【作者】 陈美香；

【导师】 左军成；

【作者基本信息】 中国海洋大学 ， 物理海洋学， 2009， 博士

【摘要】 本文利用AVISO卫星高度计资料分析了北太平洋1993-2006年海平面变化特征及主要模态,并对比容、降水和蒸发对海平面不同尺度变化的影响进行了研究。对东海黑潮和黑潮延伸体邻近海域海平面变化特征进行分析,讨论比容、蒸发、降水以及黑潮流强度对海平面变化的影响。利用包含正压Sverdrup平衡和一阶斜压罗斯贝波动力学的一层半理论模型对黑潮延伸体所在北太平洋中纬度海区的海平面进行机制研究,分析风驱动的罗斯贝波对北太平洋中纬度海区海平面变化的动力学贡献。利用POP模式对全球海平面进行了厄尔尼诺事件及时间序列模拟实验,分别讨论海平面的年际变化及长期趋势变化的主要影响机制。1993～2006年北太平洋大部分海区海平面都在不同程度的上升,少数海区存在下降趋势。上升区域基本位于洋盆西侧的中低纬海区,高纬度海区及中低纬度海区的大洋东岸海平面在下降。上升最快和下降最快的海区位于黑潮和亲潮的交汇海域。整个北太平洋在这14年间的平均上升率为2.87mm/a。北太平洋海平面以年周期的季节变化为主,最高海面出现在9月份,最低海面出现在3月份。拉尼娜年份北太平洋海平面出现年变幅增大;1998/99年太平洋年代际涛动位相转变后海平面海平面上升减慢。ENSO和PDO位相为正时,北太平洋出现较低的海平面,ENSO和PDO位相为负时,北太平洋出现较高海平面;海平面与ENSO及PDO信号的变化位相相反,这种反位相相关关系在1998/99年ENSO及PDO出现位相转换前后尤为明显。北太平洋海平面前5个模态的方差比重为48.3%。第一模态和第三模态为季节模态,第二模态为ENSO模态,第四模态为ENSO和PDO混合模态。北太平洋海平面变化约有四分之一来自季节信号,四分之一来自ENSO和PDO共同影响作用,剩余的其他信号则为更高频或更低频的变化。北太平洋比容海面在1993～2006年间在上升,比容变化趋势对海平面上升的贡献大约为57%。北太平洋比容海面的显著周期与海平面一致且各种周期的振幅相当;其前五个模态的时空分布与海平面具有高度相似性,对海平面的季节、年际及趋势变化都有影响。1993～2006年间北太平洋西部的降水明显在增加,对海平面上升有贡献;中纬度海区大洋中部和低纬度海区大洋东部出现明显的降水减少,东边界附近的降水减少对于所在海区海平面下降也有一定的影响。1993～2006年北太平洋平均降水在增加,其增加率为0.21mm/a,大约对北太平洋海平面的上升贡献了7.4%。降水的季节变化与海平面的季节变化非常吻合。ENSO信号较强年份,降水对海平面的年际变化有负贡献。1993-2006年北太平洋海区蒸发有上升趋势,其增加率为0.09mm/a;蒸发对1998年海平面高值有负贡献。蒸发降水引起的淡水通量变化对海平面上升的正贡献大约为4.2%。1993-2006年间东海黑潮流域平均海平面呈上升趋势,上升速率为5.1mm/a。黑潮右侧大洋区域上升率大于左侧浅海区域。东海黑潮流域海平面变化具有显著的季节变化特征;ENSO和PDO处于正位相时东海黑潮流域海平面变化年较差较小;而二者处于负位相时海平面年较差较大;海平面的年际变化与PDO信号存在反位相关系。ENSO和PDO位相发生转变前后,海平面变化幅度较大。1995年东海黑潮流域海平面出现异常低值,2002年以后海平面上升减慢。比容对海平面季节变化的贡献大约为81.3%。1996年以前比容的年际变化与海平面存在较大差异,1997年以后二者变化较为吻合;2001年以前比容海面在缓慢上升,而2001年后比容海面突然降至较低水平,这可能是引起海平面上升减慢的一个原因。比容对海平面趋势变化的贡献为45.6%。降水对海平面变化的贡献主要是年际尺度的,降水的趋势变化对同期海平面上升的贡献约为1.2%。1995年东海黑潮流域海平面出现极低值,同期蒸发增强而降水减弱,对海平面的下降有正贡献。1998年和2001拉尼娜信号较强时,降水增强与蒸发减弱同时发生,对该时期海平面高值也有正贡献。黑潮流量增大时,东海黑潮左侧海平面下降,右侧海平面上升,左右两侧海面高度差增加,而东海黑潮流域的整体海平面上升。黑潮流量在1995年出现异常低值,海平面也出现异常低值,这是同期比容变化所无法解释的。黑潮延伸体南北两侧的海平面变化呈现不同的特征。黑潮延伸体北侧海平面变化在1993-2006年可以分成三个时段,1993-1998年下降,1999-2003年抬升后下降,2003年以后上升。该海域海平面具有显著的季节变化特征,季节变化引起的海平面年较差为18.6cm,占海平面变化量的60%。比容季节变化信号占海平面季节变化的84%,年际变化信号约占海平面年际变化的40%,对海平面趋势变化的贡献超过70%。降水引起的海平面变化主要体现在个别年份,1993-2006年黑潮延伸体北侧降水有减少趋势,对海平面趋势变化的贡献为2-3%。蒸发对海平面的贡献很微弱,对海平面下降的贡献不足1%。黑潮延伸体增强时,北侧海平面降低,黑潮延伸体减弱,北侧海平面升高。黑潮延伸体南侧海平面在1993-2006年期间也经历了三个阶段性变化,1993-1996年下降,1997-2003年上升,2003年以后下降。南侧海平面变化速率较北侧大。南侧海域也具有显著的季节变化特征,季节变化的年较差约为22.6cm,约占海平面变化信号的50%左右。比容的趋势变化约占海平面趋势变化的三分之一;季节变化可以解释海平面季节变化88.5%的信号,年际变化中比容的贡献约占三分之一。蒸发降水引起的海平面变化的比重很小。黑潮较强时南侧海平面较高,黑潮较弱时南侧海平面较低。但是对应关系比北侧要弱很多,可能与南侧较强的不稳定性有关。利用包含罗斯贝波和Sverdrup动力学的一层半模型可以很好的刻画风对北太平洋中纬度海平面变化的影响机制。大洋内区风驱动的罗斯贝波的传播对海平面变化的贡献存在区域差异。黑潮延伸体南侧海域海平面的季节变化中比容的作用是主要的,而年际和趋势变化中风的作用是主要的;黑潮延伸体北侧海平面的变化中比容在季节和趋势变化中占主导,而风的作用在年际变化中略微显著。POP模式模拟得到的97/98厄尔尼诺期间北太平洋海平面的变化可以用温度异常引起的比容效应和流的异常引起的输运变化来解释。不同强迫得到的实验结果对比发现,97/98厄尔尼诺事件中风的作用是主要的,风引起的流场及海面高度与气候态流场及海面高度场均有很大差异,而热通量引起的变化不显著。风的作用可以很好的解释赤道海区的海平面变化过程,而热通量对中高纬度海区海平面变化过程作用显著;就量值来说,风引起的海平面异常占厄尔尼诺期间海平面异常很大比重,而热通量的作用仅有20%左右。POP模式模拟得到1993-2006年全球比容海面上升趋势明显,年上升率为0.74mm/a;比容的快速上升始于1998年;上层200米比容变化作用显著。北太平洋海水总量变化引起的上升率为0.44mm/a,海面高度在1997年出现极低值。北太平洋比容海面1998年以前存在下降趋势,1998年以后开始快速上升,这一趋势变化与同期Ishii温盐资料计算得到的比容高度变化符合;利用随机动态分析得到模拟的北太平洋比容高度的上升速率为0.92mm/a。上层200米比容的趋势变化在比容海面的整体趋势变化中占有重要比重。模拟得到全球范围水体重新分布导致的北太平洋海平面变化大约占海平面趋势变化的15.3%;比容海面上升趋势大约占北太平洋海平面上升趋势的32.1%。二者可以解释北太平洋海平面趋势变化的47.4%。更多还原

