海洋环流与波动重点实验室知识库

湍流混合是控制全球经圈翻转环流的重要因子，然而目前发现的大洋中的背景混合并不足以维持全球经圈翻转环流。粗糙海底上方内波破碎所形成的叠加在背景场上的混合增强区域则可能弥补这种混合的缺失。海底生成的内波主要包括内潮和背风波，前人对内潮驱动的混合开展了广泛的研究，但对背风波驱动的混合的关注较少。过去有研究推测背风波驱动的增强混合对全球深海水团转化率的贡献至少达到三分之一，进而显著影响全球经圈翻转环流，但是此估算结果受制于背风波驱动的增强混合的地理分布和垂直结构。因此，了解背风波驱动的增强混合的地理分布和垂直结构，对于准确评估深海水团转化来说至关重要。受此启发，在本文中，我们首先利用现有观测资料探索北太平洋西边界流区可能存在的背风波驱动的增强混合，随后基于数值模式分析控制背风波驱动的增强混合的垂直结构的关键参数和物理机制，具体的研究内容和成果如下：

1）棉兰老流通过向印尼贯穿流提供北太平洋水而连接了北太平洋和印尼海，其沿途湍流混合的真实空间分布情况还有待阐明。通过将基于Garrett-Munk（GM）内波谱的细尺度参数化方案应用到两个沿棉兰老流快速采样的Argo浮标所收集到的温盐剖面数据，我们得到了棉兰老流沿途的湍流混合空间分布特征。计算结果显示，桑义赫海峡和靠近棉兰老岛的西边界（WB）区域是棉兰老流沿途的两个混合增强区域，其耗散率和扩散率分别高达~10^-6 W kg^-1和~10^-3 m² s^-1。进一步的分析表明，桑义赫海峡内混合的增强与潮流和粗糙地形之间的相互作用有关。一般来说，潮流与海底地形相互作用会生成内潮。考虑到桑义赫海峡中突出海山上方的潮流显著增强（0.1-0.2 m/s），可能使得局地潮偏移参数大于1，因此我们推测桑义赫海峡也存在着强潮流激发的背风波。此外，值得注意的是，尽管WB区域的混合强度与桑义赫海峡相当，但WB区域的潮流速度几乎可以忽略不计。考虑到前人的观测研究表明，在WB区域存在着强深层地转流速（0.1-0.3 m/s），因此该处混合的增强很可能与底部稳定流和粗糙海底地形相互作用所生成的背风波有关。同时我们还发现，棉兰老流沿途的水团转化主要发生在桑义赫海峡，因为该处的水团长时间滞留在强混合区域。

2）目前，当内波驱动的底部增强混合被参数化到全球海洋环流模式和气候模式中时，耗散率的垂直衰减尺度ζ往往被假设为全球均一的常数500 m。基于包含背景GM内波场的非静力近似数值模式，我们发现背风波破碎所驱动的底部增强混合的垂直衰减尺度ζ实际上是可变的，其数值大小受稳定流流速和海底地形尺度的影响。当陡度参数（SP = Nh_T / U₀：h_T为海底地形高度，N为海底附近的背景浮性频率 ）小于0.3时，背风波从海底向上传播，同时与背景GM内波场进行非线性相互作用，形成底部增强混合区域。底部增强混合的垂直衰减尺度ζ随着稳定流流速U₀的增加而增大，但受地形波数k_H的影响较小。而且，当稳定流流速足够强时，比如说，U₀ ³ 0.1 m s^-1，底部增强混合能保持其强度不衰减一直延伸到上层海洋。这与过去在南大洋锋面区域观测到的现象一致，在锋面区域，南极绕极流的底部流速增强，伴随着贯穿全水深的增强混合。相反，当陡度参数SP大于0.3时，近惯性振荡逐渐在海底地形上方发展起来，抑制了海底生成的背风波的向上传播，垂直衰减尺度ζ也因此变得非常小。

3）前人认为当SP大于0.3时，稳定流和近惯性振荡分别与地形相互作用产生两种内波，然后这两种内波再通过三波共振相互作用将能量传递到具有近惯性频率的第三种内波，以加速近惯性振荡。但是我们的模拟研究显示近惯性振荡与稳定流是共同而不是分开与地形相互作用，而且背风波的生成量具有时间依赖性。具体表现为近惯性振荡叠加在稳定流上形成一支复合振荡流，而这支复合振荡流在以其流速最大值为中心的时间段内，也就是当其流速暂时稳定时，最有效地与海底地形相互作用产生背风波。随后，海底生成的背风波在向上传播过程中与近惯性振荡发生非线性相互作用。由于与近惯性振荡的强垂直剪切相关的多普勒频移效应，背风波的垂直波数逐渐增大至破碎极限，而且背风波在破碎时，会同时传递一部分能量给近惯性振荡。增强的近惯性振荡又进一步加速了背风波的多普勒频移，由于正反馈机制，近惯性振荡的振幅随时间不断增大。

