DOE Energy Exascale Earth System Model (E3SM) Version 2
E3SMv2 is the second version of the DOE Energy Exascale Earth System Model (E3SM), a model designed to address DOE mission-relevant science questions. E3SMv2 represents a significant evolution from its predecessor, E3SMv1, with substantial updates to its numerical core and improvements in the simulated climate. E3SMv2 is approximately twice as fast as E3SMv1 computationally. The atmosphere component, while based on E3SMv1, underwent significant changes in E3SMv2 to improve clouds and precipitation.
This work documents the lower resolution configuration of E3SMv2, which includes a grid spacing of 110 km in the atmosphere, 165 km in the land, 55 km in the river routing model, and a grid spacing that varies between 60 km in the mid-latitudes and 30 km at the equator and poles in the ocean and sea ice components. The model is evaluated using a standard set of Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations totaling over 5000 simulated years which have been released publicly. In addition to this lower resolution configuration of E3SMv2, other configurations, including higher resolution regionally refined configurations, will be documented in separate papers.
The E3SM project was conceived from the confluence of energy mission needs and disruptive changes in scientific computing technology. E3SM aims to optimize the use of DOE resources to meet the science needs of DOE. Efficient utilization of emerging computational architectures requires a significant evolution in present programming models in Earth System Models (ESMs), leading DOE to develop a new ESM, initially branching from the Community Earth System Model. The long-term goal of the E3SM project is to produce robust actionable predictions of Earth system variability and change, with an emphasis on the most critical scientific questions facing the nation and DOE (Leung et al., 2020).
With the release of E3SMv2, DOE now possesses improved capabilities to examine long-term changes in environmental factors impacting the energy sector. These capabilities will be further augmented with the upcoming release of a regionally refined model configuration of E3SMv2 with a 4x increase in resolution in all its components over North America.
E3SMv2 is approximately twice as fast (or efficient if measured in terms of power) compared to E3SMv1. The efficiency gains are achieved in the atmosphere and ocean components. In the atmosphere, they arise from a new semi-Lagrangian tracer transport method and a new grid for physics calculations. The gain in the ocean is due to a longer allowable timestep.
The model performance is evaluated with CMIP6 Diagnosis, Evaluation, and Characterization of Klima (DECK) simulations augmented with a five-member historical ensemble as well as simulations to evaluate the impacts of different forcing agents. All simulation data have been released publicly for further analysis by researchers outside of the E3SM project. The simulated climate in E3SMv2 has many realistic features of the climate system, with notable improvements in clouds and precipitation compared to E3SMv1. The climatology of the simulated precipitation improved substantially, with a 15% reduction in global error and notable reductions in regional biases (e.g., tropical Pacific Ocean, Maritime continent, Central America, and the Amazon). Additionally, a new trigger function for the deep convection significantly improved the phase of the diurnal cycle of precipitation, especially over the contiguous US. The amplitude of the diurnal cycle remains, however, weaker than observed.
E3SMv1 suffered from an excessively high equilibrium climate sensitivity (ECS) of 5.3 K. Through improvements in clouds and their feedbacks, the ECS is reduced to 4.0 K in E3SMv2, which is now within the plausible range based on a recent World Climate Research Program (WCRP) assessment.
However, important biases remain, including a weak Atlantic Meridional Overturning Circulation (AMOC) in the ocean and an underestimation of the warming in the second half of the 20th century. An analysis of single-forcing simulations indicates that correcting the historical temperature bias would require a substantial reduction in the magnitude of the aerosol-related forcing. Those biases are high priorities for improvement in future versions of E3SM.