Spatial and temporal variability of heat content above the thermocline in the tropical Pacific Ocean

Abstract-Heat content of the upper layer above the 20℃ isotherm in the tropical Pacific Ocean isestimated by using the sea temperature data set with a resolution 2°latitude×5°longitude (1980～1993)for the water depths (every 10 m) from 0 m to 400 m, and its temporal and spatial variabilities are an-alyzed. (1) The temporal variability indicates that the total heat in the upper layer of the equatorial Pa-cific Ocean is charcterized by the interannual variability. The time series of the equatorial heat anomaly5 months lead that of the El Nino index at the best positive lag correlation between the two, and theformer 13 months lag behind the latter at their best negative lag correlation. Therefore the equatorialheat content can be used as a better predictor than the El Nino index for a warm or cold event. In addi-tion, it is also found that less heat anomaly in the equator corresponds to the stronger warm events inthe period (1980～1993) and much more heat was accumulated in the 4 years including 1992/1     

A dipole mode at thermocline layer in the tropical Indian Ocean

Temperature data at different layers of the past 45 years were studied and we found adiploe mode in the thermocline layer (DMT): anomalously cold sea temperature off the coast of Sumatra and warm sea temperature in the western Indian Ocean. First, we analyzed the temperature and the temperature anomaly (TA) along the equatorial Indian Ocean in different layers. This shows that stronger cold and warm TA signals appeared at subsurface than at the surface in the tropical Indian O-cean. This result shows that there may be a strong dipole mode pattern in the subsurface tropical Indian Ocean. Secondly we used Empirical Orthogonal Functions (EOF) to analyze the TA at thermocline layer. The first EOF pattern was a dipole mode pattern. Finally we analyzed the correlations between DMT and surface tropical dipole mode (SDM), DMT and Nino 3 SSTA, etc. and these correlations are strong.     

Preliminary study on decadal variations of El Nino-associated adjustment in the tropical Pacific thermocline

Statistical analysis about ENSO index represented by SSTA in Nino3 with several datasets shows obviously decadal changes in the dominant period and amplitude of ENSO. Correlation analysis about the composite E1 Nino events before and after 1976 exhibits obviously decadal changes in the propagation and intensity of the oceanic anomaly related to the variation of SSTA in Nino3. In the composite E1 Nino before 1976, the coherence is relatively weak between the oceanic anomaly in the tropical Pacific and the SSTA in the Nino3 region; the area with significant correlation coefficient is relatively small; the oceanic anomaly related to Nino3 SSTA propagates faster. The above changes correspond well to the decadal changes of ENSO cycles. Some preliminary explanations are given based on the analysis of the decadal changes in the thermocline. The tropical thermocline shoals after 1976 except in the equatorial far East Pacific and the inclination of the tropical thermocline deep west and shallow east patterns weakens. Much of the oceanic anomaly from the relative higher latitude contributes to the slow propagation of the oceanic anomaly in the North Pacific. The air-sea coupling in the tropical Pacific intensifies after 1976 corresponding to decadai change of the ENSO amplitude. Decadai adjustment of the thermocline may have influenced the propagation and intensity of the oceanic anomaly related to the ENSO and intensity of air-sea interaction in the tropical Pacific, and changed the dominant period of ENSO and its amplitude.     

盐度对变化2014年东北太平洋“暖泡”的作用

A significant strong, warm "Blob"(a large circular water body with a positive ocean temperature anomaly)appeared in the Northeast Pacific(NEP) in the boreal winter of 2013–2014, which induced many extreme climate events in the US and Canada. In this study, analyses of the temperature and salinity anomaly variations from the Array for Real-time Geostrophic Oceanography(Argo) data provided insights into the formation of the warm"Blob" over the NEP. The early negative salinity anomaly dominantly contributed to the shallower mixed layer depth(MLD) in the NEP during the period of 2012–2013. Then, the shallower mixed layer trapped more heat in the upper water column and resulted in a warmer sea surface temperature(SST), which enhanced the warm"Blob". The salinity variability contributed to approximately 60% of the shallowing MLD related to the warm"Blob". The salinity anomaly in the warm "Blob" region resulted from a combination of both local and nonlocal effects. The freshened water at the surface played a local role in the MLD anomaly. Interestingly, the MLD anomaly was more dependent on the local subsurface salinity anomaly in the 100–150 m depth range in the NEP.The salinity anomaly in the 50–100 m depth range may be linked to the anomaly in the 100–150 m depth range by vertical advection or mixing. The salinity anomaly in the 100–150 m depth range resulted from the eastward transportation of a subducted water mass that was freshened west of the dateline, which played a nonlocal role.The results suggest that the early salinity anomaly in the NEP related to the warm "Blob" may be a precursor signal of interannual and interdecadal variabilities.     

