Global warming is altering soil moisture (SM) droughts in Europe with a strong drying trend projected in the Mediterranean and wetting trends projected in Scandinavia. Central Europe, including Germany, lies in a transitional zone showing weaker and diverging change signals exposing the region to uncertainties. To analyse the projected SM drought changes and associated uncertainties in Germany, we utilize a large multi-model ensemble of 57 bias-adjusted and spatially disaggregated regional climate model simulations to run the hydrologic model mHM at a high spatial resolution (0.015625°, eq. 1.2 km) for Germany over the period 1971-2098. We show that projections of future changes in soil moisture droughts over Germany depend on the emission scenario, the soil depth and the timing during the vegetation growing period. The most robust and widespread increase in soil moisture drought intensities is projected for upper soil layers (0-30 cm) in the late growing season July-September under RCP8.5. There are greater uncertainties in the changes in soil moisture droughts in the early vegetation growing period (April-June). Regional differences are visible, with the strongest Soil moisture Drought Intensities (SDI) increase in the south-west of Germany. In the north-east, we observe divergent trends between the top soils and the total soils (0-200 cm). We find stronger imprints of changes in meteorological drivers controlling the spatial disparities of SM droughts than regional diversity in physio-geographic landscape properties. Our study provides nuanced insights for an important climatic transition zone and is therefore relevant for regions with similar transitions.

Climate change threatens the sustainable use of groundwater resources worldwide by affecting future recharge rates. However, assessments of global warming’s impact on groundwater recharge at local scales are lacking. This study provides a continental-scale assessment of groundwater recharge changes in Europe, past, present, and future, at a (5 x 5) km2 resolution under different global warming levels (1.5 K, 2.0 K, and 3.0 K). Utilizing multi-model ensemble simulations from four hydrologic and land-surface models (HMs), our analysis incorporates E-OBS observational forcing data (1970-2015) and five bias-corrected and downscale climate model (GCMs) datasets covering the near-past to future climate conditions (1970-2100). Results reveal a north-south polarization in projected groundwater recharge change: declines over 25-50% in the Mediterranean and increases over 25% in North Scandinavia at high warming levels (2.0-3.0 K). Central Europe shows minimal changes (±5%) with larger uncertainty at lower warming levels. The southeastern Balkan and Mediterranean region exhibited high sensitivity to warming, with changes nearly doubling between 1.5 K and 3.0 K. We identify greater uncertainty from differences among GCMs, though significant uncertainties due to HMs exist in regions like the Mediterranean, Nordic, and Balkan areas. The findings highlight the importance of using multi-model ensembles to assess future groundwater recharge changes in Europe and emphasize the need to mitigate impacts in higher warming scenarios.

FSO is a symbolic regression method that allows for automatic estimation of the structure and parameterization of transfer functions from catchment data. The FSO method transforms the search for an optimal transfer function into a continuous optimization problem using a text generating neural network (variational autoencoder). mHM is a widely applied distributed hydrological model, which uses transfer functions for all its parameters. For this study, we estimate transfer functions for the parameters saturated hydraulic conductivity and field capacity. To avoid the influence of parameter equifinality, the remaining mHM parameter values are optimized simultaneously. The study domain consists of 229 basins, including 7 major basins for Training and 222 smaller basins for validation, distributed across Germany. 5 years of data are used for training und 35 years for validation. By validating the estimated transfer functions in a set of validation basins in a different time period, we can examine the FSO estimated transfer functions influence on model performance, scalability and transferability. We find that transfer functions estimated by FSO lead to a robust performance when being applied in an ungauged setting. The median KGE of the validation basins in the validation time period is 0.73, while the median KGE of the 7 training basins in training time is 0.8. These results look promising, especially since we are only using 5 years of training data, and show the general applicability of FSO for distributed hydrological models.

Parameter estimation is one of the most challenging tasks in large-scale distributed modeling, because of the high dimensionality of the parameter space. Relating model parameters to catchment/landscape characteristics reduces the number of parameters, enhances physical realism, and allows the transfer of hydrological model parameters in time and space. This study presents the first large-scale application of automatic parameter transfer function (TF) estimation for a complex hydrological model. The Function Space Optimization (FSO) method can automatically estimate TF structures and coefficients for distributed models. We apply FSO to the mesoscale Hydrologic Model (mHM, mhm-ufz.org), which is the only available distributed model that includes a priori defined TFs for all its parameters. FSO is used to estimate new TFs for the parameters “saturated hydraulic conductivity” and “field capacity”, which both influence a range of hydrological processes. The setup of mHM from a previous study serves as a benchmark. The estimated TFs resulted in predictions in 222 validation basins with a median NSE of 0.68, showing that even with 5 years of calibration data, high performance in ungauged basins can be achieved. The performance is similar to the benchmark results, showing that the automatic TFs can achieve comparable results to TFs that were developed over years using expert knowledge. In summary, the findings present a step towards automatic TF estimation of model parameters for distributed models.