海洋环流与波动重点实验室知识库

副热带南大西洋和南印度洋（SAIO）面积广阔，拥有丰富的海洋生物多样性和渔业资源，深受气候变化的影响。尽管远离北大西洋和热带太平洋等气候内部变率中心区，但由于热带气候遥相关导致的和局地产生的大气副热带高压和南半球海洋环流等过程的影响，SAIO的海表面温度（SST）表现出强烈的多时间尺度变率。SAIO的大范围尺度SST变率对周边国家的气候和社会经济有着深远的影响，特别是对非洲-南美-澳洲的陆地降水、区域海平面高度变化和全球海平面不均一性、以及Benguela Niño和Ningaloo Niño等海洋热浪事件的发生具有重要调控作用，牵动粮食生产、淡水供应以及疟疾和炭疽等流行病发生等方面。随着气候变化导致更多极端气候事件增多，这种影响将给周边国家带来更为严峻的考验。然而，相对于太平洋和北大西洋，学界对SAIO的研究较为薄弱，制约着气候预测水平的提高。因此，研究气候变暖背景下SAIO的SST变率的物理机制有助于气候预测，具有重要意义。本文结合了观测和再分析数据、海洋模式实验和气候模式模拟结果，围绕“SAIO的表面海温年代际变率”这一前沿问题进行了系统深入的研究，取得了如下科学发现。

首先，本文研究了SAIO海温20世纪中叶以来的多年代际变化特征，探明了该特征的主要来源，指出了火山和南极臭氧层空洞等外部强迫的关键作用。观测数据表明，南纬20°-45°之间的南大西洋和南印度洋SST的变暖速率在各个时期均较为接近，呈现多年代时间尺度的协同变化。上世纪40-60年代，SAIO的SST与全球平均值（GMSST）的演变较为接近；1963年左右，该区域SST突然下降，并在随后的1965-1975年期间迅速变暖，上升速率为每十年0.43 ℃，远高于GMSST的每十年0.02 ℃的速率；1975年之后，SAIO变暖速率减慢，保持在每十年约0.05 ℃的水平。海洋模式LICOM3的模拟较好地再现了上述低频变化，而基于LICOM3的敏感性强迫实验表明海-气热通量在这些变化中的主导作用。进一步利用CMIP6气候模式探究了海-气热通量变化的来源。CMIP6模式的集合平均（MMM）在一定程度上模拟出了1963-1965年期间SAIO海温的骤降和随后十年的迅速变暖，表明外部强迫因素的重要性。在CMIP6模拟中，SAIO的SST变暖速率很大程度上可以由表面净热通量所解释。具体分析表明，1963年印尼阿贡（Agung）火山爆发的气溶胶强迫导致SAIO海温骤降，随后的十年，随着气溶胶的耗散和海表面辐射通量的恢复，SAIO迅速变暖；1975年之后，南极臭氧洞造成了南大洋西风增强，风速和潜热释放的加强显著削弱了整个SAIO区域的变暖速率。此外，本研究还探讨了内部变率的贡献，如太平洋年代际变率（IPV）和大西洋多年代际变率（AMV）。结果显示，虽然IPV也能通过大气遥相关影响SAIO的大气环流进而调控SST，但在20世纪60年代以来的变化中，其作用总体上是次要的。

在副热带SAIO的SST变率中，除了整个海盆一致的变暖和变冷（basin mode）之外，还会出现西南-东北方向的“偶极子”结构的年代际变率，是年代际SST变率（周期长于7年）的第二模态。而且，南大西洋和南印度洋的年代际偶极子往往协同演变，本文称之为南半球大西洋-印度洋双生偶极子（Atlantic-Indian Twin Dipoles，AITD）。AITD的正位相表现为南大西洋和南印度洋均出现西南部暖SST异常和东北部冷SST异常。观测分析表明，AITD模态能够调控区域性年际变率事件。例如，在AITD的负相位，有利于在南大西洋东边界形成Benguela Niño事件，并在南印度洋东边界形成Ningaloo Niño事件，从而在年代际时间尺度上为这两种区域性年际变率建立了联系，对局地气候和沿岸生态系统产生重要影响。本研究进一步结合LICOM3的敏感性实验和气候模式CESM Large Ensemble（CESM-LE）研究了AITD的机制。结果表明，AITD与SAIO的副热带高压的变化密切耦合，而副热带高压的变化又涉及多种局地海-气过程。具体而言，风场变化驱动的东边界上升/下沉流和云量控制的辐射加热在南大西洋起到了关键作用，而大气温度平流导致的湍流热通量变化在南印度洋更为重要。LICOM3的结果与CESM-LE基本一致，证实了上述过程通过改变海-气热通量控制着AITD模态的形成；总体上，风驱动的海洋动力学过程的作用是次要的。最后，CESM-LE的分析结果还表明，太平洋的IPV和南大洋的南半球环状模（SAM）均能显著引起SAIO的副热带高压变化，进而调控AITD模态。其中IPV主要影响副热带高压的低纬度部分（如南纬25°以北），而SAM主要影响其中纬度部分（如南纬30°以南）。

