海洋环流与波动重点实验室知识库

太平洋副热带-热带经向翻转环流（Subtropical-tropical cells, STCs）是连接太平洋副热带和热带的海洋通道，对热带—副热带的海洋气候变化起着重要作用。内部STC输运是STCs向赤道输运的次表层分支，与热带太平洋年代际变率有着密切关系，并通过经向输运影响着赤道太平洋的海表温度变化。此外，厄尔尼诺—南方涛动（El Niño-Southern Oscillation, ENSO）是热带太平洋最显著的年际变率，而与赤道水体辐合辐散紧密相关的STC被认为是影响ENSO的因素之一。因此，本文基于Argo观测数据、多套再分析产品和模式数据，利用多种统计学方法，结合一层半海洋约化重力模型，研究太平洋STCs的变异规律及其与ENSO的关系。

首先研究了不同纬度上内部STC输运的季节变化规律，定性分析了不同纬度上风场和Rossby波对内部STC输运季节变化的相对贡献。通过使用多源数据分析发现沿不同纬度，内部STC输运存在不同的季节变化特征：在10°-2°S和2°-6°N，内部STC输运在春（夏秋）季较强（弱）；在7°-15°N，内部STC输运在夏（冬）季较强（弱）；在15°-11°S，内部STC输运在冬（夏）季较强（弱）。根据大洋Sverdrup理论，内部STC主要受局地表层风场和西传Rossby波调控。在不同纬度上，两者对内部STC输运季节变化的贡献并不相同：在10°S-6°N，内部STC输运的季节变化与表层风场基本一致，表明该区域风场强迫对内部STC输运变化的影响更大；在15°-11°S及7°-10°N，内部STC输运变化滞后表层风场，说明与表层风场相比，西传Rossby波对内部STC输运变化的影响更大；与近赤道区域相比，赤道外区域的Rossby波对内部STC输运的调控作用更明显。

其次，研究了内部STC输运与ENSO关系的年代际变化及其变化原因。研究结果表明内部STC输运与ENSO关系发生年代际变化：在1930-2010年，沿9°S的内部STC输运与ENSO密切相关；沿9°N的内部STC输运与ENSO在20世纪30年代末到60年代末（PD1：1937-1965）关系不显著，20世纪60年代末到21世纪初（PD2：1966-2003）两者关系显著。因此沿9°N的内部STC输运与ENSO关系在20世纪60年代末发生年代际变化，其主导因素与风场有关。在PD1，沿9°N的内部STC输运主要受副热带北太平洋东北风影响，而ENSO相关风场在该区域为西北风，说明在该时期ENSO相关风场并不能主导STC输运变化；在PD2，影响STC的风场主要位于赤道中西太平洋，且与ENSO相关风场空间分布基本一致。因此在PD2，STC主要受ENSO相关风场调控。进一步研究发现大西洋多年代际振荡（Atlantic Multidecadal Oscillation, AMO）是调控两者关系的可能物理机制之一。在AMO正位相期间，北太平洋有较强的东北风异常，AMO驱动的副热带太平洋表面风场有利于STC的发展，ENSO相关的热带风场对STC的影响较弱。在AMO负位相期间，赤道西风异常促进ENSO发展，且温跃层反馈的增强促进次表层海温对温跃层变化的响应，而表征赤道水体辐合辐散的STC输运可以影响次表层海温，从而增强与ENSO的关系。此外，在PD2，AMO通过影响赤道风场来促进ENSO的发展，ENSO振幅的增强驱动更强的热带风场。与副热带风场相比，该时期的STC主要由更强的热带风场驱动，因此STC更多地受到与ENSO相关风场的影响，STC-ENSO关系增强。

