OP-E3.1 SCREAM overview

Abstract

E3SM plans to replace the current atmosphere model with a global cloud-resolving model written in C++. A clean rewrite will allow us to embrace modern software engineering practices and to remove legacy code. To ensure computational performance across the spectrum of next-generation machines, parallelism will be abstracted using Kokkos templating. The initial focus is on computational efficiency; by making the model as simple as possible we can explicitly resolve more scales for a given number of core hours. Explicitly resolving processes (like deep convection) which were previously parameterized is important because it removes important sources of uncertainty from model predictions. However, even at the target grid of ~3km some processes need to be parameterized. We will use the Simplified Higher-Order Closure (SHOC) macrophysics/turbulence scheme, Predicted Particle Properties (P3) microphysics, and Rapid Radiative Transfer for Energetics/Radiative Transfer for GCMs – Parallel (RTE-RRTGP) radiation. In this talk, we will provide an overview of our plans for and progress on developing this new model.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. It is supported by the Energy Exascale Earth System Model (E3SM) project, funded by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research. IM Release LLNL-ABS-758567.