Abstract: Rice is one of the most important crops in Heilongjiang province.It is significant to study rice yield prediction method for food production in this region.Based on historic meteorological influence index for bumper or poor harvest of rice,a dynamic prediction model was established for region-specific rice yield,by using the data of rape yield,development stage,daily maximum and minimum temperature,daily precipitation and daily sunshine duration from individual main producing region.The key meteorological factors,which affected the rice yield,were determined by using correlation analysis method,and the corresponding yield prediction model was established.The dynamic prediction model for the trend of bumper or poor harvest was revised.Rice yield was predicted dynamically with the model from 1997 to 2006,which showed that the average accuracy of the increase or decrease trend for rice yield was 90%,70%,90% and 80% for May 31,June 30,July 31 and August 31 respectively.The dynamic prediction model realized a successive dynamic quantitative prediction of rice.

Key words: Rice, Rice, Dynamic prediction for the trend of bumper or poor harvest, Key meteorological factors, Yield prediction