最后使用CMIP6耦合模式分析历史模拟与未来情景下内部STC输运的变化趋势，明确了STC输运的未来变化趋势并讨论其变化原因。研究结果表明大部分CMIP6耦合模式能够较好地模拟内部STC的时间平均输运，在模拟时间变异性方面存在差异。在未来气候下，南北半球的内部STC将减弱，并指出上层海洋层结增强和风场减弱的共同作用导致内部STC输运的减弱。在SSP585情景下，中西太平洋风应力旋度和近赤道东向纬向风应力减弱，进而向极的Ekman输运和赤道跃层输运减弱，导致内部STC输运向赤道辐合减弱。由于全球变暖，海洋上层密度降低，层结增强，副热带海水沉降减弱，导致STCs内部分支减弱。内部STC与西边界输运的补偿关系在未来情景下仍然存在且沿9°S两者补偿关系更强。此外，与海洋再分析产品相比，耦合模式往往低估内部STC输运辐合变率，从而可能失去一些海表面温度（Sea Surface Temperature, SST）驱动力，这可能是一些模式模拟的STC-SST相关性较低的原因。未来情景模拟表明，在全球变暖的情况下，内部STC的热输运减弱，模型间存在普遍一致性。

The Pacific subtropical-tropical cells (STCs) are the oceanic channels connecting the subtropical and tropical Pacific Ocean, and play an important role in tropical-subtropical ocean climate change. The interior STC transport is the subsurface branch of the equatorward transport of STCs, which is closely related to the inter-decadal variability of the tropical Pacific and influences the sea surface temperature in the equatorial Pacific through meridional transport. In addition, ENSO (El Niño-Southern Oscillation) is the most significant interannual variability in the tropical Pacific, and STCs, which are closely related to equatorial water body irradiation and dispersion, are considered to be one of the factors affecting ENSO. Therefore, based on Argo observation data, multiple sets of reanalysis and model data, this article uses various statistical methods, combined with a 1.5-layer ocean reduced gravity model, to study the variation patterns of Pacific STCs and their relationship with ENSO.

The seasonal variability of interior STC transport at different latitudes are investigated, and the relative contributions of the local wind and Rossby waves to the seasonal variability of interior STC transport at different latitudes are qualitatively analyzed. The analysis using multi-source data reveals that there is different seasonal variability of interior STC transport along different latitudes: at 10°-2°S and 2°-6°N, the interior STC transport is stronger (weaker) in the spring (summer-autumn); at 7°-15°N, the interior STC transport is stronger (weaker) in the summer (winter); and at 15°-11°S, the interior STC transport is stronger (weaker) in the winter (summer). According to the Sverdrup theory, the interior STC is mainly modulated by the local surface wind field and the westward Rossby waves. At different latitudes, the contributions of the two to the seasonal variability of interior STC transport are not the same: at 10°S-6°N, the seasonal variability of interior STC transport are basically the same as that of the surface wind field, suggesting that the wind forcing in this region has a greater influence on the variation of interior STC transport; at 15°-11°S and 7°-15°N, the variation of interior STC transport lags behind the surface wind field, suggesting that the westward Rossby waves have a greater influence on the interior STC transport than the surface wind field, and that the Rossby waves in the off-equatorial region have a more pronounced modulation effect on the interior STC transport compared with the near-equatorial region.

Secondly, the interdecadal changes in the relationship between interior STC transport and ENSO and the reasons for these changes were investigated. The results of the study show that there has been a clear interdecadal change between interior STC and ENSO: the interior STC transport along 9°S is closely related to ENSO; The relationship between interior STC transport along 9°N and ENSO is not significant from the late 1930s to the late 1960s (PD1: 1937-1965), and is significant from the late 1960s to the early 2000s (PD2: 1966-2003). Thus, there is an interdecadal change in the relationship between interior STC transport along 9°N and ENSO in the late 1960s, with the dominant factor related to the wind field. In PD1, the interior STC transport along 9°N is influenced by the northeast wind in the subtropical north Pacific, while the ENSO-related wind field is northwest wind in this region, indicating that the ENSO-related wind field does not dominate the variation of STC transport in this period; in PD2, the wind field that influences the STC is located in the equatorial west-central Pacific Ocean and is basically the same as that in the spatial distribution of the ENSO-related wind field. Therefore, in PD2, the STC is mainly modulated by the ENSO-related wind field. Further studies identified the Atlantic Multidecadal Oscillation (AMO) as one of the possible physical mechanisms modulating the relationship. During +AMO, the north Pacific Ocean has strong northeasterly wind anomalies, and the AMO-driven subtropical Pacific surface wind field favors the STC, while the ENSO-related tropical wind field has a weaker effect on the STC. During -AMO, equatorial westerly wind anomalies promote ENSO development and enhanced thermocline feedback promotes subsurface temperature response to thermocline variability, whereas STC characterizing equatorial water mass convergence and divergence can influence subsurface temperaure and thus enhance the relationship with ENSO. Furthermore, in PD2, the AMO contributes to ENSO by influencing the equatorial wind field, and the enhanced ENSO amplitude drives a stronger tropical wind field. The STC in this period is mainly driven by stronger tropical wind fields compared to subtropical wind fields, and thus the STC is more influenced by ENSO-related wind fields and the STC-ENSO relationship is enhanced.

