Abstract: In arid and semiarid regions and their surrounding areas, sandstorms are a considerable weather-related disaster. Sandstorms often lead to pollution flashovers, high-voltage ignition, and line tripping, posing a serious threat to the safe and stable operation of the power grid. Enhancing the accuracy of numerical sandstorm forecasting can effectively reduce risks to power grid operations and considerably improve the resilience of the grid to sandstorm conditions. This study is based on a previously established mesoscale sand and dust prediction system. The simulation of the sand emission process becomes aligned with actual observational data from northern China by incorporating gust effects into the dust emission scheme of the system. To assess whether the inclusion of gust wind effects enhances sandstorm simulations, this study analyzed a severe sandstorm event that occurred in late March 2004. The evaluation focused on three aspects of the forecast results: first, the simulation of the horizontal extent of sandstorms was assessed using the TOMS (Total Ozone Mapping Spectrometer) aerosol index and station observations of weather phenomena. The findings revealed that, except for a few areas near the Tarim basin, the introduction of gust effects considerably improved the prediction accuracy of the sandstorm’s horizontal extent, particularly in Northeast China. Second, the simulation of the vertical structure of sandstorms was assessed using LiDAR (Light Detection and Ranging) observation data from stations. The results indicated that, although an overestimation of the upper boundary height of the sandstorms was observed, introducing gust effects improved the accuracy of simulating the upper boundary height and the position of the high-concentration layer of sandstorms. Third, the simulated dust concentration was quantitatively evaluated using 5-min PM10 concentration data observed at multiple ground stations. The results demonstrated that, in most areas of China, incorporating gust effects considerably enhanced the simulation accuracy of the evolution of dust storms. According to these findings, introducing gust wind effects substantially improves the accuracy of sandstorm predictions for this event.