Factor Separation Analysis of Urban Impacts on a Simulated Supercell

The effect of urban areas on weakly-forced precipitation events is well-documented, and the causes of these precipitation modifications are generally understood. However, the effects of urban areas on precipitation systems in synoptically active regimes, particularly severe convection, are relatively unstudied. While our previous investigations of this topic have shown that a large urban area can have a statistically significant effect on a simulated supercell, this topic merits further work. In particular, we aim to quantify the individual contributions of the urban heat island and slower near-surface wind speeds induced by increased surface roughness to urban-induced storm modifications . To do so, this study will use the ARW-WRF model to simulate an isolated supercell interacting with a large Plains urban area (Dallas-Ft. Worth; DFW) in 108 gridded locations within the domain, where each urban location is associated with its own individual simulation. A factor separation approach will be undertaken to determine the relative importance of urban roughness and thermal characteristics in the modification of storm properties. This process will involve performing all 108 simulations three times: once with the urban area represented only by surface roughness differences, once with it characterized only by its thermal properties, and a final time with both aspects represented. An additional initial-condition ensemble of control simulations (CTRLE) without any urban areas will be used for comparison, resulting in a total of 324 unique, high-resolution, 24-hr simulations. By comparing the various DFW simulations with CTRLE, the effects of the various aspects of the urban area on storm strength and evolution will be isolated and examined in detail. These comparisons will focus on attributing inter-simulation differences in storm structure, dynamics, and strength to pre-storm urban-atmosphere interactions.