以上发现显示，获取高分辨率的海底地形数据和深层流速数据有益于了解背风波驱动的增强混合的地理分布情况。此外，运用更合理的参数化方案将这些背风波驱动的增强混合纳入全球海洋环流模式及气候模式之中，比如说考虑底部流速和地形尺度对混合垂直分布结构的影响，可能将显著提升模式性能。

A strong candidate to cover the shortfall in diapycnal mixing required for maintaining the global overturning circulation is the enhanced turbulent mixing above the rough seafloor induced by the breaking of internal waves. The internal waves generated at the seafloor mainly include internal tides and internal lee waves. Compared to the extensive research on internal tide-driven mixing, little attention has been paid to internal lee wave-driven mixing. Previous studies showed that the internal lee wave-driven mixing hotspots could be responsible for about 1/3 or more of global abyssal water mass transformation rate thus affecting the global overturning circulation, but this assessment was subject to the geographic distribution and the vertical structure of internal lee wave-driven mixing hotspots. Therefore, understanding the geographic distribution and vertical structure of internal lee wave-driven mixing hotspots is essential for evaluating the global abyssal water mass transformation. Motivated by this, in the present study, we first use the available observations to explore the possible presence of internal lee wave-driven mixing hotspots in the western boundary current region of the North Pacific Ocean, and then we use numerical models to investigate the key parameters as well as the physical mechanisms controlling the vertical structure of internal lee wave-driven mixing hotspots. The main findings of the present study are as follows:

1) The Mindanao Current (MC) bridges the North Pacific Ocean and the Indonesian Seas by supplying the North Pacific waters to the Indonesian Throughflow, and the real spatial distribution of diapycnal mixing along the MC has long remained to be clarified. We tackle this question here by applying Garrett-Munk (GM) spectrum-based finescale parameterization to the temperature and salinity profiles obtained using two rapid-sampling profiling Argo floats that drifted along the MC. We find that, the Sangihe Strait close to the Mindanao Islands and the western boundary (WB) region are the two mixing hotspots along the MC, with energy dissipation rate ε and diapycnal diffusivity K_ρ enhanced up to ~10^-6 W kg^-1 and ~10^-3 m² s^-1, respectively. Except for the above two mixing hotspots, the mixing along the MC is mostly weak, with ε and K_ρ to be 10^-11-10^-9 W kg^-1 and 10^-6-10^-5 m² s^-1, respectively. Our analysis suggests that the enhanced mixing within the Sangihe Strait is associated with the interaction between the rough seafloor topography and strong tidal flows. In general, the interaction of tidal flows and seafloor topography generates internal tides. Considering that tidal flows near the top of prominent ridges in the Sangihe Strait are amplified to 0.1-0.2 m/s, so that the local tidal excursion parameter may exceed unity, we can speculate the existence of tidally generated internal lee waves. It is interesting to note that although the mixing intensity in the WB region is comparable to that in the Sangihe Strait, the speed of tidal flows in the WB region is almost negligible. Taking into account the previous studies showing that strong geostrophic velocities of 0.1-0.3 m/s exist in the deep ocean of the WB region, we can speculate that such enhanced mixing are related to internal lee waves generated by the interaction of bottom steady flows and the rough seafloor topography. We also find that water mass transformation along the MC mainly occurs in the Sangihe Strait where the water masses are subjected to the strong turbulent mixing during a long residence time.

2) The effects of the internal wave-driven bottom-enhanced mixing are often parameterized in global ocean general circulation models and climate models with an assumption that the vertical decay scale ζ of energy dissipation rate is universally constant at 500 m. Using a non-hydrostatic numerical model incorporating the background GM internal wave field, we find that ζ of the internal lee wave-driven bottom-enhanced mixing is actually variable depending on the magnitude of the steady flow and topographic scale. When the steepness parameter of the seafloor topography SP (SP= Nh_T/U₀: h_T is the height of the seafloor topography and N is the background buoyancy frequency near the seafloor) is smaller than 0.3, internal lee waves propagate upward from the seafloor topography while interacting with the background GM internal wave field to create a turbulent mixing area with ζ extending further upward off the seafloor as U₀ increases, but nearly independent of k_H. Especially, when the steady flow is sufficiently strong, say, U₀ ³ 0.1 m s^-1, the enhanced mixing is found to extend throughout the water column without obvious decay, which is consistent with the observed features near the frontal zones in the Southern Ocean where the Antarctic Circumpolar Current locally intensifies. When SP exceeds 0.3, in contrast, the near-inertial oscillations （NIOs） gradually develop just above the seafloor topography, inhibiting the upward propagation of the bottom-generated internal lee waves, so that ζ becomes very small.