【Abstract】 Sea level variation of the North Pacific Ocean during 1993 to 2006 is studied based on the AVISO altimeter data, and the most 5 important modes are given out. The contribution of steric effect, precipitation and evaporation to sea level variation is discussed. The sea level variation of the Kuroshio region of East Sea of China and Kuroshio Extension are also studied with a discussion of the contribution of steric effect, freshwater flux caused by precipitation and evaporation and flow intensity of Kuroshio. A two-layer ocean model is adopted that includes first-mode baroclinic Rossby wave dynamics and barotropic Sverdrup dynamics to investigate the contribution of wind-forcing to the sea level variation in the mid-latitude region of the North Pacific Ocean. The POP ocean model is applied in the simulation of global sea level variation to examine the dynamics of seasonal, inter-annual and long term sea level change.During 1993 to 2006, sea level rise occurs in the low and mid-latitude area of the western North Pacific Ocean basin while the low and mid-latitude area in the east of the basin and the high latitude area see sea level drop. The largest rise and drop rate occurs in the Kuroshio-Oyashio region. The average rate of the North Pacific sea level is 2.87mm/a. The seasonal sea level change is dominated by the annual cycle, with the highest and lowest sea level occurs in September and March. The annual sea level range increases in La Nina year and the sea level rise rate becomes smaller after the phase shift of Pacific Decadal Oscillation (PDO) in 1998/99. The Pacific Ocean has lower sea level during the positive phase of ENSO and PDO, and higher sea level during the negative phase of ENSO and PDO which is most obvious before and after the phase shift of 1998/99. The 5 most important Empirical Orthogonal Functions (EOF) can explain 48.3% of the sea level change in the Pacific Ocean with the first and the third EOFs being the seasonal modes , the second EOF being the ENSO mode and the fourth being a mixed mode of ENSO and PDO. The EOF decomposition shows one fourth of the sea level change is seasonal signal and one fourth is modulated by ENSO and PDO and the residual is higher or lower frequency signals.The steric sea level of the North Pacific Ocean is rising during 1993 to 2006 and the contribution of steric effect to sea level trend is 57%. The significant periods of the steric sea level is the same as the sea level in the North Pacific Ocean and the amplitudes of the oscillation on the significant periods are very close to each other. The 5 most important EOFs of steric sea level is similar to that of the sea level, which shows the steric effect can influence the sea level change in the North Pacific Ocean on both seasonal and inter-annual variability.During 1993 to 2006, precipitation in the western North Pacific Ocean is increasing which is making a contribution to the sea level rise in the area while the precipitation in the interior of the ocean in the mid-latitude and on the eastern boundary of the low latitude area is decreasing and the decreasing precipitation on the eastern boundary is contributing to the sea level drop in the area. The average precipitation rate of the North Pacific Ocean is 0.21mm/a which accounts for 7.4% of the sea level rate of the ocean. The seasonal