Preliminary study on decadal variations of EI Nino-associated adjustment in the tropical Pacific thermocline

Statistical analysis about ENSO index represented by SSTA in Nino3 with several datasets shows obviously decadal changes in the dominant period and amplitude of ENSO. Correlation analysis about the composite El Nino events before and after 1976 exhibits obviously decadal changes in the propagation and intensity of the oceanic anomaly related to the variation of SSTA in Nino3. In the composite El Nino before 1976, the coherence is relatively weak between the oceanic anomaly in the tropical Pacific and the SSTA in the Nino3 region; the area with significant correlation coefficient is relatively small; the oceanic anomaly related to Nino3 SSTA propagates faster. The above changes correspond well to the decadal changes of ENSO cycles. Some preliminary explanations are given based on the analysis of the decadal changes in the thermocline. The tropical thermocline shoals after 1976 except in the equatorial far East Pacific and the inclination of the tropical thermocline deep west and shallow east patterns wea     

Salinity effects on the 2014 warm “Blob” in the Northeast Pacific

A significant strong, warm "Blob"(a large circular water body with a positive ocean temperature anomaly)appeared in the Northeast Pacific(NEP) in the boreal winter of 2013–2014, which induced many extreme climate events in the US and Canada. In this study, analyses of the temperature and salinity anomaly variations from the Array for Real-time Geostrophic Oceanography(Argo) data provided insights into the formation of the warm"Blob" over the NEP. The early negative salinity anomaly dominantly contributed to the shallower mixed layer depth(MLD) in the NEP during the period of 2012–2013. Then, the shallower mixed layer trapped more heat in the upper water column and resulted in a warmer sea surface temperature(SST), which enhanced the warm"Blob". The salinity variability contributed to approximately 60% of the shallowing MLD related to the warm"Blob". The salinity anomaly in the warm "Blob" region resulted from a combination of both local and nonlocal effects. The freshened water at the surface played a local role in the MLD anomaly. Interestingly, the MLD anomaly was more dependent on the local subsurface salinity anomaly in the 100–150 m depth range in the NEP.The salinity anomaly in the 50–100 m depth range may be linked to the anomaly in the 100–150 m depth range by vertical advection or mixing. The salinity anomaly in the 100–150 m depth range resulted from the eastward transportation of a subducted water mass that was freshened west of the dateline, which played a nonlocal role.The results suggest that the early salinity anomaly in the NEP related to the warm "Blob" may be a precursor signal of interannual and interdecadal variabilities.     

中国科学院气候系统模式模拟的ENSO循环

On the basis of more than 200-year control run, the performance of the climate system model of Chinese Academy of Sciences(CAS-ESM-C) in simulating the El Ni?o-Southern Oscillation(ENSO) cycle is evaluated, including the onset, development and decay of the ENSO. It is shown that, the model can reasonably simulate the annual cycle and interannual variability of sea surface temperature(SST) in the tropical Pacific, as well as the seasonal phase-locking of the ENSO. The model also captures two prerequisites for the El Ni?o onset, i.e., a westerly anomaly and a warm SST anomaly in the equatorial western Pacific. Owing to too strong forcing from an extratropical meridional wind, however, the westerly anomaly in this region is largely overestimated. Moreover, the simulated thermocline is much shallower with a weaker slope. As a result, the warm SST anomaly from the western Pacific propagates eastward more quickly, leading to a faster development of an El Ni?o. During the decay stage, owing to a stronger El Ni?o in the model, the secondary Gill-type response of the tropical atmosphere to the eastern Pacific warming is much stronger, thereby resulting in a persistent easterly anomaly in the western Pacific. Meanwhile, a cold anomaly in the warm pool appears as a result of a lifted thermocline via Ekman pumping. Finally, an El Ni?o decays into a La Ni?a through their interactions. In addition, the shorter period and larger amplitude of the ENSO in the model can be attributed to a shallower thermocline in the equatorial Pacific, which speeds up the zonal redistribution of a heat content in the upper ocean.     

