High resolution simulations of the 3 May 1999 Oklahoma tornado outbreak: impact of microphysics on cold pool intensity and storm morphology

We present a series of high resolution convection-permitting numerical simulations of the 3 May 1999 Oklahoma tornado outbreak, using the University of Oklahoma ARPS model, with a goal of examining the importance of model physical parameterizations, particularly microphysical parameterizations on the evolution of the simulated convection. The outbreak was characterized by several tornadic supercells in relatively close proximity during the tornadic phase of their life-cycles. This condition is often seen as detrimental to the development of tornadoes in supercells due to storm interaction, particularly the ingestion of rain-cooled outflow from adjacent storm cells. Successful numerical simulation/prediction of such events in which individual storm cells are explicitly resolved by the model is challenging due to the tendency of current popular microphysical schemes to produce overly-intense cold pools that promote the upscale growth of model convection into an MCS far more quickly than often is the case in reality. In particular, on 3 May 1999, the individual storm cells maintained discreteness, resisting the tendency to develop upscale into an MCS, for several hours. Preliminary high resolution (3km and 1 km) simulations using the Lin and WRF Single-Moment 6-class (WSM6) ice microphysics schemes produce individual discrete cells which quickly merge or promote additional development along the leading edge of the cold pool, producing a squall line after only 1-3 hours. The goal of this study is to investigate whether overly-intense cold pools produced as a result of the microphysics scheme promote this erroneous storm evolution. We also examine possible ways to curtail the problem, including adjusting various microphysical parameters, particularly the values of the intercept parameters of rain, snow, and graupel/hail. A modification to the WSM6 scheme will be tested in which the intercept parameter for rain is diagnosed from the rainwater content based on empirical relations. The goal is to examine the impact on cold pool intensity and storm morphology.