding edge began to rise slowly 1 month after the start of ponding; it leveled off at 8 m above the pre-recharge water table depth of 75 m and did not recede 2 months after termination of ponding. Water levels in wells bottomed in the original vadose zone suggested that an inverted water table migrated downward to meet the rising water table. Minor, local perching occurred at 14 m depth, as indicated by the presence of moist ground near one monitoring well and by hydraulic responses during a 20 day intermission in percolation. As it percolated through the sediments, the imported northern California water gained Ca but lost Mg, so that the Mg/Ca ratio resembled that of local ground water. Such cation exchange has also been demonstrated by leaching experiments in the laboratory. However, the characteristics of the original source waters appear to be retained by D/H isotope ratios and Cl concentrations, as well as cross-plots of SO 4 vs. Cl and B vs. Cl. Such unreactive tracers could serve to monitor transport and mixing of the chemically diverse water used in future recharge programs in the San Jacinto basin.