Spatio-temporal variations of water sources and mixing in a riparian zone

Riparian zones are known for their role in regulating water quality in stream corridors. Specifically, riparian zones can act as buffers for high-concentration nutrient inputs into the stream, which can be harmful for the aquatic ecosystem. This natural attenuation capacity is controlled by variable water and solute exchanges bringing together different reactants via the mixing of stream water (SW) and local groundwater (GW). The degree and the extent of this mixing can regulate the potential for turnover processes for certain solutes. Here, we couple a previously calibrated transient and fully-integrated 3D numerical flow model with a Hydraulic Mixing Cell (HMC) method to map the different water sources in the stream corridor of the 4^th-order Selke stream and track their spatio-temporal evolution. This allows us to identify areas where waters from different sources mix enhancing the potential for turnover of groundwater-borne solutes such as nitrate. We evaluate HMC results with hydrochemical field data, and outline mixing hot-spots defined by high degrees of mixing (i.e. balanced volume fractions of the mixing endmembers in a model cell) expressed in terms of a threshold mixing degree (d=d_h) within the stream corridor. Our results show that around 50% of the water in the aquifer originates from infiltrating SW. Especially around the stream (within 250m from the stream), aquifer water is almost exclusively made up of infiltrating SW with minimal amount of water from other sources being mixed in. On average, 9% of the floodplain aquifer are characterized by high degrees of mixing (d=d_h), but this value can be nearly 1.5 time higher following big discharge events. Our modeling results further suggest that peak intensity of events is more significant for the increase of mixing degrees than event duration. We also found that discharge events mainly facilitate high mixing degrees at greater distances from the stream; while near the stream growing SW influxes dominate water composition in the aquifer and decreasing water transit times reduce exposure-times of water and solutes to the conditions in mixing hot-spots. With this easy-to-transfer modeling framework we seek to show the applicability of the HMC method as a complementary tool for the identification of SW-GW mixing hot-spots at the floodplain-scale, when simulating the spatio-temporal patterns of SW-GW exchange in stream corridors.