Assimilation of non-conventional data to improve model initialization for the study of severe weather events using the WRF and/or ICON numerical models.

Weather forecasting, and in particular the forecasting of severe events, depends strongly on the quality of the initial conditions and on the physical complexity of the numerical models, especially over complex-terrain regions. The research aims to improve the initialization of the WRF and/or ICON models through the assimilation of non-conventional data, such as observations from weather radar, GNSS networks, satellites, and Personal Weather Stations. To this end, simulations will be carried out on representative case studies (both severe and non-severe), assessing the impact of the different data sources on forecast quality.