中文摘要：

在模拟逐日降水数据方面,近些年来的一种通用方法是联合应用一阶马尔科夫链和伽玛分布函数建立随机模拟模型,这种方法在国外很大的环境范围内被证明是有效的。为此,应用黑龙江省14个站点56年的实测降水数据,针对马尔科夫两状态转移概率和伽玛分布参数的不同来源,即统计分析实测降水数据直接获得和利用实测降水数据建立回归模型间接推求,分别模拟了各站点的1 000年逐日降水过程。结果表明：应用一阶马尔科夫链和伽玛分布函数可以很好地模拟黑龙江省14个站点的逐日降水过程,模型的模拟精度很高,证明了这种方法是有效的;但国外推求转移概率和伽玛分布参数的经验模型和长系列法在黑龙江省不完全适用,应用重新建立的回归模型推求的转移概率和伽玛分布参数模拟的逐日降水数据的精度同样很高。

英文摘要：

In modeling and simulation daily rainfall,a combination of Markov chain and gamma distribution function has been recognized as a particularly popular method and demonstrated to be effective in generating daily rainfall data for large range of environment abroad in recent years.According to different data sources of the Markov transitional probabilities and gamma distribution parameters,two plans were used to simulate the daily rainfall process based on 56-year daily rainfall data of 14 stations in Heilongjiang Province.The research results show that the method can be used to generate daily rainfall data in Heilongjiang province based on the model parameters estimated with historical rainfall records and its accuracy is high,but the empirical formula and the long method presented abroad are not applicable.The regression model re-established with measured data of 14 stations in Heilongjiang province can well decide the transitional probabilities and gamma distribution parameters,and it could generate satisfactory rainfall data with good accuracy as well.