Abstract:

Heilongjiang Province serves as the primary cultivation region for high-quality japonica rice production in China. The occurrence of heavy rainfall during the flowering period significantly impacts the seed setting rate of rice in cold regions, resulting in a decrease in yield. It is important to establish the identification index of the occurrence of rainy and low light conditions and quantitatively assess the yield loss for ensuring the domestic supply of high-quality rice and national food security. In this study, a multi-layer grey correlation analysis method was adopted to investigate the grey relationship among disaster factors, crop yield structure and final yield by combining meteorological, growth period, yield data and disaster historical data, and to construct the rice rainy and low light composite index (RSCI) in cold region, and establish the correlation degree model between complex adversity and yield loss rate. Based on K-mean clustering analysis method and historical typical disaster years, the critical value and grade of disaster were determined, and the evaluation index of yield loss caused by rainy and low light weather was established. The critical threshold and yield loss rate of rice with mild, moderate and severe rainfall were studied. The results of historical disaster verification showed that the identification rate of rainy and low light was 100%, and the identification accuracy rate of yield loss rate was higher than 80%. From 1958 to 2021, the frequency of rice rainy and low-light disasters in different degrees in the province was as follows: mild was higher than moderate and severe, and the frequency of rice rainy and low-light disasters in the northern agricultural area was higher than that in the southern agricultural area. In this study, the identification index of the occurrence of rain and light combination was constructed to provide technical support for quantitative evaluation of yield loss.

Key words: Heilongjiang Province, japonica rice, high-quality japonica rice, flowering period, rainy and low light combined damage, composite index, yield loss, grey relational analysis, K-means clustering analysis