次月尺度季風震盪與伴隨的極端天氣預報技術研發---子計畫:次月與季內尺度季風模對西北太平洋颱風活動的影響:綜觀指標之建立與校驗
No Thumbnail Available
Date
2010/08-2011/07
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
本研究擬在三年期間，根據第一期計畫所發展的次月尺度準雙週震盪(QBWO; Quasi-Biweekly Oscillation)與季內尺度MJO(Madden-Julian Oscillation)等兩個季 風模的綜觀氣候，進一步建立西北太平洋颱風季期間熱帶氣旋發生前至少一週大尺度 環境隨時間變化的綜觀指標。並且以再分析及再預報資料為平臺對其加以驗證，研發 利用季風模訊號進行展期颱風活動的預報技術。 為達上述目標，本子計畫規劃了下列5 大研究重點。(1)擴大個案年分析到31 年 (1979-2009)的JRA-25 與NCEP/CFSRR 再分析資料庫，檢驗同解析度對上述季風模之 綜觀氣候軌跡的影響;(2)發展變異量保守的統計迴歸模式VCCRM (Variance Conserved Canonical Regression Model)，據以建立上述季風模在不同相位時，環流暨熱力場在 特定颱風軌跡前兆期的綜觀指標;(3)根據再分析資料對上述綜觀指標進行統計交叉驗 證，評估包括軌跡在內的定量誤差;(4)發展颱風濾除技術，比較有與無颱風的環境流 場，所造成對上述低頻季風模的衝擊;即評估颱風對季風模的可能升尺度作用及其對所 建立之綜觀指標的影響;(5)分析上述季風模在NCEP/CFSRR 再預報資料庫的訊號。結合 子計畫一的颱風暖心渦旋辨識法，將嚐試對已建立之綜觀指標進行操作型驗証，提供 氣象局未來要利用NCEP 新一代的GFS 與多模式系集預報系統進行展期颱風預報時的參 考。 待上述重點4 和5 獲得一定程度的釐清後，發展的統計迴歸模式不排除在計畫後 期與多模式系集預報系統結合，實驗進行展期颱風活動統計預報。
Based on the developed synoptic climatology of two monsoon modes: the quasi-biweekly oscillation (QBWO) in the sub-monthly timescale and Madden-Julian oscillation (MJO) in the intraseasonal timescale established during the phase-1 integrated project, this 3-yr study plans to further construct and validate a temporally varying synoptic index (TSI) in the large-scale environment prior to the occurrence of tropical cycle at least one week during the typhoon season of western North Pacific. Utilizing the signals of monsoon modes, the TSI identified in both reanalysis and reforecast products will serve as the basis for developing the extended typhoon forecast technique. Five tasks will be conducted to achieve this goal. (1) Our case analysis will be extended to cover 31-yr (1979-2009) JRA-25 and NCEP/CFSRR reanalyses to examine the impacts of various resolutions on the synoptic tracks of monsoon modes. (2) A variance conserved canonical regression model (VCCRM) will be developed to help constructing the TSI for specific typhoon track types based on the changes in circulation and thermal fields associated with these two monsoonal precursors in different phases. (3) Using reanalysis data, the TSI will be statistically cross validated and the errors including the track deviation will be quantified. (4) A typhoon removal technique will be used to examine the possible upscaling effects of typhoon on the low-frequency monsoon modes and the resulting impacts on the TSI. (5) The signals of monsoon modes in the NCEP/CFSRR reforecast data will be analyzed. Combining the results of subproject 1 which will identify the modeled typhoons through the use of warm-core identification procedure, the established TSI will try to conduct an operational validation. As CWB plans to use the products of NCEP’s new generation GFS (Global Forecast System) and multi-model ensemble forecast systems in its operation, such an experiment will facilitate CWB in the development of the extended typhoon forecast. After clarifying the results in tasks 4 and 5, there exists the possibility that the developed regression model might be used as a statistical forecast model in the final stage of project, if the multi-model ensemble forecast system is ready.
Based on the developed synoptic climatology of two monsoon modes: the quasi-biweekly oscillation (QBWO) in the sub-monthly timescale and Madden-Julian oscillation (MJO) in the intraseasonal timescale established during the phase-1 integrated project, this 3-yr study plans to further construct and validate a temporally varying synoptic index (TSI) in the large-scale environment prior to the occurrence of tropical cycle at least one week during the typhoon season of western North Pacific. Utilizing the signals of monsoon modes, the TSI identified in both reanalysis and reforecast products will serve as the basis for developing the extended typhoon forecast technique. Five tasks will be conducted to achieve this goal. (1) Our case analysis will be extended to cover 31-yr (1979-2009) JRA-25 and NCEP/CFSRR reanalyses to examine the impacts of various resolutions on the synoptic tracks of monsoon modes. (2) A variance conserved canonical regression model (VCCRM) will be developed to help constructing the TSI for specific typhoon track types based on the changes in circulation and thermal fields associated with these two monsoonal precursors in different phases. (3) Using reanalysis data, the TSI will be statistically cross validated and the errors including the track deviation will be quantified. (4) A typhoon removal technique will be used to examine the possible upscaling effects of typhoon on the low-frequency monsoon modes and the resulting impacts on the TSI. (5) The signals of monsoon modes in the NCEP/CFSRR reforecast data will be analyzed. Combining the results of subproject 1 which will identify the modeled typhoons through the use of warm-core identification procedure, the established TSI will try to conduct an operational validation. As CWB plans to use the products of NCEP’s new generation GFS (Global Forecast System) and multi-model ensemble forecast systems in its operation, such an experiment will facilitate CWB in the development of the extended typhoon forecast. After clarifying the results in tasks 4 and 5, there exists the possibility that the developed regression model might be used as a statistical forecast model in the final stage of project, if the multi-model ensemble forecast system is ready.