查詢結果分析
來源資料
頁籤選單縮合
題 名 | 衛星微波資料對降水型態分類及其應用=Using TRMM Microwave Data to Classifing Rain Types and Applications |
---|---|
作 者 | 簡宏彬; 陳萬金; 劉振榮; | 書刊名 | 氣象預報與分析 |
卷 期 | 192 2007.09[民96.09] |
頁 次 | 頁29-39 |
分類號 | 328.63 |
關鍵詞 | 多頻道微波資料; 降雨型態; 貝氏機率法; |
語 文 | 中文(Chinese) |
中文摘要 | 本研究係利用Tropical Rainfall Measuring Mission (TRMM)衛星上TRMM Microwave Imager (TMI)多頻道微波資料,辨識臺灣空域海上劇烈天氣系統中對流及層狀兩種降雨型態(Rain Type)後,再估算其定量降水;首先以貝氏機率法進行降雨型態的辨識,再以相匹配的Precipitation Radar (PR)近地面降雨率(PR, Rain Rate)與衛星微波亮度溫度(Tb, Brightness Temperature)分別建立迴歸關係式(PR-Tb),最後與PR所辨識的降雨型態及近地面降雨率進行驗證。 PR的標準產品可提供對流及層狀降雨型態(2A23)及近地面降雨率(2A25),然其掃描範圍較窄(220 km),本研究目的,即利用掃描範圍較寬(760 km)之TMI多頻道微波資料進行降雨型態辨識,以擴大應用範圍,並提升定量降雨反演能力。由1998~2002年TMI與PR所匹配的統計資料分析發現,(T19v-T37v)、(T85v-T85h)及(T85v+T85h)/2等三項微波頻道特徵具有較好的降雨型態辨識能力;以2003~2004年驗證結果顯示,梅雨鋒面在對流、層狀及整體的成功識率分別為48%、97%及92%,颱風則為43%、97%及94%。 此外,對流降雨型態中觀測到「冰態降水」的個案資料發現,其頻道特性與輻射傳遞模擬的結果有顯著的差異,因此在迴歸微波降雨估算式時須分開處理;定量降雨估算驗證方面,梅雨鋒面及颱風之相關係數分別為0.87及0.86,均方根誤差則分別為2.64mm/hr及2.6mm/hr。 |
英文摘要 | The study is to distinguish rain types into convective and stratiform for severe weather system over ocean in Taiwan area using TRMM Microwave Imager onboard Tropical Rainfall Measuring Mission Satellite. After classifing rain types, we estimate their quantitative precipitation. The procedure is at first to classify rain types by using Bayes' probability methodology and then derive the regressionship betweenn PR rainfall rate at near surface and satellite brightness temperatures. Finally, we provide validation comparisons with rain types and near sureface rainfall rate of PR. PR has standardized products to provide different rain types (2A23) and rainfall rate at near surface (2A25), but the swath is relatively narrow (220 km). The goal of this research is to broaden the swath with classification of rain types and to improve the ability of quantitative rainfall rate estimation with TMI multiple microwave data (760 km). Through analyzed statistical data, make-ups by TMI and PR from 1998 to 2002, we can found three feature of channels, (T19v-T37h), (T85v-T85h) and (T85v+T85h)/2), to have ability of distinguishing rain types. The result of successful classification of convective, stratiform and overall are 48%, 97% and 92% for Mei-Yu fronts, and 43%, 97% and 94% for Typhoons, respectively. Furthermore, “Ice hydrometeor precipitation” in convective rain type has been detected in this study. There is quite obvious difference between those properties on microwave channels and the result of radiation transfer simulation. It must be deal with separately while we create microwave rainfall rate regression equations. The coefficient of correlation were 0.87 and 0.86 for Mei-Yu front and Typhoon respectively, and the Root-Mean-Square were 2.64 mm/hr and 2.6 mm/hr between estimatied quantitative rainfall rate and PR rainfall rate for oceanic validation. |
本系統中英文摘要資訊取自各篇刊載內容。