干旱区研究2025，Vol.42Issue(6)：957-969,13.DOI:10.13866/j.azr.2025.06.01
基于雷达和遥感卫星的新疆区域降水反演
Precipitation retrieval for Xinjiang region based on radar and remote sensing satellites
摘要
Abstract
To more accurately obtain precipitation distributions in remote areas,this study combined the high-res-olution advantages of radar and the wide-coverage detection of satellites.By integrating radar and satellite-de-rived precipitation,we generated high-precision quantitative precipitation estimation products.Using the strong convective events in Xinjiang on August 12 and 13,2023,as an example,we used radar reflectivity for precipita-tion inversion based on cloud classification and Z-R relationships.We fed the Himawari 9 satellite brightness tem-perature and IMERG precipitation into a BP neural network model to establish the relationship between the aver-age brightness temperature and the average rainfall intensity.Subsequently,we used the instantaneous brightness temperature of the Himawari 9 satellite to invert the momentary precipitation through the BP neural network mod-el.We also proposed two precipitation data fusion schemes:Scheme I uses a uniform correction value to integrate radar and satellite precipitation,whereas SchemeⅡfurther considers the precipitation intensity levels for com-parison.Finally,we obtained high-precision precipitation inversion products for Xinjiang.The results showed that:(1)Cloud classification based on brightness temperature can finely estimate precipitation within the radar range,and brightness temperature differences can reduce the impact of non-precipitating clouds to some extent.(2)The root mean square error(RMSE)of the satellite precipitation inversion was 1.793 mm·h-1,with a coeffi-cient of determination(R2)of 0.572,indicating reasonable model accuracy.The binary classification score indicat-ed that the model can accurately invert precipitation in over 70%of the areas.(3)The fusion of precipitation by the two schemes slightly improved the accuracy of short-duration light rain distributions.Scheme Ⅱ outper-formed Scheme I for short-duration moderate rain but showed a slight decline for short-duration heavy rain com-pared with Scheme I,indicating that the asynchrony between satellite observation and near-surface precipitation had some impact.(4)Under a 95%confidence interval,the P-values for the RMSE and R2 differences between the two schemes and satellite inversion were all less than 0.005,while the P-value for SchemeⅡcompared with Scheme I was greater than 0.05.Both fusion schemes significantly improved the accuracy of the satellite precipi-tation;however,the improvement of SchemeⅡ,which considers the precipitation intensity levels,over Scheme I was minimal.关键词
多雷达/Himawari 9卫星/亮温/降水反演/BP神经网络/降水数据融合/新疆Key words
multi-radar/Himawari 9 satellite/brightness temperature/precipitation retrieval/BP neural net-work/precipitation data fusion/Xinjiang引用本文复制引用
郭建茂,吴登国,韩金龙,张茹水,王勇..基于雷达和遥感卫星的新疆区域降水反演[J].干旱区研究,2025,42(6):957-969,13.基金项目
新疆维吾尔自治区重点研发任务专项"基于多源卫星的新疆水库安全监测技术研发"(2022B03027-1) （2022B03027-1）
