Land surface heterogeneity has major control on land surface processes; however, current Earth System Models (ESMs) use spatial scales too coarse to capture its effects on critical land surface processes. Recently, several new features have been introduced to improve representation of land surface processes in the Energy Exascale Earth System Model (E3SM) Land Model (ELM) including a topography-based subgrid structure, capability to downscale atmospheric forcing from grid cell mean to the corresponding subgrids and a scheme to represent subgrid topographic effects on solar radiation. In addition, recently developed high resolution land surface data enables to represent subgrid scale land cover variability. While recent studies have shown effects of the new improvements in ELM simulations separately, the relative effects of different sources of land surface heterogeneity combined with downscaling of atmospheric forcing has not been investigated. This study focuses on evaluation of the relative effects of improved representation of land surface heterogeneity influenced by topography, land cover, and climate on water and energy fluxes and stream flow simulated in ELM. For this purpose, ELM simulations with and without subgrid land cover heterogeneity, with and without downscaling of atmospheric forcing, with the plane-parallel radiative transfer scheme and with the scheme accounting for subgrid topographic effects on solar radiation. Comparison of the simulations will aim at isolating the impacts of atmospheric forcing, impacts of representing subgrid land cover heterogeneity and impacts of accounting subgrid topographic effects on solar radiation over various regions of complex terrain, regions receiving their major precipitation during different seasons as well as regions with streamflow dominated by rain-fed vs. snow-fed runoff. Furthermore, the simulations will be compared against observations in topographically complex regions.