云的可降水概率遥感分析及在气象干旱中的应用

There are some limitations in current drought monitoring most of which are based on visible light. This application aims to increase the utilization rate and take good advantage of high temporal resolution, of remote sensing data, conducting cloud classification of FY-2 data based on object-oriented classification method, using TRMM radar precipitation data to analyze the precipitation probability of different types of cloud , resulting in a muti-stage precipitation probability function, getting a truncation level based on the theory of run, extracting three index: maximum continuous precipitation index, maximum continuous precipitation-free index, precipitation and precipitation-free ratio from time series cloud classification data and precipitation probability function, establishing CRGD based on cloud classification information and precipitation information. Moreover, spatial-temporal correction functions will be established based on the analysis of the model’s spatial-temporal applicability and the accuracy assessment will be conducted through the comparison to Palmaer drought index based on observed data like precipitation, air temperature, etc. Finally, the stability of CRGD model will be analyzed based on the statistical data of typical drought events.

现有的干旱遥感监测方法主要依赖于可见光、遥感，存在一定的局限性。本申请为了提高遥感数据利用率及发挥遥感数据的高时间分辨率特性，利用面向对象的方法进行FY-2数据云分类，基于模糊数学理论利用同步TRMM雷达降水数据分析不同类型云的可降水概率，形成分阶可降水概率函数；基于游程理论确定截断水平，从时序云分类影像及可降水概率函数分布中提取最大连续可降水指数、最大连续无降水指数、可降水比指数等，结合TRMM月产品指数，构建基于云分类信息及降水信息的气象干旱遥感监测模型（CRGD）；分析模型的时空适应性，提炼时空修正函数，利用降水、气温等实测数据计算Palmer干旱指数，比较两种干旱指数，评定CRGD模型准确性；利用典型干旱事件的统计资料，评价CRGD模型的稳定性。

现有的干旱遥感监测方法主要依赖于可见光、遥感，存在一定的局限性。本申请为了提高遥感数据利用率及发挥遥感数据的高时间分辨率特性，利用面向对象的方法进行FY-2数据云分类，基于模糊数学理论利用同步TRMM雷达降水数据分析不同类型云的可降水概率，形成分阶可降水概率函数；基于游程理论确定截断水平，从时序云分类影像及可降水概率函数分布中提取最大连续可降水指数、最大连续无降水指数、可降水比指数等，结合TRMM月产品指数，构建基于云分类信息及降水信息的气象干旱遥感监测模型（CRGD）；分析模型的时空适应性，提炼时空修正函数，利用降水、气温等实测数据评定CRGD模型准确性；利用典型干旱事件的统计资料，评价CRGD模型的稳定性。