In order to study the effects of weather seasonal changes on airport arrival capacity distribution, the capacity probabilistic distribution of historical weather was obtained through statistical analysis of the original weather data. A periodic time-series model based on the capacity probabilistic distribution was built on the basis of the 1-parameter arrival capacity distribution model. Then, according to the similarity of the monthly capacity probabilistic distribution in the periodic time series, months are classified by genetic algorithm into four seasons: (3,4),(5,6,7,8,9),(10,11) and (12,1,2), and the capacity distribution and probability of each season are obtained. The results indicate that during the stratus season from May to September, the probability of reduced arrival capacity for 8 hours is 25.76% while normal capacity for 8 hours is 9.64%. The weather effects of the seasons based on the capacity probabilistic distribution are in accordance with the actual situation.