Customization of Limited Area Model for complex orography territories: the WRF and the WRF-Chem UniMe model

Since ancient times, knowledge of weather conditions has been essential for human survival and for planning activities. In this context, atmospheric forecast models play a key role in weather studies, contributing significantly to understand and predict short- and long-term weather changes. In the field of atmospheric modelling, weather forecast models can be classified into Global Models (GMs) and Limited Area Models (LAMs) in relation to their both domain coverage and spatial resolution. Especially in the last few years, the application of LAMs with high spatial resolution has become more and more important, as they can perform high-efficiency predictions for the increasingly frequent severe weather phenomena. In particular, the Weather Research and Forecasting (WRF) model operates in the field of Numerical Weather Prediction (NWP); it was developed as a result of a collaboration between National Center for Atmospheric Research (NCAR), National Center for Environmental Prediction (NCEP) and Earth System Research Laboratory (ESRL). Furthermore, the WRF-Chem model, which can be considered as the chemical extension of the WRF model, is widely employed for modelling atmospheric pollutants. In this framework, the present contribution is focused, at first, on the optimization steps performed on the WRF model applied to Sicily, a region characterized by a complex orography and peculiar geophysical conditions. In addition, due to the presence of different active volcanoes in Sicily, the WRF-Chem model is also used for forecasting volcanic ash as well as desert dust intrusion. Furthermore, the hardware and software structures of the customized WRF and WRF-Chem models developed at Messina University and the adopted management procedures will be described. In particular, the software routines to follow for the forecast chain and the procedures to be adopted in case of faults will be described step by step. Finally, in order to show the prediction capability of the customized WRF and WRF-Chem models, two case studies will be presented and discussed.