A Climate Model Ensemble for Understanding Future Changes to Extreme Weather

PI Paul Ullrich , University of California
Co-PI Sara Pryor, Cornell University
Colin Zarzycki, Pennsylvania State University
Stefan Rahimi-Esfarjani, Univeristy of California Los Angeles
Melissa Bukovsky, National Center for Atmospheric Research
Alan Rhoades, Lawrence Berkeley National Laboratory
Ullrich Tropical Cyclones Graphic

A tropical cyclone as found in climatological simulations with the Energy Exascale Earth System Model (E3SM) run at low resolution (top, 110km grid spacing) and high resolution (bottom, 28km grid spacing).  Columns show (from left to right) upwelling longwave radiative flux, column integrated water vapor, and precipitation rate.  While the low resolution tropical cyclone is discernible as little more than a blob of convective activity, the high resolution simulation produces a storm with a clear eye and spiral rain bands.

Project Summary

An understanding of the effects of climate change on extreme weather and atmospheric hazards is essential to ascertain future socioeconomic and infrastructural impacts from these events. The awarded effort from this project will produce the world’s first high-resolution “medium” ensemble from a single global modeling system, using regional refinement in the Department of Energy’s recently released Energy Exascale Earth System Model (E3SM) version 2.

Project Description

An understanding of the effects of climate change on extreme weather and atmospheric hazards is essential to ascertain future socioeconomic and infrastructural impacts from these events. However, because extreme events are inherently rare and often localized, a large number of high-resolution simulations are required to ensure sufficient statistical fidelity when making statements regarding projected changes to these features. While large ensembles of climate model simulations have been invaluable in recent years for putting tight bounds on future changes in synoptic meteorology, the finest of these model simulations use a grid spacing of ~80km, which precludes their use for fine scale weather impacts from extreme features such as tropical cyclones or atmospheric rivers.

The awarded effort will produce the world’s first high-resolution “medium” ensemble from a single global modeling system, using regional refinement in the Department of Energy’s recently released Energy Exascale Earth System Model (E3SM) version 2. Specifically, 10 simulations will be performed covering the period from 1950 through 2100 with a refinement region for E3SM that covers the conterminous United States (CONUS) at a grid spacing of approximately 22km. This grid spacing captures aspects of the large-scale structure, frequency and location of tropical cyclones, atmospheric rivers, wind storms, and winter storms, as well as quantify their changes in the future. These simulations will provide better estimates of mountain snowpack in the Western US, and will provide valuable feedback to the E3SM developers on model performance for extreme weather events. In addition, the most extreme 20 events will be simulated at high resolution using the Simple Cloud-Resolving E3SM Atmosphere Model (SCREAM) at 3.5km grid spacing and Weather Research and Forecasting (WRF) system at < 2 km grid spacing. This will enable us to investigate if phenomena generated by E3SM are as hazardous as historically-based events. Both the large ensemble simulations and downscaled simulations will further allow us to understand if historical extremes are close to the “worst case” for such extremes, or if there is risk for even more damaging extremes beyond the observational record

Allocations