本研究採用系集法(Ensemble-based method)，首次在台灣藉由天氣個案模擬，針對 對流尺度之預報誤差結構進行分析，個案選取 2008 年西南氣流聯合觀測實驗 (SoWMEX IOP8)期間，於 06 月 15 至 16 日之間生成之中尺度對流系統(Mesoscale Convective Systems, MCSs)，使用 WRF 單向巢狀網格，高解析度(3-km)之網格涵蓋台 灣本島與台灣海峽，以及部分巴士海峽。以 72 組系集預報結果作為樣本，運用統計方 法估算出背景(預報)誤差協方差(Background Error Covariance)，其提供我們瞭解系集卡 爾曼濾波器(EnKF)，在同化應用上之索引，藉此推斷觀測資料在同化期間，資訊導入 與傳遞的情況。 在 MCSs 影響期間，比較不同解析度(9, 3-km)之誤差結構，顯示高解析度的方差 量值較大，特別是垂直速度有最顯著的差異，高解析度能夠顯現較小尺度的不確定性。 在方差的時間序列中，方差隨著降雨率的強度增強而增加，降雨率減弱而趨緩。水平 風的方差結構中，擁有多重尺度作用的複雜結構，較小尺度的不確定性是由濕對流過 程(moist processes)產生；此外，降雨過程產生的冷池影響近地表溫度的方差。本研究 同時探討時間與空間上的誤差相關性，其亦受到濕對流過程影響。在交相關的部分， 溫度與垂直速度之正相關結構與潛熱釋放有關，低層之水平風 U 與 V，在台灣西南部 出現負相關之特徵，為西南氣流與地形交互作用產生。本研究對於預報誤差結構的評 估，期望能夠提供資料同化有利的策略，進而改進台灣地區雷達資料同化能力。;This study focus on the short-term forecast error structures with different resolution (9/ 3 km) at meso- and convective scales. With a set of 72-member ensemble forecasts by Weather Research and Forecasting (WRF) model, the error covariances are presented. A case study during Southwest Monsoon Experiment intensive observing period 8 (SoWMEX-IOP8) in 2008 is investigated. The characteristics of forecast error covariances are examined by the spread and error correlation in state variables. Compared with different resolution, the variance of state variables are larger in higher resolution, particularly in vertical velocity. It indicates higher resolution run (3-km) can better-represent the smaller scale uncertainties in this severe weather event. The time-series of ensemble spread reveal that significant variances are associated with the strength of rainfall rate. In the magnitude of horizontal wind, multi-scale interactions are found over the southwesterly flow region. The temperature near surface has relatively large quantity of variance in association with cold pool performances. Moist processes not only impact on the distribution of variance, but paly important role of error correlation in temporal and spatial structure. The cross correlation of the temperature and vertical velocity is strongly positive at high level dominated by latent heat release. The negative cross correlation between zonal and meridional wind over southwest quadrant of Taiwan illustrates the range affected by orography. The information of forecast error provides optimal strategies of data assimilation, especially for assimilating radar network over Taiwan area.
