Sub-grid heterogeneity, characterized by spatial variabilities in land use/land cover (LULC) types, vegetation characteristics, soil properties, and topography, plays a pivotal role in surface energy balance and land-atmosphere interactions. However, most land surface models generally use a computationally-efficient tiling scheme to account for LULC heterogeneity but neglect the heterogeneity of vegetation, soil, and topography. Besides, how different methods of representing sub-grid heterogeneity affect surface energy balance and land-atmosphere interactions remain unclear. Here we report recent progress in the sub-grid heterogeneity representation in the land model (ELM) of the Energy Exascale Earth System Model (E3SM).

(1) We implemented a new sub-grid topographic parameterization in ELM to account for the effects of sub-grid topography on solar radiation flux. Incorporating sub-grid topographic effects overall reduces the biases of ELM in simulating surface energy balance and snow cover, especially in the high-elevation and snow-covered regions over the Tibetan Plateau.

(2) We quantified the impacts of representing sub-grid topography with different complexities on surface energy balance and surface boundary conditions for turbulent heat flux and scalar (co-)variances in ELM. The topounit-based sub-grid structure provides better performance than the default sub-grid structure.

(3) We used advanced machine learning (ML) methods to develop a novel sub-grid structure in ELM to better consider the sub-grid heterogeneity of vegetation, soil, and topography. Our ML methods can be used to identify optimal sub-grid configuration for representing surface heterogeneity.

(4) We developed a new land-atmosphere coupling scheme in E3SM via accounting for the impacts of sub-grid heterogeneity on the lower boundary condition for the atmosphere model. Using the Southern Great Plain site as a testbed, we found that sub-grid heterogeneity can modify the structure of the atmospheric boundary layer and further affect the atmospheric and cloud dynamics.

Our work advances the modeling and understanding of the impacts of sub-grid heterogeneity on terrestrial processes and land-atmosphere interactions. Further study will focus on integrating all the advancements mentioned above in global-scale Earth system modeling and investigating the climate impacts of sub-grid heterogeneity.