I varied the initial model profile in a series of supercell simulations from a baseline temperature (``cool" simulations) to temperatures that ranged from 0.125 to 1.5 K warmer than the baseline profile (``warm" simulations). These temperature variations minimally affected severe weather parameters that are commonly used to forecast supercell potential. I gave all simulations a wind profile with directional shear in the lowest 1 km of the atmosphere and straight shear above 1 km, which are common characteristics of tornado outbreaks. In all simulations, a left-and-right storm split occurred from an initially solitary updraft. In the cool simulations, the cyclonically rotating supercell turned southwestward, intensified, and eventually developed into an MCS. In contrast, the cyclonically rotating supercell quickly decayed in the warm simulations. The supercell's temperature perturbation at low-levels was larger in the cool simulations than in warm simulations. This difference led to the cool simulations having comparatively enhanced low-level ascent and dynamic pressure perturbations, which drove their ``rightward" propagation.
These results highlight a key predictability limit of supercells within predominantly straight wind profiles, where small variation in low-level temperature that are not apparent to a forecaster may determine whether or not supercells will sustain themselves.