Modeling and Machine Learning Analysis of Surface Mesovortices Formation and Maintenance in the St. Louis Metropolitan Area

This study aims to evaluate the formation of surface mesovortices using a high-resolution Weather Research and Forecasting (WRF) model. A simple Machine Learning (ML) tracking algorithm was used to identify relevant areas of mesovortex formation in the high spatial and temporal resolution surface mesonet data, thus reducing the amount data to be processed.The WRF model assimilated the surface mesonet data to investigate the three dimensional structure of the mesovorticies. Since WRF used was used as an analysis tool the surface mesonet data was assimilated over the entire forecast period. Simulations used a 1km and 300 m grid over a three period centered on the observed mesovortex time. Preliminary results show that these mesovorticies form as a result of the interaction between the Saint Louis Urban Heat Island, the Mississippi river and local topography