Abstract: Using the original data of forest fires frequency from 1952-2010 and the weather stations data in Jilin province, an analysis is made on the temporal variations of forest fires frequency and its relationship with meteorological conditions. The moisture balance method and statistical regression analysis method are used to establish climate impact assessment models of forest fires frequency respectively, and then one model is contrasted with another one. On this basis, a time variation series of forest fire meteorological indexes is established by using moisture balance model. The results show that forest fires frequency is closely related to precipitation, relative humidity, temperature, wind speed and other meteorological factors, and the climate impact assessment result of the moisture balance model is better than that of the statistical regression model. Both index and frequency of forest fires in Jilin province are in increase trend in the latest 62 years, which show climate change is in favor of the occurrence of forest fires. After 2000 forest fires in spring and autumn occur frequently in Jilin.