美國國家大氣研究中心(NCAR)所發展之都卜勒雷達變分分析系統(VDRAS)，主要使用四維變分方法同化都卜勒雷達觀測資料，可以提供高時間、空間解析度之對流尺度大氣分析場。但是，由於海洋上之觀測資料缺乏，前人研究皆說明VDRAS運用於台灣島鄰近區域之分析頗具挑戰性。然而，衛星資料能提供海洋上大範圍的觀測與反演資料，例如ASCAT產品可以提供海表面10米風場資料，且不受雲雨區屏蔽。故本篇研究希望利用VDRAS同化ASCAT風場反演資料，改善海洋上觀測資料不足的問題，進而提升其對於對流系統之分析能力。 研究中使用已建置具地形解析能力之VDRAS版本（IBM_VDRAS），在此基礎上並改善前人使用之地面資料同化方法。其中，除了ASCAT風場資料，陸地上之地面站風場也一併同化。用以測試及分析之個案為2013年6月23日13 UTC於台灣西南沿海生成之颮線系統。在探討不同地面資料同化方法之實驗中發現，改善後的地面資料同化方法能避免分析場中觀測資料周圍梯度過大之情形，並能將觀測資訊合理散佈於三維空間。接著，吾人利用此改善之地面資料同化方法，進行各種資料同化效益實驗。研究結果指出，尚未加入地面資料同化時，VDRAS無法分析出颮線系統底層陸風及盛行西南風輻合之特徵。在同化ASCAT與地面站風場後，對於熱力、動力與雲微物理結構都有顯著改變。從垂直結構分析上，底層資料也合理反應至三維空間中。最後，與觀測資料之驗證結果也證明，同化ASCAT風場資料能顯著改善原本低估的西南風場，結合陸地與海上資料的同化後，更能提升三維空間雨水混合比的同化效果。 ;The Variational Doppler Radar Analysis System (VDRAS) developed by National Center for Atmospheric Research (NCAR) is able to assimilate Doppler radar observations by using 4DVar method, and to provide high temporal and spatial meteorological states in convective scale. Due to lack of sea surface observation, it has been challenging to generate realistic analysis by using VDRAS in Taiwan and vicinity. As we know, satellite can measure in wide range on the ocean. For example, ASCAT (Advanced SCATterometer) data can be retrieved as product of sea surface 10 meters wind data, which avoids contamination caused by hydrometeors in the atmosphere. Therefore, this study is aimed to improve the accuracy of analysis by additionally assimilating satellite data by using VDRAS. An improved surface data assimilation scheme has also been developed in updated version of VDRAS that was implemented with a terrain-resolving scheme (IBM_VDRAS). Note that horizontal wind data measured by mesonet and retrieved by the ASCAT are both processed and assimilated. A real case of squall line observed at 13UTC of June 23, 2013 near offshore of southwestern Taiwan is selected in current study. Test on different surface assimilation schemes is conducted first, which indicates that, compare to previous scheme, modified surface data assimilation scheme can generate smoother and more reasonable analysis. With the updated surface data assimilation scheme, a series of experiments is then conducted to investigate the impacts of assimilation of different kinds of observational data. It reveals that the low-level wind convergence associated with the squall line is successfully analyzed with additional surface data assimilation. The enhanced prevailing southwesterly flow and seaward flow are clearly shown, which even modifies analyzed thermodynamic, dynamical and microphysical structures of the squall line. In addition, evaluations also demonstrate that, with merging of surface data, significant correction of underestimated southwesterly flow on the ocean and better minimization of rainwater mixing ratio are present.
