未來台灣將由S與C波段雙偏極化雷達組成降雨雷達網，而雙偏極化雷達參數（如Kdp、Zdr)會 因電磁波波長差異而有不同的特性，進而影響其應用於降雨估計的表現。為促進各波段雷達定量降雨估 計產品的整合，本研究利用國立中央大學二维雨滴譜儀資料，依月份針對春雨、梅雨、午後對流、颱風 以及冷鋒鋒面等降雨類型進行分類，並透過T-Matnx散射模擬雙偏極化雷達參數，求得不同波段雷達各 降雨類型之降雨估計關係式：R(Z)、R(Kdp)、R(Z，Zdr)及R(Kdp，Zdr)。本研究評估資料處理程序與各類 降雨估計方法，分析其對S與C波段雙偏極化雷達降雨估計的影響。研究個案選自2014至2015年由 氣象局S波段與中央大學C波段雷達共同觀測之7組個案累計15小時，透過地面雨量站資料校驗雷達 降雨估計的表現。分析結果顯示：（一)套用對應降雨類型公式可改善雷達降雨估計的表現；（二)利用〇dp 修正能量衰減、自洽法(Self-consistency)修正Z偏差量與濕天線罩效應的衰減量，皆能明顯改善S與C 波段雷達降雨估計的表現；（三)兩雷達降雨估計皆以Kdp參數之組合型公式(Hybrid)有最佳表現；（四)當 雷達資料樣本數相近時，短波段雷達其Kdp參數於降雨估計略具優勢，有較低的相對均方根差與較高的 相關係數。

The operational radar network in Taiwan will be composed with S and C-band dual-polarimetric radars to monitor extreme precipitation. Various polarimetric parameters of different wavelength radars have distinct advantages and disadvantages in term of quantitative precipitation estimation (QPE). Four different QPE relationships, namely R(Z), R(Kdp), R(Z,Zdr) and R(Kdp,Zdr), of each rain type are obtained from National Central University 2D-Video disdrometer data and applied to seven events observed by both RCWF S-band and NCU C-Pol radars from 2014 to 2015. The goal of this study is to investigate the performances of radar-based QPE from different wavelengths by comparing with 96 rain gauges. The results indicate that: (1) The QPEs are improved after applied into the corresponding rain-type coefficients with respect to general ones. (2) Both the attenuation correction by 〇dp algorithm and the wet radome correction by self-consistency method greatly improve the QPEs from RCWF and NCU C-Pol radars. (3) The QPEs from KDP-based relationships which combine with R(Z) are more accurate than Z-based relationships. (4) When two radars have comparable sampling data, NCU C-Pol radar shows the advantage of Kdp parameter for QPE at shorter wavelength with lower relative root mean square error and higher correlation coefficient.