干旱是全球影响范围最广、 危害最重的自然灾害之一。标准化降水指数（SPI）是干旱监测业务和科研中使用最为广泛的气象干旱指数之一。目前， 中国已建设了大量高密度的区域自动气象观测站， 这些站由于缺乏长历史序列数据， 无法用于计算SPI， 如何利用这些短时间序列区域站数据开展精细化干旱监测评估是目前关注的重点。利用1960 -2020年2032个国家气象站和2010 -2020年云南省1009个区域站日降水数据， 选取31个国家站作为方法试验站， 构建了通过伽马分布参数插值法来拟合区域站的SPI（Iab）， 并与常用的邻站替代法、 多元线性回归法进行了对比， 进行交叉检验和误差分析。通过对比拟合值与真值的相关系数、 均方根误差和平均绝对误差得出， 除在中国西北站点稀疏的地区外， 拟合值Iab均明显优于其他两种方法的拟合结果。在不同季节和不同时间尺度下， 参数插值法拟合得到的SPI（Iab）效果最好， 特别是在中国中东部地区的Iab值较其他方法平均误差减小0.02~0.30； 北京、 昆明历年拟合值Iab与真值Iz误差变幅最大为0.16， 不到半个干旱等级0.25； 拟合值Iab与真值Iz的相关系数达0.999， 通过0.001的显著性检验。采用该方法进行区域站干旱过程的监测结果显示， 利用高密度区域站计算的干旱指数SPI比仅利用国家站插值得到的SPI更接近干旱灾害实况。总体而言， 伽马分布参数插值法可以用于高精度短序列区域站降水资料推算干旱指数SPI， 从而实现气象干旱精密监测、 预报和评估服务。

Drought is one of the most widespread and harmful natural disasters in the world.The standardized precipitation Index （SPI） is one of the most widely used meteorological drought indexes in the drought monitoring operation and research.At present， China has built a large number of high-density regional automatic weather stations， which can not be used to calculate the SPI due to the lack of long history sequence data.How to use these short-term data to carry out refined drought monitoring and assessment is the focus of attention.Based on the daily precipitation data of 2032 national meteorological stations from 1960 to 2020 and 1009 regional stations in Yunnan Province from 2010 to 2020， 31 national stations are selected as the method test stations.The SPI（Iab） is constructed by the Gamma distribution parameter interpolation method to fit the regional stations， compared with the commonly neighbor station substitution method and multiple linear regression method， and cross-test and error analysis were carried out.By comparing the correlation coefficient， root mean square error and average absolute error between the fitting value and the true value， it can be concluded that the fitting value Iab is significantly better than the fitting results of the other two methods except in the areas with sparse stations in Northwest China.In different seasons and different time scales， the SPI （Iab） obtained by parameter interpolation is the best.In Mid-Eastern China， the average error of Iab decreases from 0.02 to 0.30 compared with the other two methods.The maximum error change of fitting value Iab and true value Iz in Beijing and Kunming over the years is 0.16， less than half the drought grade 0.25.The correlation coefficient of Iab and the true value （Iz） is above 0.999， passed the significance test of 0.001.The monitoring results of the drought process of the regional stations using this method show that the drought index SPI calculated by the high-density regional stations is closer to the actual drought disaster than the SPI obtained by the interpolation of the national stations.In general， the gamma distribution parameter interpolation method can be used to calculate the drought index SPI from the precipitation data of high-precision short series regional stations， so as to realize the precise monitoring， prediction and evaluation services of meteorological drought.