In East Africa are conflict over water and vegetation prominent. On top of that, water abstraction of commercial farms are increasing the competition of water. Therefore, this study has developed a model which can investigate what the influence is of these farming activities on the water balance of the region and people's livelihood activities in times of dry periods. We do that by ‘replacing’ the farms in the model, and see what the effect would be if there were communities or forests instead.

Understanding how reactive solutes move through nearly saturated soil is important for many environmental processes. However, this movement is influenced by both soil deformation and chemical reactions, which are often treated separately. This paper proposes a flexible framework that links these coupled effects. The results show that changes in loading, moisture, and mineral reactions can shift both the rate and pattern of solute movement, improving long-term predictive capability.

The Explore2 project has provided an unprecedented set of hydrological projections in terms of the number of hydrological models used and the spatial and temporal resolution. The results have been made available through various media. Under the high-emission scenario, the hydrological models mostly agree on the decrease in seasonal flows in the south of France, confirming its hotspot status, and on the decrease in summer flows throughout France, with the exception of the northern part of France.

As natural hazards evolve, understanding how extreme events interact over time is crucial. While single extremes have been widely studied, joint extremes remain challenging to analyze. We present a framework that combines advanced statistical modeling with copula theory to capture changing dependencies. Applying it to historical data reveals dynamic patterns in extreme events. To support broader use, we provide an open-source tool for improved hazard assessment.

Pumping wells may not be officially registered or documented. We develop a new framework to jointly estimate spatially variable conductivity and identify unknown pumping well locations and rates. Our results support the ability of the new approach to accurately estimate conductivity and identify well location and rates under diverse configurations, attaining a quality of performance similar to its traditional counterpart while computational time is reduced by nearly an order of magnitude.

Common metrics used to evaluate hydrologic models make it relatively easy to achieve high performance scores in highly seasonal catchments. However, we analysed 18 hydrologic models and found that almost all were worse at simulating interannual variability and change in seasonal streamflow regimes. This suggests that climate change impacts on streamflow may not be accurately predicted in highly seasonal tropical, alpine, and polar regions, which are highly vulnerable to climate change.

This study investigated groundwater-driven nutrient fluxes in a Yangtze Basin oxbow lake using a year of high-frequency monitoring. Water level and ²²²Rn tracing revealed a seasonal groundwater discharge pattern. This dynamic was regulated by net precipitation, which controlled lake level fluctuations and subsequent groundwater discharge rates. Associated N and P loads varied synchronously with discharge. The N/P closely matched that of lake water and correlated with chlorophyll-a dynamics.

We compared global vs. local deep learning models for groundwater level prediction using ~3,000 wells across Germany. Unlike surface water, groundwater is complex and data-scarce. Results: global models show no systematic accuracy advantage over local ones. Data similarity matters more than quantity for better predictions. Successful groundwater modeling requires strategies tailored to these unique complexities, not just larger datasets.

We apply and evaluate a statistical method to adjust biases in high-resolution climate projections for England. By comparing modelled rainfall and temperature with observations, we show that the adjusted projections agree much more closely with real conditions for both average patterns and extremes, providing a more reliable basis for high-resolution impact modelling and assessments that depend on hourly precipitation data.

We estimate daily evapotranspiration (ET) across the United States using the ‘surface flux equilibrium’ approach, which assumes that the balance of temperature and humidity in the atmosphere reflects recent ET on land. Using triple collocation, we compare our estimates to three other ET datasets and find that the surface flux equilibrium ET method performs well. Surface flux equilibrium ET may therefore be useful for hydrologic studies where simple, parameter-free ET estimates are advantageous.

This study quantifies irrigation water pressure and upstream–downstream interactions in the Lancang–Mekong Basin under climate change. Results show increasing irrigation pressure, with dry-season demand reaching 59 % of available water by 2040. Key subregions exhibit a structural shift from internally to externally driven pressures, highlighting spatial heterogeneity and nonlinear dynamics. Findings inform vulnerable components, transboundary responsibility, and differentiated water governance.

Hydrological models are vital for water management but often fail to predict water flow in dynamic catchments due to model simplification. This study tackles it by developing an optimized calibration framework that considers dynamic catchment characteristics. To overcome potential difficulties, multiple schemes were tested on over 200 U.S. catchments. The results enhanced our understanding of simulation in dynamic catchments and provided a practical solution for improving future forecasting.

We developed a new way to understand how droughts grow, persist, and decline over time and space.  Applied to Northeast Brazil, a region frequently hit by severe droughts, the approach identified four recurring patterns of drought behavior. Most droughts there expand quickly, last for a long time, and end suddenly. Our method can help decision-makers better understand and manage droughts, supporting stronger preparedness and resilience in regions at risk.

This study presents an experimental framework (EF-SPM) to disentangle and evaluate the benefits of spatially explicit parameterization and multi-gauge calibration in distributed hydrological modelling. Experiments in a nested catchment show that both strategies consistently improve streamflow simulations across sub-basins, jointly alleviating multi-objective competition and the trade-off between spatial complexity and parameter identifiability.

This study investigates how soils respond to major climatic disturbances, such as the extreme drought in Germany in 2018. Using long-term lysimeter observations and an artificial intelligence model, we show that persistent shifts in soil water dynamics indicate changes in hydraulic properties that may affect soil health, emphasizing the need for continuous monitoring under a changing climate.