多平台(雷达、卫星、雨量计)降水信息的融合技术初探  

Preliminary Studies on Merged Techniques Based on Precipitation Information from Multiplatform(Radar,Satellite and Rain Gauge)

作　　者：高晓荣[1,2] 梁建茵[2] 李春晖[2] 张维[1] 

机构地区：[1]广州中心气象台,广东广州510080 [2]中国气象局广州热带海洋气象研究所,广东广州510080

出　　处：《高原气象》2013年第2期549-555,共7页Plateau Meteorology

基　　金：广东省科技厅项目(2007B030401007);中国气象局气象新技术推广项目"雨量计和多平台遥感降水资料的融合技术"(CMATG2008Z11)共同资助

摘　　要：为提高卫星降水估计的精度,利用CMORPH资料和雨量计资料,基于最优插值法OI(OptimumInterpolation)有效合成,得到0.125°×0.125°降水分析产品;基于广东省境内5部新一代多普勒天气雷达(CINRAD/SA)的降水估计拼接产品,即0.01°×0.01°降水分析产品。为进一步扩展降水的探测范围,将以上两种不同分辨率产品进行区域定量协调,并以各自均方根误差平方的倒数作为权重系数将两者融合。误差分析表明,前期雨量计订正的CMORPH能有效减小系统偏差;交叉检验表明,CMORPH融合了雨量计资料后,比单纯OI雨量计的效果要好。两种不同分辨率产品有明显的区域差异特征;获取的一套降水产品既能体现雷达资料的优势,又能体现CMORPH融合雨量计的优势,在降水估测精度和扩大降水估测范围上都有了较好改观,能够反映不同尺度平台的降水信息。To improve satellite Quantitative Precipitation Estimation （QPE） techniques, combining CMORPH precipitation data and rain gauge observations based on OI （Optimal Interpolation）, the QPE with 0. 125°×0. 125° resolutions was obtained. In order to enlarge the scale of precipitation, the CMORPH QPE with 0. 125°×0. 125° resolution and multi-radar QPE with of 0.01°×0.01°resolutions were merged by weight coefficient of square multiplicative inverse of its root mean square error. Error statistical analysis indicated that using rain gauge data to correct CMORPH data can reduce bias cross-validation, some conclusions were obtained that the satellite QPE merging with rain gauge data is better than gauge＇s OI. QPE output with different resolutions has obvious regional difference. Merging is an effective way to improve the precision and enlarge the scale of QPE from multiple sources of precipitation information and applied to take advantage of precipitation information from multiple platform.

关 键 词：定量降水估计 融合技术 交叉检验 

分 类 号：P413[天文地球—大气科学及气象学]