Research on precipitation estimation method based on Himawari-8 data

• 1、School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731
• 2、School of Mathematical Sciences, University of Electronic Science and Technology of China, Chengdu 611731

Abstract：Precipitation is an important part of the earth\'s water cycle system and has important meteorological, climatological, and hydrological significance. Therefore, accurate precipitation retrieval has always been a significant scientific research. This paper selects Himawari-8 satellite image data and ground precipitation data, adopts random forest method and least square method to fit and establish a precipitation inversion model, and conducts precipitation estimation research in Dongchuan County, Yunnan. The research results show that the relationship between cloud top brightness temperature and rain intensity can be fitted to an exponential function relationship, and it decreases monotonously. The application of random forest algorithm can greatly improve the inversion accuracy, and the inversion effect is better for moderate to heavy rain. The precipitation inversion model established in this paper has a high accuracy rate and has reduced the rate of unannounced precipitation. To a certain extent, the problem of unannounced and error is solved.

Keywords： Himawari-8 Satellite Random Forest Least Squares Method Quantitative Precipitation Estimation