Abstract: The 5km×5km meteorological dataset from 1981 to 2010 is used in this study to analyze the delayed chilling injury risk of rice in Northeast China and the pure premium ratemaking thereof. The availability of the grid dataset is evaluated by calculating the correlation coefficients and root mean square errors (RMSE) between the grid database of average temperature from May to September from 1981 to 2010 in Northeast China and the data observed from 99 meteorological stations. The number of days with daily average temperature stably higher than 10°C and 18°C is taken as an indicator for defining climatologically safe rice-growing areas, within which the temporal and spatial change characteristics of chilling injury of rice in Northeast China are analyzed to determine the pure premium rate thereof. The results show a strong correlation coefficient and low RMSEs between the grid data and the observed May to September mean temperatures in Northeast China from 1981 to 2010, indicating that grid data is available in Northeast China. The climatologically safe rice-growing areas, which occupied 56.5% of Northeast China, are mainly distributed in Liaoning, central and western Jilin, southwestern and northeastern Heilongjiang, the western region in eastern Inner Mongolia and the eastern regions bordering Liaoning and Jilin. In climatologically safe rice growing areas, the frequency of delayed chilling injury in rice is typically characterized by a distribution pattern of low in the south and higher in the north, and low in the middle and elevated in the east and west. Moreover, severe delayed chilling injuries were the most frequent. The spatial distribution of the risk index of delayed chilling injury is similar to the distribution of the frequency, with high value areas in the western and northeastern parts of Inner Mongolia, northern Heilongjiang, and western Jilin. The spatial distribution of the pure premium rate of delayed chilling injury for rice in Northeast China from 1981 to 2010 is similar to its risk index distribution, i.e., low in the south, high in the north, low in the center, and high in the east and west. The pure premium rate is between 0.010 and 0.094, which can be regarded as a reference for insurance companies in making specific rates.

Key words: Delayed chilling injury, Agrometeorological hazards, Pure premium rate, Root mean square error, Climatologically safe rice-growing areas