Abstract: Based on the characteristic of the 500 hPa atmospheric circulation characteristic that could characterize the weather situation and control the weather conditions, and the principle of air-sea interaction, the two techniques of linear and monotonic curve correlation and optimization factor correlation were used to conduct census and comparative analysis on circulation and sea temperature factors. A batch of factors of atmospheric circulation characteristics and sea temperature that were extremely significantly related to the yield component factors （effective panicle number, grain number per panicle and thousand-grain weight） of rice （Oryza sativa L.） in Jiangsu Province were selected. Then, the stability of the factors obtained from the general survey was tested by the sliding correlation test, and the independence of the factors obtained from the general survey was tested by the principal component identification method. The factors of atmospheric circulation characteristics and sea temperature that were stable, independent and significantly related to the rice yield component factors were determined as the forecasting factors. Finally, based on the least squares method, circulation and sea temperature models for dynamic prediction of rice yield component factors in Jiangsu Province were established, the fitting effects of these models were good and could be put into business use.

Key words: rice（Oryza sativa L.）, atmospheric circulation characteristic, sea temperature, yield component factor, dynamic forecast, Jiangsu Province