variability of the precipitation is in favor of the seasonal sea level variability; the precipitation affects to the interannual sea level variation during the year with strong ENSO signals. The evaporation in the North Pacific Ocean during 1993 to 2006 is increasing at a rate of 0.09mm/a and it is in favor of the low sea level in 1998 and 2002. Contribution of the freshwater flux caused by precipitation and evaporation to the sea level trend is 4.2%.During 1993-2006, sea level of the Kuoshio region in the East Sea of China (ESC) is rising with a rate of 5.1mm/a and the rate on the right side of the Kuroshio is larger than the left side. The sea level of the Kuoshio region in the ESC shows prominent seasonal variability; the annual range is smaller during the positive phase of ENSO and PDO, and larger during the negative phase of them. The sea level changes in an opposite phase of PDO on the inter-annual timescale and the sea level shows large variation before and after the phase shift of ENSO and PDO. In 1995 the sea level drops dramatically and after 2002 the sea level rise becomes slower.The steric effect contributes about 81.3% to the seasonal variability of the sea level of the Kuoshio region in the ESC. Large difference can be found before 1996 between the sea level and steric sea level on the inter-annual timescale; low steric sea level after 2001 may be a reason for the decreasing of rising rate of the sea level. The steric sea level contributes 45.6% to the sea level trend. The precipitation contributes to the sea level change mainly on inter-annual timescale, and the precipitation rate accounts for 1.2% of the sea level rate. During 1995 when the sea level in the area drops dramatically, the precipitation weakens and the evaporation strengthens which is in favor of the sea level change; In the La Nina year of 1998 and 2002 when the area has high sea level, the precipitation strengthens while the evaporation weakens which may also contributes to the sea level change. During the strengthening period of the Kuroshio, the sea level on the left side usually drops while the sea level on the right side rises and the difference between them increases and the mean sea level in the Kuroshio region of ESC rises. The prominent low sea level occurs in 1995 can be explained by the dramatic decrease of the Kuroshio transport which is obviously not caused by steric effect.The sea level change south and north of the Kuroshio Extension(KE) is different. The sea level change on the north side of KE can be divided into three periods: the sea level drops during 1993-1998; in 1999 the sea level rises dramatically and then drops till 2003; after 2003 the sea level rises. The sea level on the north side of KE also shows prominent seasonal variability and the annual range is 18.6cm which accounts for about 60% of the sea level variation. The steric effect can explain 84% of the seasonal variability and 40% of the inter-annual variability of the sea level; the steric sea level trend contributes 70% to the sea level trend. The contribution of precipitation to sea level change is important in some years and the weakening of precipitation on the north side of KE contributes about 2-3% to the sea level trend. The contribution of