Based on the long time series gale data of 9 national meteorological stations in Liaoning province and the data of blueberry growth period, the optimal machine learning model is chosen to extend the time series of maximum wind speed, using the maximum wind speed and disaster information data of the corresponding stations, considering the frequency and duration comprehensively, the meteorological index thresholds of different grades of high wind disaster risk during the ripening period of blueberry were determined, the risk grades of mild, moderate and severe gale disasters were established, and the spatial and temporal distribution characteristics of blueberry gale disaster risk were analyzed by the frequency of disasters and the ratio of stations. The results showed that the stochastic forest model had a high simulation precision and could extend the maximum wind speed time series, and the maximum wind disaster risk threshold was ≥ 13.9 ms−1 in the mature period of blueberry, the results were verified to be in accordance with the actual situation. During the whole mature period of blueberry in 30 years, the impact of Gale Disaster Risk tended to be mitigated, and the frequency of occurrence showed a non-significant decreasing trend, the Xiuyan area showed the most significant decrease, while the Kuandian Manchu Autonomous County area showed a significant increase trend. The probability of occurrence of gale hazard over the whole mature period of 30 years is 83.3%, and the probability of occurrence of gale hazard over two years (≥ 50%), among which the Fushun region has the highest risk degree, the middle risk probability is 17.8%, the Qingyuan region has the highest risk degree, and the severe risk degree is 5.2%, mainly in Fushun and Benxi. In general, the northwest part of the Green Economic Zone is a high risk area for gale disasters, which are widespread, frequent and severe, mainly distributed in Fushun, Benxi, Qingyuan Manchu Autonomous County and Xifeng County.