Characteristics and Cause of Changes in Mesoscale Convective Systems within a Convection Permitting Regional Climate Model

Mesoscale convective systems (MCS) are organized clusters of thunderstorms responsible for a significant portion of total annual precipitation within the central United States. At the coarse grid spacings employed within most global climate models, MCS and their attributes are poorly resolved, making it difficult to determine how these storms may change in a future climate. We apply an automated MCS classification and tracking algorithm to a set of continental-scale convection-permitting regional climate model output encompassing several 15-year periods. Three periods are analyzed, consisting of a historical (HIST) and two end-of-century periods respectively forced by bias-corrected output from the Community Earth System Model under an RCP8.5 (EoC85) and RCP4.5 (EoC45) emissions scenario. Overall, differences in the probability distribution function between these three periods reveal a greater likelihood of MCS that are larger, longer in duration, and produce higher maximum hourly rainfall intensities in the future climate states. Furthermore, statistically significant increases in MCS count of up to 33% for the Northeast during the warm-season and 48% for the Midwest in the spring indicate that MCS frequency is also increasing. Combined increases in the frequency and intensity of MCS results in a greater fraction of total annual rainfall that can be attributed to these systems. An analysis of the synoptic-scale environment preceding each MCS is performed to isolate the relative importance of changes in the mean dynamic and thermodynamic environments towards the overall response.