evaporation is less than 1%. The sea level on the north side of KE drops as the flow strengthening and rises as the flow weakening.The sea level on the south side of the KE is also experienced three periods, the sea level drops during 1993-1996 and then rises during 1997-2006 and after 2003 the sea level drops. The sea level rate on the south side of KE is larger than north side. The seasonal variability caused an annual range of 22.6cm of sea level which accounts for about 50% of the sea level variability. The steric sea level contributes 88.5% to the seasonal sea level variability and one third to the inter-annual variability and trend of the sea level. The freshwater flux caused by precipitation and evaporation contributes little to sea level change. The sea level on the south side rises as the KE strengthening and drops as the KE weakening.The one and a half-layer ocean model shows the contribution of wind forcing to the sea level variation is different in different regions of the mid-latitude North Pacific Ocean. In the area south of the KE, the wind forced seasonal signal through barotropic Sverdrup transport is smaller than the inter-annual signals through boroclinic rossby wave; but in the area north of the KE, the seasonal variability is larger than the inter-annual variability caused by wind forcing. Considering the effect of steric sea level, the seasonal variability of the sea level south of the KE is mainly caused by steric effect while wind forcing is mainly causes the inter-annual and long term variability of the sea level. The seasonal and long term variability of the sea level north of the KE is mainly caused by steric effect and wind forcing is important in the inter-annual sea level variability.The anomalous sea level during 97/98 El Nino can be explained by the change of sea water temperature and transport caused by change of circulation. Controlling experiments show that wind dominates the 97/98 El Nino event; there is large difference in the current field and sea surface height between the El Nino event forced by real wind and climatology, while little difference can be found between the El Nino event forced by real heat flux and climatology. The sea level variation in the equatorial region can be well explained by wind forcing and the sea level variation in the mid and high latitude region is mainly caused by heat flux. The sea level variation during 97/98 El Nino caused by heat flux is about 20% and the residual is caused by wind.The simulated global mean steric sea level during 1993-2006 is rising at a rate of 0.74mm/a with most contribution coming from the upper 200m and the fast rising begins from 1998. The sea level rise of North Pacific Ocean caused by volume redistribution is 0.44mm/a; in 1997 the Pacific Ocean has a low sea level. The steric sea level of the North Pacific Ocean drops before 1998 and after 1998 begins to rise fast and this trend evolution is in agreement with the observation. The rising rate of the steric sea level of the North Pacific Ocean during 1993-2006 is 0.92mm/a and the contribution from the upper 200m is important. The sea level rise coming from water volume redistribution and steric effect can explain 47.4% of the sea level rise of the North Pacific Ocean.更多还原