How does downscaling of atmospheric forcing vary across different regions of CONUS in ELM?
First Author
Teklu K Tesfa
teklu.tesfa@pnnl.gov
All Authors
Ruby Leung, Michael Brunke, Peter Thornton, Zhuoran Duan
Topic
Land - River - Human
Project
E3SM
Abstract
There are various ongoing research efforts to improve the representations of the effects of small-scale land surface heterogeneity, which are not well represented in ESMs. In this study, the effects of topography-based subgrid structure (TGU) combined with downscaling of atmospheric forcing introduced in the Energy Exascale Earth System Model (E3SM) Land Model (ELM) are evaluated across different regions. For this purpose, ELM offline simulations are performed using two model configurations with (ELMD) and without (ELMNoD) downscaling of the gridcell mean atmospheric forcing onto the TGUs of the land model, using the same land surface properties. Here downscaling of atmospheric forcing is performed using only information of the subgrid topography. The effects of TGU and downscaling of atmospheric forcing on land surface processes are then evaluated over the contiguous US (CONUS) and within regions determined from the season of maximum precipitation and topography-based indices. The CONUS level results generally suggest that ELMD generates more snowfall and SWE, and higher runoff and less ET during spring and summer. Regional comparisons show that the effects of downscaling vary from region to region. The effects of downscaling are more pronounced over regions which receive their maximum precipitation during SON and DJF compared to those of MAM and JJA and in regions where most of the TGUs are above the gridcell mean elevation. Furthermore, ELMD predicted observed SWE better than ELMNoD at 83% of SNOTEL sites, with the best performance over more topographically heterogeneous areas. The results in this study highlight the importance of improving the representation of small-scale climatic heterogeneity due to topography in ESMs and motivate future research to understand the impacts of small-scale surface heterogeneity on land-atmosphere interactions over mountainous regions.
