T-REX: the tile-based representation of lateral exchange processes in ICON-Land

The vast majority of land-surface models uses a tiling-approach to capture the effects of subgrid-scale spatial heterogeneity in the land-surface properties. In most cases, however, the tiles, which represent patches with homogeneous characteristics at and below the surface, are treated independently of each other and the lateral exchange between them is not being taken into consideration. The present manuscript describes an approach for a tile-based representation of lateral exchange processes in heterogeneous landscapes that was recently implemented into ICON-Land, the land surface component of the ICON framework. The scheme captures the horizontal fluxes on a broad range of spatial scales and represents 5 lateral exchange processes, namely gravity-driven moisture fluxes and the corresponding advective heat transport, diffusive- and conductive fluxes of water and heat as well as the redistribution of snow between small-scale topographic features. In the approach, the relationships between any two tiles are determined by a set of characteristic connectivities which are treated as inherent properties of the pair of tiles – invariant in time and independent of the location – and derive from the internal logic underlying the definition of the tiles. The characteristic connectivities are used to calculate the spatio-geometric relationships between two tiles, such as the (geometrical) contact length or the characteristic center-to-center distance between two adjacent surface clusters. These, in turn, define gradients as well as the time-lag factors that govern the lateral transport in the model. In addition to a description of the model development, we present two example applications of the new scheme, which address the effect of sub-grid scale fluxes on the model's ability to capture the spatial variability in the state of the surface and sub-surface and the overall terrestrial water storage. Here, our results suggest that lateral exchange processes, especially on small horizontal scales, are highly relevant for the spatial variability in the soil temperatures and for the simulated extent of surface water bodies, while the effects on the grid-cell mean state and the turbulent exchange with the atmosphere appear to be largely negligible.