Finally, the CMIP6 coupled models are used to analyze the linear trend of interior STC transport under historical simulation and future scenarios, clarifying the future trend of STC transport and discussing its reasons for change. The research results indicate that most CMIP6 coupled models can simulate the time averaged transport of interior STC well, there are differences in simulating time variability. In the future climate, the interior STC in the northern and southern hemispheres will weaken, and it is pointed out that the combined effect of enhanced upper ocean stratification and weakened wind fields will lead to the weakening of interior STC transport. In the SSP585 scenario, the wind stress curl in the western-central Pacific Ocean and the zonal wind stress near the equator is weakened, which in turn weakens the poleward Ekman transport and the equatorial pycnocline transport, leading to the weakening of the interior STC transport convergence toward the equator. Due to global warming, the density of the upper layer of the ocean decreases, stratification strengthens, and subtropical seawater subsidence weakens, resulting in a weakening of the interior branches of the STCs. The compensation relationship between interior STC and western boundary transport still exists in the future scenario, and the compensation relationship between the two is stronger along 9°S. In addition, compared with ocean reanalysis products, the coupled models tend to underestimate the variability of the interior STC transport convergence, and thus may lose some sea surface temperature (SST) driving force, which may be the reason for the low STC-SST correlation simulated by the model. The future scenario simulation shows that the heat transport of interior STC is weakened under global warming, with a general agreement across models.

第1章 绪论    

1 1.1 STCs研究现状   1

1.1.1 STCs季节-年际-年代际变异规律    1

1.1.2 STCs影响机制    6

1.2 STCs与热带太平洋海温的关系    8

1.3 内部STC与ENSO的关系    11

1.3.1 ENSO多样性    11

1.3.2 内部STC与ENSO的关系    12

1.4 本文研究内容与研究意义    15

1.4.1 主要研究内容    15

1.4.2 研究意义    15

第2章 数据和方法    17

2.1 数据    17

2.1.1 观测和再分析数据    17

2.1.2 CMIP6数据    19

2.2 方法    21

2.2.1 地转流计算方法    21

2.2.2 STC输运计算方法    21

2.2.3 位势涡度计算    22

2.2.4 合成、回归和相关分析    23

2.2.5 梁氏-克里曼信息流    23

2.2.6 指数定义    24

2.2.7 滤波    24

第3章 不同纬度下内部STC的季节变化规律    27

3.1 引言    27

3.2 不同纬度下内部STC的季节变化规律    27

3.3 内部STC季节变化可能的物理机制    32

3.4 小结    39

第4章 内部STC输运与ENSO关系的年代际变化    41

4.1 引言    41

4.2 内部STC与ENSO振幅的年代际变化    41

4.2.1 内部STC输运的年代际变化    41

4.2.2 ENSO发展过程及振幅的年代际变化    43

4.3 内部STC与ENSO关系的年代际变化    48

4.4 内部STC与ENSO关系年代际变化的可能机制    50

4.5 小结    56

第5章 全球变暖下CMIP6模式中STCs的变化    59

5.1 引言    59

5.2 内部STC输运在CMIP6模式中的表现及其未来变化    59

5.2.1 内部STC在CMIP6模式中的表现    59

5.2.2 内部STC在未来情景下的变化趋势    62

5.2.3 内部STC与西边界输运的关系    65

5.3 内部STC输运与赤道太平洋SST的关系    68

5.4 内部STC热输运    74

5.5 小结    75

第6章 总结与展望    77

6.1 主要结论    77

6.2 论文创新点    78

6.3 未来展望    79

参考文献    83

致  谢    93

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    95