全球暖化下西北太平洋颱風負回饋機制變化之探討

Abstract

颱風是自然界中重要的系統之一，而海洋在颱風生成期以及發展期扮演著關鍵能量供給的角色，因此了解颱風及海洋的交互作用顯得相當重要。根據Emanuel, K.A.,(1999), Bender and Ginis(2000), Lin et al.(2005);(2009a,b);(2013)，可得知颱風不僅與表層海水，亦和整個次表層海水有交互作用。當交互作用產生時，海洋表層海水會和次表層較冷的海水混合，產生降溫作用（Ocean cooling effect），因而減少海洋供給颱風的能量，更進一步地抑制颱風強度，此為相當重要的颱風海洋負回饋機制。 本研究將探討在全球暖化下，颱風所引起的海洋降溫作用之變化。此降溫作用受到四個因子所控制，包含海洋的初始場、颱風移動速度、颱風大小以及颱風風速。本研究將針對海洋的初始場和颱風移動速度在暖化之下的變化情形，進一步地討論其對颱風引起的海洋降溫作用所造成的變化。 本研究利用第五期耦合模式比對計畫（CMIP5）在暖化情境RCP8.5下，探討西北太平洋颱風經過區域，海洋次表層垂直熱力結構變化及颱風移度速度變化，使用Camargo(2013)及Zhao et al.(2009)的颱風資訊作分析。研究中使用十四個CMIP5模式討論二十一世紀末(2091-2100)與二十一世紀初(2006-2015)颱風經過區，海洋垂直熱力結構變化。發現颱風經過區海洋表層暖化程度較次表層多，此暖化差異與Huang et al. (2015)之結果相同，但海洋次表層溫度梯度更大，代表考慮颱風經過區域，海洋次表層垂直溫度梯度將變得更大。本研究亦發現西北太平洋颱風路徑有向北偏移的趨勢，此偏移趨勢與HiRAM結果一致。向北偏移原因可能和颱風駛流場、生成位置改變有關，北移的主因需做更多變量分析才能下定論。 本研究分析十四個CMIP5模式及高解析度模式HiRAM的颱風移動速度變化，發現在二十一世紀間，西北太平洋颱風移動速度沒有明顯的趨勢存在，同時印證Kim,H.-S. et al.(2014)使用單一模式之結果。本研究利用環境風場、地區性信風及副熱帶高壓變化情形，佐證本研究颱風移動速度無趨勢的結果。綜合本研究兩個颱風所引起的海洋降溫作用之控制因素變化，在暖化下颱風所引起的海洋降溫作用可能會增強，此推測的可信度需要做更多的模式分析才能下結論。

Tropical cyclones(TC) are common in many regions of the world and affect nearly all tropical areas. When TC passing by ocean, TC will not only interact with surface ocean but interact with subsurface ocean.(Emanuel, K.A., 1999, Bender and Ginis, 2000, Lin et al., 2005; 2009a; 2009b; 2013 etc.)Cooling of the upper ocean by TC-induced mixing which is so-called ocean cooling effect is an important negative feedback that impacts TC intensity. It is very important to understand the interaction between TC and ocean. This research objective of this study is to investigate the change of the ocean cooling effect in the Western North Pacific Ocean under global warming. The ocean cooling effect is controlled by four factors, including the initial vertical temperature profile of the ocean, TC translation speed, TC size and TC wind speed. This study focus on the climate change of first two control factors of the ocean cooling effect, using TC track data from fourteen models output of the Coupled Model Intercomparison Project Phase 5 (CMIP5)(Camargo, 2013) and HiRAM model (Zhao et al., 2009). CMIP5 models used in this research shows that the initial vertical temperature profile of the subsurface ocean can be sharpened along TC track for the Western North Pacific Ocean .This study shows the same results as Huang et al. (2015), but have bigger gradient of the initial vertical temperature profile of the subsurface ocean. The CMIP5 models and HiRAM model show that TC track poleward shift in the Western North Pacific Ocean under global warming. This may be related to the change of steering flow or TC genesis position, this part need more parameters analysis in the future work. This study shows that TC translation speed have no significant trends in both CMIP5 models and HiRAM model in the Western North Pacific Ocean. Just like Kim, H.-S. et al. (2014) using GFDL CM2.5 model shows that TC translation speed have no significant trend under RCP8.5 scenario in the Western North Pacific Ocean. The trend of TC translation speed is related to the change of the steering flow, the equatorial Pacific trade winds and the Western North Pacific subtropical high. Consider two of the ocean cooling effect controlling factors, the ocean cooling effect can be enhanced in the Western North Pacific Ocean under global warming, it need more models analysis in the future work.