Subsurface lateral groundwater flow plays an important role in controlling water table dynamics. Due to the relatively coarse spatial resolutions of land surface and Earth system models, this process is often omitted even though it can be significant due to subgrid heterogeneity. In this study, we developed a physical based model to simulate subsurface lateral groundwater flow using hillslopes to represent subgrid spatial variability in topography. This model explicitly considers the smooth transition between different water table scenarios (e.g., with or without a seepage phase in the lower elevation). We coupled this model to the land component (ELM) of the Energy Exascale Earth System Model (E3SM) and applied it at global scale. Current simulations show that lateral groundwater flow is affected by topography through its impacts on water table slopes. We will further calibrate the model using observational water table depth dataset. Additional analysis will be performed to understand how lateral flow contributes to the river discharge and influences the spatial distributions of soil moisture along the hillslope. We will evaluate our simulations and investigate how surface topographic heterogeneity modulates hydrological processes.