3) Previous studies suggested that, when SP exceeds 0.3, the steady flow and the NIOs individually interact with the topography to generate two waves, which then transfer energy to a third wave with near-inertial frequency via wave triad interactions to accelerate the NIOs. However, our simulation results show that NIOs and steady flow interact with the topography jointly rather than individually, as shown by the oscillatory flow consisting of the superposition of the steady flow and the NIOs most efficiently generates internal lee waves during the period centered on the time of its maximum when it becomes temporarily steady. Bottom-generated internal lee waves propagate upward while interacting with the background NIOs. At the depths where their vertical wavenumbers are Doppler shifted up to the breaking limit by the strong vertical shear associated with the NIOs, the internal lee waves break while supplying part of their energy to the background NIOs. The enhanced NIOs, in turn, further accelerate the doppler shifting of the internal lee waves propagating upward from the seafloor topography, and the NIOs are continuously enhanced due to positive feedback.

The above findings indicate that obtaining high-resolution bathymetry data and deep current velocity data will benefit the understanding of the geographic distribution of internal lee wave-driven mixing hotpots. Furthermore, incorporating these internal lee wave-driven mixing hotspots into global ocean circulation models and climate models using more reasonable parameterization schemes, such as taking into account the effects of bottom velocity and topographic scale on the vertical structure of internal lee wave-driven mixing hotspots, may significantly improve model performance.

第1章 绪论... 1

1.1 研究背景... 1

1.1.1 湍流混合的研究意义... 1

1.1.2 海洋内波的产生与分布... 3

1.1.3 Garrett-Munk大洋内波谱... 6

1.1.4 三波共振相互作用... 7

1.1.5 波致混合参数化方案... 9

1.2 背风波驱动混合的研究现状... 10

1.2.1 对经圈翻转环流的影响... 10

1.2.2 地理分布情况... 12

1.2.3 垂直结构特征... 14

1.3 科学问题与本文研究内容... 16

第2章 数据与方法... 17

2.1 Argo剖面数据... 17

2.2 格点化数据集... 18

2.2.1 海底地形数据... 18

2.2.2 TPXO8潮流数据... 19

2.2.3 ASCAT风场数据... 19

2.2.4 WOA13温盐数据... 19

2.3 细尺度参数化方案... 19

2.4 MITgcm数值模式... 22

2.5 在MITgcm中引入GM内波场... 23

第3章 棉兰老流沿途的混合热点区域及其主要能量来源... 25

3.1 研究背景... 25

3.2 数据处理... 26

3.2.1 选取快速采样的Argo浮标... 26

3.2.2 混合强度的估算与验证... 28

3.3 结果与讨论... 30

3.3.1 棉兰老流沿途的湍流混合分布情况... 30

3.3.2 与前人的混合观测研究结果进行对比... 34

3.3.3 混合热点区域的主要能量来源... 36

3.3.4 棉兰老流沿途的水团转化... 39

3.4 本章小结... 43

第4章 背风波驱动的底部增强混合的垂直结构特征... 44

4.1 研究背景... 44

4.2 数值模式设置与数值实验介绍... 45

4.3 结果与讨论... 48

4.3.1 小陡度参数下的混合热点垂直结构... 50

4.3.2 较大陡度参数下的混合热点垂直结构... 53

4.3.3 与前人的模拟和理论研究进行对比... 55

4.3.4 与过去南大洋的混合观测结果进行对比... 55

4.4 本章小结... 57

第5章 陡度参数较大时海底上方近惯性振荡的演变过程... 59

5.1 研究背景... 59

5.2 数值模式设置与数值实验介绍... 59

5.3 结果与讨论... 61

5.3.1 稳定流与近惯性振荡共同与地形相互作用... 61

5.3.2 近惯性振荡引起的背风波的多普勒频移... 63

5.3.3 背风波与近惯性振荡之间的能量传递... 65

5.3.4 关于近惯性振荡发展过程的设想... 66

5.4 本章小结... 67

第6章 总结与展望... 68

6.1 主要结论... 68

6.1.1 棉兰老流沿途的混合热点区域及其主要能量来源... 68

6.1.2 背风波驱动的底部增强混合的垂直结构特征... 69

6.1.3 陡度参数较大时海底上方近惯性振荡的演变过程... 69

6.2 特色与创新... 70

6.3 不足之处与未来工作展望... 70

参考文献... 71

附录... 83

附录A: 推导内波速度扰动和密度扰动（公式2.7-2.9）... 83

附录B: 推导背风波的垂直波数随时间的变化（公式5.7）... 86

致  谢... 87

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... 89

作者简历... 89

已发表（或正式接收）的学术论文... 89

在读期间的出海调查经历... 89

参加的研究项目及获奖情况... 89