The storm in the control simulation forms in a convectively unstable, horizontally homogeneous boundary layer (BL) east of a dryline and subsequently experiences a progressively cooler, drier, and more stable BL featuring a ~ 4 °C decrease of surface potential temperature as it moves eastward over a distance of ~ 100 km. East of this linear thermal gradient zone, the mesoscale inversion region is horizontally homogeneous. Several storm simulations have been performed in which the horizontal temperature and moisture gradients are altered to examine the impact of varying BL stratification on simulated storm evolution and lifetime. Although the simulated storm could not survive if initiated in the stable inversion region, a key finding is that the storm may be long-lived provided it initiates in the unstable BL and subsequently propagates into the stable BL. A sensitivity test lacking a stable inversion region produces a (unobserved) mesoscale convective system, indicating the role of the stable inversion to suppress secondary convection and upscale growth along the forward flank downdraft boundary.
The control simulation produces a supercell storm similar in several key respects to the observed Binger storm, including: (1) an intense and persistent, bell-shaped rotating updraft with flanking cloud striations; (2) a mid-level bounded weak echo region (BWER) and low-level hook echo; (3) hail; (4) a deep mesocyclone extending from low-levels (< 1 km) to 10 km above ground level (AGL) with peak low-level mesocyclonic intensity at ~ 5 hours after CI; and (5) cyclic low-level mesocyclone evolution. After rapidly intensifying by ~ 1.5 hours, the simulated supercell slows and turns right in response to the upshear-propagating cold pool boundary which increasingly resists ambient low-level shear and provides BL lifting to help force the continuous redevelopment of flanking line convection. Eventually, the storm weakens in the cooler BL, as a steadily decreasing cold pool temperature deficit leads to weaker solenoidally-forced BL updrafts. The simulated storm subsequently develops a narrowing, tilted low-precipitation updraft and evolves into a short-lived multicell mode accompanied by decreasing magnitudes of updraft and precipitation content. Updraft demise is triggered by weaker BL convergence and stronger stratification as updraft parcels ultimately remain below the level of free convection following simple up-down trajectories.