| 摘要: | 本計畫擬利用變分方法同化目前可資利用之遙測資料（包括傳統衛星輻射值、亮度溫度、近地風場，GPS ground-based 及掩星衛星資料等），選定歷史颱風（如Chanchu, Sepat, Shanshan, Krosa 等）及未來颱風進行資料同化對颱風模擬（強度、路徑及局地降水）影響之研究，同時使用位渦反演技術，探討颱風準平衡動力結構及侵台前受地形作用或於鋒前受鋒面環流影響之變化特性，最終為發展一實用的虛擬位渦渦旋同化方法，俾以改善颱風初始化問題。本研究所使用的位渦反演方法，可以定出一時間平均的軸對稱渦旋及非軸對稱分量，同時由片段位渦切割，可以探討某片段位渦對平衡流場的貢獻，最後透過求解 方程及連續方程，可以將輻散風場（含垂直速度）求出，可以提供完整的三維環流結構（風場，壓力場及溫度場）。本計畫為期三年，資料同化研究主要以WRF 三維變分(3DVAR)導入QuikSCAT 衛星觀測近地風場、SSM/I，微波衛星（Advanced Microwave Sensing Unit）AMSU 觀測反演之三維風場與溫濕度場、GPS ground-based 可降水量、CHAMP 及FORMOSAT-3/COSMIC 等GPS 掩星觀測之折射率，及總計畫飛機觀測之dropsonde 探空資料，探討各該同化資料對侵台颱風數值模擬與風雨預報改善之影響。第一年（97 年度）著重於使用WRF 及WRF 3DVAR 進行個案同化模擬及敏感度實驗，同時針對歷史颱風及未來颱風個案進行位渦反演分析。第二年（98 年度）除延續第一年工作，亦將進行WRF cycling 3DVAR 研究，探討颱風spin-up 及bogussing 等問題，同時針對位渦反演之颱風進行個案動力分析，探討絕熱及非絕熱作用對準平衡流場之影響，據此提出適當的準平衡位渦渦旋方法(簡稱PVV- potential-vorticity vortex)，並與梯度風平衡的對稱渦旋方法(簡稱GWV- gradient-wind vortex)及BDA (bogus data assimilation) 的非對稱渦旋(簡稱PEV- primitive-equation vortex)方法比較。最後一年（99 年度）除評估上述颱風虛擬渦旋同化方法(PVV、GWV and PEV)之成效，亦將透過片段位渦反演分析，進一步分析個別位渦區於颱風侵台前及接近登陸時的貢獻以了解主要的動力過程及可能機制。 This project will utilize 3DVAR to assimilate remote sensing data, now available including conventional satellite radiance, bright temperature, near-surface wind, GPS ground-based and occultation observations, etc. With a duration of 3 years, the main goals will be the assimilation using WRF/MM5 3DVAR for QuikSCAT satellite near-surface wind, AMSU-retrieved 3D wind or moisture/temperature, GPS ground-based PW, CHAMP and FORMOSAT-3/COSMIC occultation refractivity or phase excesses and the dropsonde data from the experiments conducted by the general project, to investigate the impact of the assimilated data on the prediction of wind and rainfall associated with typhoons impinging Taiwan. Potential vorticity inversion will be performed to investigate quasi-balanced state of the typhoon vortex, with application to vortex initialization. The method of this PV inversion may provide a complete set of meteorological fields (including wind, pressure and temperature). In the first year, we will focus on sensitivity tests WRF 3DVAR with different data assimilated and conduct PV inversion for past and future typhoons. In the second year, we will conduct cycling 3DVAR to investigate typhoon spin-up and bogussing and explore the dynamics from PV inversion for diabatic and adiabatic processes on balanced flow. Based on the results, we will propose three different initialization schemes for the vortex as PPV (potential-vorticity vortex) from the PV inversion, GWV (gradient-wind vortex) from the gradient-wind equation, and PEV (primitive-equation vortex) from bogus data assimilation (BDA). In the last year, the above three schemes (PVV, GWV and PEV) will be compared in typhoon simulation. In addition, piecewise PV inversion will be conducted to analyze the contribution of each specific PV in the typhoon vortex for typhoons before landfall and nearing landfall. 研究期間：9908 ~ 10007 |
