Using Case Studies to Test and Tune a Regional Climate Model for Severe Weather Applications in the United States

     The prediction of extreme weather can be quite challenging due to its rare occurrence and a lack of observations, yet it is generally projected to become more frequent and impactful as our climate warms into the future. In regions like the southeast United States (US), where extreme weather is a frequent occurrence, generating accurate projections of extreme weather outbreaks is becoming more relevant. Regional climate models are useful tools for generating such projections, especially when used at resolutions capable of directly simulating convective processes. Location and model configuration can impact a model’s ability to represent severe weather and climate properly, and the apparent model performance can differ depending on the assessed quantity (i.e. precipitation, temperature, wind, etc.). This study aims to understand which configurations of the International Centre for Theoretical Physics regional climate model RegCM version 5 (RegCM5) can best represent convective environments in the US, specifically applying the model to severe weather outbreaks.
    We utilize RegCM5 at a convection-resolving resolution of 3 km to test over 100 different configurations with changes to the model’s vertical resolution, cloud microphysics and planetary boundary layer parameterizations, soil moisture initialization, and soil moisture time step. This presentation assesses four quantities: convective available potential energy, convective inhibition, updraft helicity, and precipitation, and compares the output to Storm Prediction Center tornado track data, NASA’s GPM IMERG precipitation dataset, and ERA5 Reanalysis data.