Abstract
Although some prior research suggested little benefit in forecasting skill when horizontal grid spacing is reduced below 3 km for simulation of convection, it appears some important aspects of thunderstorms are better depicted with 1-km grid spacing than with 3-km. In particular, upscale growth happens faster in 1-km runs, because new convection is initiated by gravity waves, whereas in 3-km runs, new cells wait until cold pool boundaries collide. In addition, 1-km runs simulate more bow echoes, with a frequency better agreeing with observations, as it appears storm-scale processes are able to initiate bowing, whereas in 3-km runs, other larger-scale features such as strong storm-relative inflow or the presence of mesoscale boundaries, may need to be present to trigger bowing in most cases. Finally, the challenges of forecasting the most damaging single thunderstorm event in United States history, the 10 August 2020 derecho, along with the mechanisms that allowed winds to exceed 60 m/s and last for up to an hour in some locations, will be discussed using 1-km and 3-km runs from the FV3, WRF, and MPAS models.
BIO
Dr. William Gallus is a Distinguished Professor of Meteorology at Iowa State University, starting his career there in 1995. He received his PhD in Atmospheric Science in 1993 at Colorado State University. During his career, he has published over 120 refereed papers, and given over 250 conference presentations, generally focused on the use of numerical weather prediction models to better understand and forecast mesoscale phenomena, particularly heavy thunderstorm rainfall and severe weather. He has mentored 140 undergraduate student research projects and nearly 60 graduate students. He has received several teaching and research awards at his university, and the T. Theodore Fujita Research Achievement Award from the National Weather Association for his significant research contributions to operational meteorology.