Simulating subsurface heterogeneity improves large-scale water resources predictions

Abstract:

Heterogeneity is abundant everywhere across the hydrosphere. It exists in the soil, the vadose zone and the groundwater. In large-scale hydrological models, subsurface heterogeneity is usually not considered. Instead average or representative values are chosen for each of the simulated grid cells, not incorporating any sub-grid variability. This may lead to unreliable predictions when the models are used for assessing future water resources availability, floods or droughts, or when they are used for recommendations for more sustainable water management.

In this study we use a novel, large-scale model that takes into account sub-grid heterogeneity for the simulation of groundwater recharge by using statistical distribution functions. We choose all regions over Europe that are comprised by carbonate rock (~35% of the total area) because the well understood dissolvability of carbonate rocks (karstification) allows for assessing the strength of subsurface heterogeneity.

Applying the model with historic data and future climate projections we show that subsurface heterogeneity lowers the vulnerability of groundwater recharge on hydro-climatic extremes and future changes of climate. Comparing our simulations with the PCR-GLOBWB model we can quantify the deviations of simulations for different sub-regions in Europe.