日本地球惑星科学連合2015年大会

講演情報

口頭発表

セッション記号 A (大気水圏科学) » A-CG 大気水圏科学複合領域・一般

[A-CG32] 熱帯におけるマルチスケール大気海洋相互作用現象

2015年5月26日(火) 16:15 〜 18:00 202 (2F)

コンビーナ:*時長 宏樹(京都大学防災研究所・白眉センター)、長谷川 拓也(独立行政法人海洋研究開発機構)、清木 亜矢子(海洋研究開発機構)、東塚 知己(東京大学大学院理学系研究科地球惑星科学専攻)、名倉 元樹((独) 海洋研究開発機構)、大庭 雅道(電力中央研究所 環境科学研究所 大気海洋環境領域)、今田 由紀子(東京大学大気海洋研究所)、座長:名倉 元樹((独) 海洋研究開発機構)、今田 由紀子(気象庁気象研究所)

17:40 〜 17:55

[ACG32-12] On the role of internal atmospheric processes in interannual equatorial variability

*Ingo RICHTER1 (1.JAMSTEC, APL)

キーワード:equatorial variability, ENSO, IOD, Atlantic Nino, predictability, surface winds

Major modes of tropical variability, such as El Nino-Southern Oscillation (ENSO), the Indian Ocean Dipole (IOD), or the Atlantic zonal mode, have been found to arise from coupled air-sea interaction. An often invoked mechanism in this context is the Bjerknes feedback, in which equatorial zonal winds respond to sea-surface temperature (SST) anomalies in such a way as to reinforce the original anomaly. Recent studies, however, have reexamined the role of coupled feedbacks and found that they might be less important than previously thought. Here we examine the issue by focusing on equatorial surface winds, which undoubtedly play an important role in driving oceanic variability in the equatorial region. We compare fully coupled general circulation models (GCMs) from the Coupled Model Intercomparison Project Phase 5 (CMIP5) with an experiment in which the atmospheric component is forced with the climatological seasonal cycle of SST. For the equatorial Atlantic, the analysis reveals that surface wind variability decreases by only about 25% when climatological SSTs are prescribed. This suggests that a large portion of equatorial Atlantic surface wind variability is due to internal atmospheric processes. In the equatorial Pacific and Indian Ocean, on the other hand, surface wind variability reduces substantially when climatological SSTs are prescribed, indicating the importance of coupled feedbacks. Even there, however, the intrinsic atmospheric component can be quite large depending on the season and is subject to a large inter-model spread. Potential reasons for the model spread will be discussed.