| 摘要: | 熱帶氣旋是個具有強烈的旋轉氣流且伴隨著旺盛對流雲系的暖心低壓系統，而由於位渦這個物理量同時結合了動力以及熱力因子，且在絕熱無摩擦的條件下具有保守的特性，故用它來分析熱帶氣旋變化會是個相當好的物理量。近年來有不少與熱帶氣旋位渦相關的研究，但採用位渦收支(Potential Vorticity Budget)對實際颱風個案進行診斷分析的研究並不多。因此，分析位渦收支可以解析出颱風登陸過程中何種動力及熱力過程之演變，而臺灣的陡峭地形又會對位渦收支造成何種影響，是值得探究的課題。 本研究參考Pedlosky(1987)及Schubert et al.(2001)，得出一非靜力可壓縮之位渦收支方程式，針對Yang et al.(2008)利用PSU/NCAR MM5數值模式模擬納莉(2001)颱風所輸出之高時間、空間解析度的模式輸出，進行位渦診斷分析，來探討納莉颱風從海上到登陸臺灣過程中的位渦收支之變化情形。隨後修改臺灣地形高度，分別透過(1)移除臺灣地形的敏感度實驗，以及(2)進一步將地面替換為純海洋的敏感度實驗，來討論臺灣地形及陸地對位渦收支的影響。 從控制組實驗的模擬結果來看，納莉颱風在尚未登陸前其位渦分佈呈現一典型成熟熱帶氣旋的分佈型態，颱風登陸瞬間由於地形的影響使其對流增強及眼牆收縮。在位渦收支分析的結果中，水平平流作用會將位渦順著颱風環流傳遞，並隨著徑向入流將位渦帶入颱風內核區域，垂直平流會將於低層經由潛熱作用所生成的位渦向上傳遞，平流作用僅扮演著於三度空間中使位渦重新分佈的角色。潛熱作用項在低層中是位渦的主要貢獻項，而在中高層則是位渦的主要消耗項。摩擦項之影響僅在低層較為顯著，但由於此項中包含了亂流混合以及邊界層摩擦力的作用，故並不一定都是造成位渦的負貢獻。 在地形敏感度的實驗結果中，若將臺灣地形移除，則颱風在登陸後摩擦作用項於近地面會產生明顯的負貢獻，顯示地表摩擦力的影響，同時在該處潛熱加熱項也會使位渦生成；與控制實驗相比，臺灣地形的存在會使摩擦項影響較大，且並不一定都是對位渦造成負貢獻，而由潛熱加熱項所生成的位渦也較多，故可得知臺灣地形的存在對於摩擦作用項來說，可能會加強低層的亂流混合，從而導致位渦的增加。此外，臺灣地形及地表摩擦力的存在都能夠促使對流發展，進而使位渦增強。在此無臺灣地形的免感度實驗中，隨著颱風繼續進入陸地，低層的位渦環會逐漸消失，相較於純海洋實驗中則沒有此現象，由此我們可以發現海表面通量是維持颱風內核位渦環存在的重要要素之一。當颱風離開臺灣再次進入海洋時，不對稱的潛熱加熱作用會使颱風重新形成一多邊形的位渦環，並可能使颱風路徑出現小幅度的擺線震盪。 Potential Vorticity (PV) is a quantity that combines both dynamic and thermodynamic information, and it is conserved under the adiabatic and frictionless condition. Therefore it is suitable for analyzing tropical cyclones’ (TC) dynamic. In recent years, there have been considerable studies between TCs and the PV. However, it appears that only a few studies using the potential vorticity budget to analyze a real TC case. Thus, what kind of dynamical / thermodynamical interaction processes could be obtained from the PV budget analysis, and what phenomena will be caused by the steep topography of Taiwan are worthy of study. A compressible nonhydrostatic PV budget equation, derived base on Pedlosky (1987) and Schubert et al. (2001), is used here to gain insights into the PV budget evolution of a typhoon from its oceanic stage to landfall stage. The budget is conducted using high spatial resolution (2-km horizontal grid size) hourly outputs from Yang et al. (2008), in which the Pennsylvania State University / National Center for Atmospheric Research Fifth Generation Mesoscale Model was used to simulate Typhoon Nari (2001) and reproduced reasonably-well results as verified against observations. Subsequently, a series of terrain-sensitivity tests were performed to examine the effect of Taiwan’s topography on PV budget. When Nari was located on the ocean, its PV distribution exhibited the typical feature in a mature oceanic TC. By the time of landfall, its eyewall was contracted and convection was intensified due to the topography. From the budget perspective, PV was redistributed via horizontal and vertical advections. Latent heating term accounted for major PV generation in lower levels during the oceanic and early landfall stage. And it also acts as a major PV sink term at mid-upper levels. The friction term included both effects of eddy mixing and surface friction; hence, it did not just act as a PV sink term. In the terrain-sensitivity experiments, if the Taiwan topography was removed, the friction term began to reduce PV over the Taiwan area in lower levels, which was opposed to that for the full-terrain run. As a result, the existence of Taiwan topography could enhance vertical eddy mixing. When comparing latent heating term and friction term, it is evident that both the Taiwan topography and surface friction are prone to trigger convection, releasing more latent heat and leading to the increase of PV. And the cut-off of ocean fluxes such as sensible heat and latent heat flux will cause the dissipation of the PV ring. In the no-terrain experiment, after the typhoon moves into the ocean again, a larger new PV ring formed. The asymmetry latent heating effect occurred on the land-sea interface not only contributed to the formation of this new PV ring, but make this new PV ring became polygonal as well. This phenomenon may also be a reason that causes the typhoon move in a trochoidal manner afterward. |
