Multiple tracers reveal different groundwater recharge mechanisms in deep loess deposits

Abstract

Interpretation of groundwater recharge mechanisms is problematic because of the muted instantaneous response of subsurface water to rainfall and limited recharge rates, particularly in semi-arid environments with deep loess deposits. Here we identify the possible groundwater recharge mechanisms in 200-m thick loess deposits with unsaturated zone thickness of over 40 m. We collected soil samples up to 15 m deep under four land use types (one grassland and three apple orchards with stand ages 15, 24 and 30 years old), and used three-year precipitation and groundwater samples to determine the contents of stable water isotopes, chloride, and tritium. Our overarching goal is to determine the relative importance of piston and preferential flow in groundwater recharge using multiple tracers and quantify the effects of land use change on groundwater recharge. We find that while both piston and preferential flows are important in groundwater recharge, the unsaturated and saturated zones have yet to come to hydraulic equilibrium. This suggests different groundwater recharge mechanisms: tracers in the unsaturated zone suggest piston flow, while the detectable tritium in the saturated zone implies preferential flow. Recharge rates in the unsaturated zones range between 23 and 82 mm year−1, accounting for 4%–14% of mean annual precipitation, and increasing with depth presumably because of land use and/or climatic conditions. Total recharge rate in the saturated zone is 112.6 ± 44.1 mm year−1, accounting for 19 ± 9% of mean annual precipitation. Overall, our study finds that piston flow contributes more to total recharge (53%–69%) than does preferential flow. Nevertheless, piston flow may become less important because of land use change (farmland to apple orchard conversion). Our findings have implications for the need to strike a delicate balance between the economic gains from afforestation and the possible risks to groundwater supply sustainability.