Abstract: The losses and risks caused by meteorological disasters are determined by the potential hazard of the meteorological events, vulnerability of the hazard affected body, as well as natural and manmade resilience in the disaster pregnant environment. Among these factors, two of them related to hazard affected body or disaster pregnant environment are comparatively stable for a certain place and period, and meteorological hazards are however changeable. The spatial and temporal distribution patterns of meteorological hazards largely determine the place and evolution of disasters. Accurate diagnosis of the hazard degree is the base for assessing the loss and risk of meteorological disasters. In this paper, the definition of meteorological hazard and its diagnostic model for both local and regional areas are put forward, and the relationship between the value of the random variables and its probability is applied to describe the distribution of weather event hazards and build up the risk threshold field that is composed of hazard threshold values at each risk probability level. The meteorological hazard degree is defined as the integration　of the similarity and approach degree of the observational values for an event to the threshold value at various probability risk levels in the risk field. Using the random variable probability distribution model, the meteorological hazard threshold values under the specific risk probability levels in any given place and their threshold field are determined, and the quantitative hazard degree diagnostic model is established by combining the spatial similarity with the distance parameters. As a demonstration, the gale(strong wind) hazard diagnostic model is set up and then the hazard degrees about TC0509,TC0515 and the other gale events in the thirty cases in Shanghai are estimated. The results are as follows : the largest gale hazard degree is from TC 7708,its overall hazard degree level is mostly close to the threshold of maximum wind speed of 8-year return period. The gale hazard degrees of TC 8615 and the case of other weather process in 3 June, 1983 are secondary, with the value close to that of 7-year return period, and the maximum wind speed hazard degrees of TC 9711 and TC 8114 are close to that of 4-year return period. The gale hazard degrees of TC7413, TC7503,TC7909 and TC8506 are close to the 3-year return period, and the TC0509 "Matsa" and the TC0515 “Kanu” are close to 2-year return period. The results also show the main distribution characteristic of annual maximum wind speed with return periods from 1 to several hundreds years in Shanghai. The coastal regions in Shanghai such as Nanhui, Chongming, Jinshan are the areas of high wind speed and also the high-value areas of the maximum wind speed under a certain risk probability level, and, at the same time, the high frequency and high-risk areas of strong winds more than eight grade are located here as well. In contrast, the inland regions in Shanghai are the low wind speed areas and the low-value areas of maximum wind speed under the certain risk levels where strong winds more than eight grade are rare observed.Application results show that the diagnostic approach is objective and quantitative in assessing the hazard degree of weather events, which provides a reference for the meteorological disaster evaluation.