Model land-atmosphere interactions in a Multiple Atmosphere Multiple Land (MAML) framework of SP-E3SM

Abstract:

Coupling between the atmosphere and the heterogenous land surface can play an important role in the climate system. Traditional global climate models have coarse horizontal grid resolutions on the order of 100 km that poorly capture mesoscale atmospheric processes and their interactions with the heterogenous land surface. To overcome this limitation, we have implemented a Multiple Atmosphere Multiple Land (MAML) framework in the Super-Parameterized Energy Exascale Earth System Model (SP-E3SM). In the standard SP-E3SM, all subgrid-scale atmospheric processes (e.g., radiation, clouds, and precipitation) are represented by a nonhydrostatic Cloud Resolving Model (CRM) embedded in each E3SM grid. As each atmospheric column of the CRM interacts with the same land surface represented by the E3SM Land Model (ELM), this configuration is called Multiple Atmosphere Single Land (MASL). In contrast, in the MAML framework, each atmospheric column of the CRM interacts with its own underlying land surface so non-linear effects of land-atmosphere interactions can be better represented.

To investigate the impact of the MAML approach, we perform two simulation experiments with the MASL vs. the MAML approach. Preliminary results show similar global precipitation patterns but some improvements in regional precipitation are apparent in the MAML simulation over the Amazon and the Indian Monsoon regions. Associated with the precipitation changes are changes in soil-moisture-precipitation coupling. For example, in the Indian Monsoon region, the moisture-limited evapotranspiration regime in MASL becomes an energy-limited evapotranspiration regime. The normalized gross moist stability (NGMS) framework is used to understand the links between the surface latent and sensible heat flux change and the precipitation change from an energetic perspective. Interestingly, local surface forcing change is found to be only a minor driver for the regional precipitation change, indicating important non-local impacts of land-atmosphere coupling. Further Analysis on the links between the surface heat flux and precipitation in the two simulations and future development to improve modeling of land-atmosphere interactions in the MAML framework will be discussed.