热带印度洋海平面低频变化及机理研究

发布时间：2018-07-15 07:06

【摘要】：本文采用卫星高度计数据、验潮站数据以及海平面重构数据较为系统的分析了热带印度洋海平面低频变化特征；结合温盐、流场和风场与海平面变化的关系，探讨海平面年际变化机理。热带印度洋海平面年际变化具有显著的区域性特征，主要集中在西南热带印度洋和苏门答腊-爪哇岛沿岸，标准差可达8cm。西南热带印度洋海平面变化滞后苏门答腊-爪哇海区3个月时，相关系数达到最大负值-0.88，印度半岛沿岸和澳大利亚西部沿岸的验潮站海平面变化序列有很好的滞后关系。IOD（印度洋偶极子）和ENSO事件共同发生期间海平面变化较IOD事件期间更为显著，IOD事件期间的海平面变化较ENSO事件期间强且范围广，ENSO事件期间的海平面变化则主要位于南热带印度洋。澳大利亚西岸海平面卫星高度计数据、以及Fremantle验潮站均与ENSO指数有很好相关关系，同步相关系数超过-0.6。海平面EOF第1模态的方差贡献为44%，呈东西反向变化，第2模态则主要表现为“三明治”结构。热带印度洋海平面存在年代际变化，主要集中在年际变化显著的海域（西南热带印度洋和苏门答腊-爪哇海区）。海平面年代际变化具有显著的空间差异，重构数据显示年代际变化空间分布与年际变化相似，且与太平洋年代际振荡（PDO）有关。52年的重构数据第1模态的方差贡献为73%，与PDO指数呈负相关，分别在1974和1994年达到年代际的极值；澳大利亚西岸验潮站海平面序列也与PDO密切相关，两者的相关系数达到-0.75。区域海平面年际变化的主要影响因素是比容效应和大气风场强迫，，而海洋环流与净热通量在局部区域也有着重要作用。大气风场通过Ekman效应和斜压Rossby波影响海平面变化，且区域特征显著。洋盆尺度上，斜压Rossby的作用明显大于局地Ekman效应，前者能解释30%的海平面年际变化；仅在阿拉伯海和南热带印度洋中部，风场强迫的局地Ekman效应对海平面年际变化的影响更为显著。IOD和ENSO事件期间的风场与海平面年际变化有重要联系，使得ENSO事件期间的海平面年际变化偏南偏弱。洋盆尺度上，比容效应能解释至少50%的海平面变化。主温跃层的抬升（下沉）伴随海水的增温（降温），相应出现海平面的抬升（下沉），海水温度的垂直结构变化与海平面年际变化有很好的一致性。热带印度洋大部分海域海表净热通量引起的海平面变化与实测海平面变化呈负相关，而在阿拉伯海域热通量有重要作用。合成分析显示，在西南热带印度洋出现逆时针流涡，与顺时针的背景流场相反，IOD+ENSO事件期间最强，IOD事件期间次之，ENSO事件期间最弱；逆时针流场的存在使得在三种事件期间将会在西南热带印度洋产生海平面正异常，流场的输运对海平面变化有调整作用。印尼贯穿流与澳大利亚西岸海平面变化有密切关系。
[Abstract]:In this paper, satellite altimeter data, tide gauge data and sea level reconstruction data are used to analyze the characteristics of low frequency variation of sea level in tropical Indian Ocean, and the relationship between temperature and salt, current field and wind field and sea level change. The mechanism of interannual sea level variation is discussed. The interannual variation of sea level in the tropical Indian Ocean has obvious regional characteristics, mainly in the southwest tropical Indian Ocean and the coast of Sumatra and Java, with a standard deviation of up to 8 cm. When sea level changes in the tropical southwest Indian Ocean lag behind the Sumatra-Java Sea region for 3 months, The correlation coefficient reaches the maximum negative value -0.88. There is a good lag relationship between the series of sea level changes in the Indian Peninsula coast and the western coast of Australia. IOD (Indian Ocean Dipole) and ENSO events occurred together, the sea level change was higher than that of IOD. The sea level change during IOD event is stronger than that during ENSO event and the sea level change during ENSO event is mainly in the southern tropical Indian Ocean. The sea level altimeter data on the west coast of Australia and the Fremantle tide gauge are well correlated with the ENSO index, and the synchronous correlation coefficient is more than -0.6. The variance contribution of the first mode of sea level EOF is 44, which is inversely changed from east to west, and the second mode is mainly "sandwich" structure. Interdecadal sea level changes in the tropical Indian Ocean are mainly concentrated in the sea areas with significant interannual changes (tropical Indian Ocean and Sumatra-Java Sea region). The spatial distribution of decadal variation is similar to that of interannual variation, and the reconstructed data show that the spatial distribution of decadal variation is similar to that of interannual variation. The variance contribution of the first mode of the reconstructed data in 52 years is 73 and negatively correlated with the PDO exponent, reaching the Interdecadal extremum in 1974 and 1994, respectively. The sea level sequence of tide gauge stations in the west coast of Australia is also closely related to PDO, and the correlation coefficient between them is -0.75. The main influencing factors of regional sea level interannual variation are specific volume effect and atmospheric wind forcing, while ocean circulation and net heat flux also play an important role in the local area. The atmospheric wind field influences sea level changes through Ekman effect and baroclinic Rossby waves, and the regional characteristics are significant. On the scale of ocean basin, baroclinic Rossby plays a more important role than the local Ekman effect, the former can explain 30% of the interannual sea level change, only in the Arabian Sea and the central part of the southern tropical Indian Ocean. The effect of local Ekman effect of wind forcing on the interannual variation of sea level is more significant. IOD and the wind field during ENSO events are closely related to the interannual variation of sea level, which makes the interannual variation of sea level weaker to the south. On the scale of ocean basin, the specific volume effect can explain at least 50% of sea level change. The uplift (subsidence) of the main thermocline is accompanied by the temperature increase (cooling) of the sea water and the rise (subsidence) of the sea level. The vertical structure variation of the seawater temperature is in good agreement with the interannual variation of the sea level. The sea level change caused by sea surface net heat flux in most of the tropical Indian Ocean is negatively correlated with the measured sea level change, but plays an important role in Arab sea area. Synthesis analysis shows that in the southwest tropical Indian Ocean there is an anticlockwise current vortex, which is contrary to the clockwise background flow field, which is the weakest during the strongest IOD ENSO event and the second one during the ENSO event. The existence of counterclockwise flow field will lead to positive sea level anomalies in the southwest tropical Indian Ocean during the three events, and the transport of the current field can adjust the sea level change. Indonesian penetration is closely related to sea level changes in the west coast of Australia.
【学位授予单位】：中国海洋大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：P731.23

【参考文献】