综上所述，本文研究了上世纪中叶以来SAIO区域SST在年代际和多年代际尺度上的变化规律，揭示了两个大洋准同步变化的鲜明特征，并初步阐明了其变化来源和相关海-气过程。以往在热带和北半球多年代际振荡研究普遍强调气候内部变率的主导性，本研究则凸显了火山和臭氧等外部强迫因素对SAIO区域的重要作用。这体现了SAIO区域的独特性：SAIO内部并未形成显著的低频气候振荡，且远离IPV和AMV的核心区域，故总体上内部变率信号较弱，使得外部强迫因素在该区域的“指纹”（fingerprint）更易显现（emergence）。这一科学认知对气候变化信号的追踪和认定具有一定的启示，也有助于评估和改进气候模式。另一方面，年代际AITD模态将南大西洋和南印度洋各自的复杂变异联系在一起，并指明其调控两大洋同步发生东边界极端海温事件的潜在作用。这些研究结果对改善全球变暖速率、南半球降水和区域极端气候的年代际预测具有重要意义。

The subtropical southern Atlantic and Indian Oceans (SAIOs) are vast, with abundant marine biodiversity and fisheries resources which are deeply affected by climate change. Despite being distant from climate variability centers such as the North Atlantic and tropical Pacific, the sea surface temperatures (SSTs) of the SAIOs exhibit strong multi-scale variability due to the impacts of the local atmospheric subtropical highs, Southern Hemisphere ocean circulations, and remote teleconnections associated with the tropical climates. The large-scale SST variability in the SAIOs has profound impacts on the climate and socio-economy of surrounding countries, especially on regulating the continental precipitation in the Africa-South America-Australia，regional sea level changes and global sea level inhomogeneity, as well as the occurrences of marine heatwaves such as the Benguela Niño and Ningaloo Niño. These influences extend to agricultural production, freshwater supply, and the occurrence of diseases like malaria and anthrax. With climate change leading to an increase in extreme events, this influence will pose even greater challenges to surrounding countries. However, compared to the tropical and Northern Hemisphere oceans, research on the SAIOs is relatively limited, constraining the improvement of climate prediction capabilities. Therefore, studying the physical mechanisms of SST variability in the SAIOs under the background of climate warming is of great significance for climate predictions. This study combines the observational and reanalysis data, ocean model experiments, and climate model simulations to systematically investigate the forefront issue of “decadal variability of SSTs in the SAIOs,” yielding the following scientific findings.

Firstly, this study investigates the multidecadal SST variations in the SAIOs since the mid-20^th century, elucidating their primary origins and highlighting the crucial role of external forcings such as the volcanic eruption and the Antarctic ozone depletion. Observations indicate that the warming rates of SSTs between 20°S and 45°S in the South Atlantic and southern Indian Ocean are relatively consistent across different periods, exhibiting synchronous changes on multidecadal timescales. During the 1940s to 1960s, the evolution of the SAIO SST closely resembled the global mean SST (GMSST). Around 1963, there was a sudden decline in SSTs in this region, followed by rapid warming during the period from 1965 to 1975, with a rate of increase of 0.43 °C per decade, significantly higher than the GMSST rate of 0.02 °C per decade. After 1975, the warming rate in the SAIOs slowed down, maintaining a trend of about 0.05 °C per decade. The ocean model LICOM3 effectively reproduces these low-frequency variations, and the sensitivity experiments based on LICOM3 suggest the significant role of the atmosphere-ocean heat fluxes in these variations. Further exploration of the origins of atmosphere-ocean heat flux changes is conducted using CMIP6. The ensemble mean of CMIP6 models also partially captures the abrupt cooling of SAIO SST around 1963-1965 and the subsequent rapid warming, indicating the importance of external forcings. In the CMIP6 simulations, the warming rates of SAIO SST can largely be explained by surface net heat fluxes. Specifically, the aerosol forcing from the eruption of the Indonesian volcano Agung in 1963 led to the abrupt cooling of SAIO SST, followed by rapid warming in the subsequent decade as aerosols dissipated and surface radiative fluxes recovered. Post-1975, the Antarctic ozone depletion caused an intensification of the Southern Ocean westerlies, significantly weakening the overall warming rate in the entire SAIO region due to the increased wind speeds and enhanced latent heat release. Furthermore, this study examines the contributions of the internal variability, such as the Interdecadal Pacific Variability (IPV) and the Atlantic Multidecadal Variability (AMV). The results indicate that while the IPV can influence the SAIO SST by modulating atmospheric circulations through atmospheric teleconnections, its role in the overall changes since the 1960s is relatively secondary.

