Abstract: In order to improve the accuracy of gust forecast,this paper used ECMWF gust forecasting method to calculate the turbulence and convection coefficients for each sub-surfaces and wind levels,and establish the gust forecast equations based on hourly wind observation data,friction velocity and low-level wind speed of RMAPS-Dongbei mesoscale model for Liaoning region from May 2020 to July 2021.Its performance was compared with that of the ECMWF gust forecast.The results show that turbulence coefficients generally range from 1 to 4 over land and from 3 to 8 over sea,convection coefficients are close to 0 over land and range from 0.2 to 0.3 over sea.Turbulence coefficients increase with increasing wind speed over land but decrease over sea,convection coefficients decrease with increasing wind force in all regions.The calculated coefficients are generally smaller than those specified in ECMWF original settings.Compared with the ECMWF gust forecast,the improved objective gust forecast reduces wind speed errors by 1~3 m·s^-1,decreases forecast errors by 20%~50%,and improves wind force level forecast accuracy by 10%~20%.Error reductions are more pronounced over land,and some stations still exhibit good forecast performance under strong wind conditions,demonstrating the advantages of mesoscale models in improving gust forecasts.