Using HydroTiles to represent different hydrological regimes in a global Earth System model

Land surface regions that are of crucial importance for climate dynamics, such as Arctic permafrost landscapes, are often extremely heterogeneous. In these areas, hydrological processes and heat fluxes, which are influenced by topographic features on the scale of a few meters, can affect processes such as permafrost thaw over large regions. Despite the emergence of Earth system models that can operate at a resolution down to one kilometer, hydrological heterogeneity at smaller scales is often overlooked. In addition, high-resolution models are computationally intensive, making them unsuitable for the time scales required to study the climate impacts of processes such as permafrost thaw. In this study, we present an extension to the tiling infrastructure of the ICON Earth system model that enables the representation of different hydrological regimes within individual grid cells. This innovative approach facilitates the representation of lateral water flow connections between different areas within grid cells and the simultaneous representation of different surface water and soil moisture states, such as dry and wet conditions, within a single grid cell. The impact of this improvement is twofold. First, it provides a more accurate representation of surface and soil hydrology. Second, it is expected to improve the representation of land-atmosphere coupling, allowing us to better capture feedbacks across landscapes affected by strong hydrologic contrasts. By enabling the representation of hydrological features in subgrids through tiles, which we call HydroTiles, we hypothesize that the HydroTiles setup could replicate some features of high-resolution simulations even at lower resolutions. This approach offers the potential to make simulations more computationally cost-efficient. In our presentation, we would like to highlight the advantages and disadvantages of the HydroTile setup compared to high-resolution simulations.