In the SST variability of the subtropical SAIOs, in addition to the basin-wide consistent warming and cooling (“basin mode”), the decadal variability of the southwest-northeast “dipole” structure, which is the second mode of the decadal SST variability (with a period longer than 7 years), also occurs. Moreover, the decadal dipoles in the South Atlantic and the southern Indian Ocean tend to evolve in concert, which is called the Southern Hemisphere Atlantic-Indian Twin Dipoles (AITDs) in this paper, and the positive phase of the AITDs is manifested by the warm SST anomalies in the southwest and cold SST anomalies in the northeast of both the South Atlantic and the southern Indian Ocean. Observational analyses show that the AITD mode can modulate regional interannual variability events. For example, negative phase of the AITDs facilitates the formation of the Benguela Niño events at the eastern boundary of the South Atlantic and the Ningaloo Niño events at the eastern boundary of the southern Indian Ocean, thus establishing a link between these two regional interannual variability on decadal timescales, with important impacts on the regional climate and coastal ecosystems. This study further investigates the mechanism of the AITDs with the combination of the sensitivity experiments of LICOM3 and the climate model CESM Large Ensemble (CESM-LE). The results show that the AITD mode is closely coupled with the variability of subtropical highs in the SAIOs, which involves a variety of local air-sea processes. Specifically, wind-induced upwelling/downwelling and cloud-controlled radiative heating play crucial roles in the South Atlantic, while turbulent heat flux variations induced by atmospheric temperature advection are more important in the southern Indian Ocean. The results from LICOM3 are consistent with those from CESM-LE, confirming that the aforementioned processes modulate the formation of the AITD mode by altering atmosphere-ocean heat fluxes; overall, the role of wind-driven ocean dynamics are secondary. Lastly, the analysis from CESM-LE indicates that both the IPV in the Pacific Ocean and the Southern Annular Mode (SAM) in the Southern Ocean significantly influence the variability of the subtropical highs in the SAIOs, thereby regulating the AITD mode. IPV primarily affects the low-latitude sectors of the subtropical highs (north of about 25°S), while SAM predominantly impacts the higher-latitude sectors (south of about 30°S).

In summary, this study investigates the SST variations in the SAIOs at decadal and multidecadal timescales since the mid-20^th century, revealing the distinct features of quasi-synchronous changes in the two oceans and providing preliminary insights into their origins and related atmosphere-ocean processes. While previous research on tropical and Northern Hemisphere multidecadal oscillations emphasized the dominance of internal climate variability, this study highlights the significant role of external forcings such as volcanoes and ozone in the SAIO region. This reflects the uniqueness of the SAIO region: it has not developed significant low-frequency climate oscillations internally and is far from the core regions of the IPV and AMV. Consequently, the overall signal of internal variability is relatively weak, making the emergence of “fingerprint” of the external forcings more readily. This scientific understanding provides insights into tracking and identifying climate change signals and aids in assessing and improving climate models. On the other hand, the decadal AITD mode connects the complex variations in the South Atlantic and the southern Indian Ocean, indicating its potential role in regulating extreme sea temperature events simultaneously occurring along the eastern boundaries of both oceans. These findings hold significant implications for improving the decadal predictions of the global warming, Southern Hemisphere precipitation, and regional extreme climate events.

第1章 绪论........................................................................................ 1

1.1 研究意义............................................................................................................ 1

1.2 研究现状............................................................................................................ 5

1.2.1 南大西洋的SST变率的主要模态............................................................ 5

1.2.2 南印度洋的SST变率的主要模态.......................................................... 11

1.2.3 南大西洋-南印度洋与其他区域的气候联系.......................................... 16

1.3 科学问题和主要研究内容.............................................................................. 20

1.3.1 关键科学问题........................................................................................... 20

1.3.2 主要研究内容........................................................................................... 21

第2章 数据资料和研究方法........................................................... 23

2.1 数据集.............................................................................................................. 23

2.2 气候指数的定义.............................................................................................. 24

2.3 海洋模式实验.................................................................................................. 26

2.4 气候模式模拟.................................................................................................. 27

2.5 提取外强迫...................................................................................................... 29

2.6 其他统计方法.................................................................................................. 30

第3章 南大西洋-南印度洋SST变暖速率的多年代际变化.......... 33

3.1 引言.................................................................................................................. 33

3.2 数据和方法...................................................................................................... 33

3.3 观测数据中的多年代际SST变化.................................................................. 35

3.4 外部强迫因素的影响...................................................................................... 41

3.5 IPV和AMV的影响......................................................................................... 46

3.6 本章小结.......................................................................................................... 50

第4章 南大西洋-南印度洋的年代际双生偶极子模态................... 53

4.1 引言.................................................................................................................. 53

4.2 数据与方法...................................................................................................... 55

4.3 年代际双生偶极子的主要特征...................................................................... 57

4.4 南大西洋-南印度洋的海-气过程.................................................................... 59

4.4.1 CESM模拟结果........................................................................................ 59

4.4.2 LICOM3实验的验证................................................................................. 65

4.5 太平洋和南大洋的影响.................................................................................. 67

4.6 本章小结.......................................................................................................... 72

第5章 总结与展望.......................................................................... 75

5.1 对目前工作的总结.......................................................................................... 75

5.1.1 南大西洋-南印度洋SST的多年代际变化............................................. 75

5.1.2 南大西洋-南印度洋年代际双生偶极子.................................................. 77

5.2 本文的创新性.................................................................................................. 79

5.3 对未来工作的展望.......................................................................................... 80

附录：中英文缩写对照....................................................................... 81

参考文献............................................................................................... 83

致 谢................................................................................................. 105

作者简历及攻读学位期间发表的学术论文与其他相关学术成果 107