Assessing WRF simulations for thermal forecasting in paragliding flights

Abstract Generated by differential surface heating within the atmospheric boundary layer (ABL), thermals are essential for sustaining flight in non-motorized aviation such as paragliding, hang gliding, and gliding. Despite their relevance, operational thermal forecasting remains limited, particularly at the spatial scales relevant for recreational flights. In this study, the Weather Research and Forecasting (WRF) model is applied at very high spatial resolution to simulate the development of thermals through vertical wind velocity fields in a pre-Alpine area of northern Italy. Model outputs are evaluated against Global Positioning System (GPS) flight tracks collected during paragliding activities in some summer days of July 2023. The results show the WRF model captures the spatial patterns of regions where thermals are unlikely to occur with good skill, while its ability to correctly predict the presence of thermals remains limited. The model tends to underforecast thermal events, particularly at the spatial and temporal scales required for flight navigation. Nevertheless, most comparisons perform better than a random classifier, indicating the model contains physically meaningful information on thermal-related processes, suggesting that high-resolution, physically based atmospheric models might provide a valuable basis for thermal forecasting in recreational aviation.