Performance Evaluation of the Makkonen Model Driven by the Output of the WRF Model in Simulating Power Line Icing Thickness in the Mountainous Regions of Southwest Sichuan, China

Abstract Power line icing frequently occurs in the mountainous regions of southwestern Sichuan, China, posing significant risks to power system operations. Numerical simulation offers an effective tool for predicting icing processes; however, the applicability of the Makkonen model driven by the output of the weather research and forecast (WRF) model in this region remains uncertain. In this study, a series of simulations were conducted using the WRF and Makkonen models with four nested domains (9, 3, 1, and 0.33 km) and two cloud microphysical schemes (Thompson and Morrison), based on eight observed icing events at Erlangshan station. Results show that the WRF output‐driven Makkonen model can reasonably reproduce the icing process although the icing thickness is generally underestimated. Increasing the horizontal resolution can improve the simulations of near‐surface air temperature and moisture, leading to more accurate icing estimates. This improvement is primarily attributed to a better representation of terrain effects, including enhanced ridge‐induced cooling, wind acceleration, and orographic lifting, which promote the formation of super‐cooled liquid water and increase collision and freezing efficiencies. Moreover, adopting the Morrison scheme produces larger icing thickness and shows better agreement with observations than using the Thompson scheme. This difference is mainly due to microphysical processes. Adopting the Morrison scheme increases rain number concentration and shifts the particle size distribution toward smaller droplets, thereby reducing the conversion efficiency of cloud water to precipitation and ice. As a result, more super‐cooled liquid water is retained, enhancing icing growth and reducing model bias.