Karstic recharge processes have mainly been explored using discharge analysis despite the high influence of the heterogeneous surface on hydrological processes. In this paper, we introduce an event-based method which allows for recharge estimation from soil moisture measurements. The method was tested at a karst catchment in Germany but can be applied to other karst areas with precipitation and soil moisture data available. It will allow for a better characterization of karst recharge processes.

We used a combination of aerial, thermal, hydrologic, and radionuclide monitoring to investigate intertidal springs flowing into a coastal lagoon with a threatened ecosystem. Field data highlight the critical hydrologic and thermal role of these springs in the nearshore zone, and modelling results reveal that the groundwater springs will likely warm substantially in the coming decades due to climate change. Springs sourced from shallower zones in the aquifer will warm first.

We show that major groundwater resources in the UK exhibit strong multi-year cycles, accounting for up to 40 % of total groundwater level variability. By comparing these cycles with recorded widespread groundwater droughts over the past 60 years, we provide evidence that climatic systems (such as the North Atlantic Oscillation) ultimately drive drought-risk periods in UK groundwater. The recursive nature of these drought-risk periods may lead to improved preparedness for future droughts.

Agricultural operations can result in nitrate contamination of groundwater, lakes and streams. At two confined feeding operations in Alberta, Canada, nitrate in groundwater from temporary manure piles and pens exceeded nitrate from earthen manure storages. Identified denitrification reduced agriculturally derived nitrate concentrations in groundwater by at least half. Infiltration to groundwater systems where nitrate can be naturally attenuated is likely preferable to off-farm export via runoff.

To localize the source of a contaminant in the subsurface, based on concentration observations at some wells, we propose to test different possible locations and minimize the misfit between observed and simulated concentrations. We use a global optimization technique that relies on an expected improvement criterion, which allows a good exploration of the parameter space, avoids the trapping of local minima and quickly localizes the source of the contaminant on the presented synthetic cases.