中文摘要：

在温度控制试验和文献查阅的基础上，分析了水稻开花期开花概率密度分布、日开花分布规律以及结实率的变化规律，提取了高温对水稻结实的影响因子。开花概率密度随开花历期呈Possion分布，日开花结实率可用二次方程描述，日开花量呈正态分布，高温对开花结实率的影响可用S型曲线描述。进一步结合气温日变化规律，并综合考虑高温对已开和未开颖花的影响，构建了水稻高温败育模拟模型。利用独立的试验资料对模型进行了检验。结果表明，当温度低于40℃时，模拟值与实测值有较好的一致性；当温度高于40℃时，因高温对开放颖花的结实率仍有影响，模拟值与实测值有一定误差。总体上，模拟值与实测值的均方根差（RMSE）为0．12。由于我国大部分地区水稻开花期的日极端最高气温较少超过40℃，因此所建立的模型可较好地模拟高温对水稻结实率的影响。

英文摘要：

Temperature is one of the most important determinants for rice yield, particularly when a high temperature event occurs during flowering period. Under the background of global warming, frequent high temperature event during rice flowering period often leads to a higher sterility of rice. In order to quantify the effects of high temperature on the sterility of rice crops, the control experiments at different high temperature （35, 37, 39 and 41℃） and with various durations （1, 3 and 5 d） in two cultivars were conducted during rice flowering period in 2004, and quantitative characteristics of rice flowering was cited from the literatures. The results showed that the daily probability density of flowering amount could he expressed as Possion distribution function, the daily fertility could he expressed in a quadratic equation, the flowering number in one day obeyed the normal distribution, and the relationship between temperature and spikelet fertility could be approximated by logistic curve. Based on the daily variation law of air temperature and the effect of high temperature on both opening and unopened spikelets, a model was developed to simulate high temperature-induced sterility. The model was validated with independent experimental data and the results showed that it could simulate the effects of high temperature on rice fertility well while tem- perature was below 40℃, and a larger error resulted from the excessive temperature （40℃） which would affect the fertility of those opened spikelets. Considering the daily maximal temperature over 40℃ is very few in open fields, the model developed is applicable for most of rice producing regions.