Hyper-resolution water modeling: PCR-GLOBWB at 1 km across Europe

GHMs are typically used for consistent global analyses of water availability such as flood risk (WRI Aqueduct Water Risk Atlas) or water shortage (National Geographic World Water Map). With climate change already impacting the water cycle, GHMs could become increasingly important in risk management. However, GHMs are currently applied at relatively coarse resolution (~100 km2), which conceals much of the local detail needed for local water availability assessments. 

Enabling GHM to run at 1 km spatial resolution is therefore paramount for studies applied everywhere and at locally relevant scales. While hydrologists have been discussing hyper-resolution GHMs for almost a decade, data availability, computational demand and storage requirements were limiting factors to put the idea into practice. With this publication, the researchers overcame most of these problems and aimed to contribute to the next development cycle of GHMs. 

The research: The first hyper-resolution model

The main objective of this study was to not only apply the model, but also to map needed improvements with respect to the currently used coarser resolutions. 

To that end, different model set-ups with various spatial resolutions were evaluated against available observations. Outcomes of the evaluation were then compared across model set-ups to determine the strengths and weaknesses of this first hyper-resolution model. 

The results: A step closer to meeting the ‘grand challenge’

The results confirmed that improvements are needed: while streamflow estimates improved over coarser models, other important hydrological parameters such as soil moisture and evaporation, do not. Despite these first gains and promising streamflow results, there is still a lot of work to do, as the uncertainty in model outputs and analysis are still large. 

Unravelling the causes for the mixed results and improving model certainty are now the most important tasks at hand. A first attempt has already been made to scale up the model application from the continental to the global scale to test possible ways forward. But many more experiments are needed to meet this “grand challenge” of a consistent hydrological model applicable at all scales, from local to global.