Wildfires play an important role in the Earth climate system by means of complex two-way couplings with climate change. Climate change is expected to increase the wildfire risk by altering fire-driving variables, such as temperature, soil moisture, wind speed, relative humidity, land use, and vegetation. Observational evidence shows that wildfire frequency and intensity have been increasing in many regions, including the western contiguous United States (US), over recent decades. Although its importance is widely recognized, wildfire simulation suffers from large uncertainties with the current global Earth system models (ESMs).

At least two major factors limit the fidelity of wildfire representation in these simulations: i) The grid resolution is too coarse to resolve critical wildfire processes; ii) Some important physical processes (e.g, chemistry-aerosol interactions) are often missing in the high-resolution global ESMs. In the present study, we will attempt to overcome these limitations based on the recently released Energy Exascale Earth System Model version 2 (E3SMv2) and further developments tailored to large wildfires. We will push the E3SMv2 regionally refined model resolution to the convection-permitting (a few kilometers) scale in California and leverage the Simple Cloud Resolving E3SM Atmosphere Model developments on the global uniform grids. Furthermore, we will take advantage of the E3SM next generation developments carried out by our group to improve the chemistry and aerosol treatments for the stratospheric fire smoke.

Here we will present the initial results from this new global multiscale simulation framework focusing on case studies, for example the 2020 Creek fire that are known to have generated pyrocumulonimbus and to have contributed to considerable changes in atmospheric radiation and air quality across the western US.

This work is performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNL-ABS-839106