中文摘要：

针对旱作区农业旱情等级的划分依据问题,以黑龙江省甘南县为例,建立了能够模拟时段降水量的随机模型,验证了该模型的有效性并模拟了1 000年的降水序列,提出考虑降水量和作物需水量的缺水率指标,应用模拟的降水序列,采取降水量距平百分比和缺水率两种指标分析了甘南县遭受不同等级干旱的概率,并应用提出的缺水率指标划分了甘南县的农业旱情等级。结果表明,将降水的随机模拟与缺水率指标结合起来划分甘南县农业旱情等级的方法是可行的,与甘南县的实际干旱情况吻合良好。

英文摘要：

According to the problem of how to divide agricultural drought levels,a stochastic model is established to simulate the long series rainfall data,and a method is presented based on the rainfall data.This paper verifies the effectiveness of this model and simulates a 1 000-year rainfall data of Gannan county in Heilongjiang Province,and puts forward a new drought index that gives consideration to rainfall and crop water requirement（Dw）.Dw and the percentage of rainfall anomaly（Dp） are used as examples and the probability of different agricultural drought levels in Gannan are analysed and the interval of drought level is divided.The results show that the method is valid,and a combination of Dw and stochastic rainfall model is feasible in describing the agricultural droughts in Gannan.