基于浮标实测数据的WindSat海洋反演产品精度分析

To evaluate the ocean surface wind vector and the sea surface temperature obtained from Wind Sat, we compare these quantities over the time period from January 2004 to December 2013 with moored buoy measurements. The mean bias between the Wind Sat wind speed and the buoy wind speed is low for the low frequency wind speed product(WSPD_LF), ranging from –0.07 to 0.08 m/s in different selected areas. The overall RMS error is 0.98 m/s for WSPD_LF, ranging from 0.82 to 1.16 m/s in different selected regions. The wind speed retrieval result in the tropical Ocean is better than that of the coastal and offshore waters of the United States. In addition, the wind speed retrieval accuracy of WSPD_LF is better than that of the medium frequency wind speed product. The crosstalk analysis indicates that the Wind Sat wind speed retrieval contains some cross influences from the other geophysical parameters, such as sea surface temperature, water vapor and cloud liquid water. The mean bias between the Wind Sat wind direction and the buoy wind direction ranges from –0.46° to 1.19° in different selected regions. The overall RMS error is 19.59° when the wind speed is greater than 6 m/s. Measurements of the tropical ocean region have a better accuracy than those of the US west and east coasts. Very good agreement is obtained between sea surface temperatures of Wind Sat and buoy measurements in the tropical Pacific Ocean; the overall RMS error is only 0.36°C, and the retrieval accuracy of the low latitudes is better than that of the middle and high latitudes.     

冬季婆罗洲岛西北沿岸上升流的时空特征及机理研究

Winter coastal upwelling off northwest Borneo in the South China Sea(SCS) is investigated by using satellite data, climatological temperature and salinity fields and reanalysis data. The upwelling forms in December, matures in January, starts to decay in February and almost disappears in March. Both Ekman transport induced by the alongshore winter monsoon and Ekman pumping due to orographic wind stress curl are favorable for the upwelling. Transport estimates demonstrate that the month-to-month variability of Ekman transport and Ekman pumping are both consistent with that of winter coastal upwelling, but Ekman transport is two times larger than Ekman pumping in January and February. Under the influence of El Ni?o-Southern Oscillation(ENSO), the upwelling shows remarkable interannual variability: during winter of El Ni?o(La Ni?a) years, an anticyclonic(a cyclonic) wind anomaly is established in the SCS, which behaves a northeasterly(southwesterly) anomaly and a positive(negative) wind stress curl anomaly off the northwest Borneo coast, enhancing(reducing) the upwelling and causing anomalous surface cooling(warming) and higher(lower) chlorophyll concentration. The sea surface temperature anomaly(SSTA) associated with ENSO off the northwest Borneo coast has an opposite phase to that off southeast Vietnam, resulting in a SSTA seesaw pattern in the southern SCS in winter.     

The low frequency oscillations of the sea surface temperature in the Equatorial Eastern Pacific and El Nino formation

-In this paper the variations of the sea surface temperature anomalies (SSTA) in the Equatorial Eastern Pacific are analysed. The results show that there are two peaks in the spectrum. One is the low frequency oscillation with a period of 3 - 5 years, and the other is the quasi-biennial oscillation. The former shows a westward migration in the warm episode of SSTA and the latter has the opposite trend. The El Nino events will be formed while the two frquency bands are in phase in the warming stage of SSTA in the Equatorial Eastern Pacific     

赤道太平洋温度、流场距平EOF分析及与厄尔尼诺的关系

利用EOF分析方法,讨论了最近20a赤道太平洋次表层温度、纬向流距平与厄尔尼诺的关系.结果表明:赤道太平洋海温距平EOF分析第一、二主分量占总量的近80%,其中第一主分量类似于厄尔尼诺模态,第二主分量类似于暖池模态;后一模态存在着突变和渐变两种过程,其中由冷位相变暖位相过程为渐变过程,而暖位相变冷位相过程为突变过程.厄尔尼诺事件是赤道西太平洋暖池突变过程的结果.赤道太平洋纬向流距平EOF的第二主分量代表赤道西太平洋潜流和东太平洋南赤道流的变化,这个模态存在着半年左右的振荡和与厄尔尼诺同位相的年际振荡两种频率.另外,它还存在明显的年代际变化.赤道西太平洋潜流和东太平洋南赤道流减弱是产生厄尔尼诺的必要条件.统计回归分析表明,赤道太平洋海温距平和纬向流距平EOF的第二特征向量的时间系数对厄尔尼诺和拉尼娜均有一定的预报意义.     

西太平洋暖池时空分布特征及其与ENSO的关系

应用NCEP/NCAR SST资料和SODA海温资料,分析研究了热带太平洋海温场的变化特征,讨论了气候突变前后热带西太平洋暖池(以下简称WPWP)形态的显著变化及其差异,由此重新界定了WPWP的范围,并进一步分析了WPWP的时空变化特征。结果表明,新界定的WPWP气候平均场与前人定义的气候平均场分布特征基本相同,但也存在一定的差异。新界定的WPWP的优点在于它不仅能够客观反映出气候(海洋)突变前后西太平洋暖池的时空变化特征,而且重要的是可以避免由前人定义的WPWP与东太平洋暖池合为一体的现象发生,从而避免人为地计算WPWP面积变化带来的结果差异。新界定的WPWP平均深度可达130 m左右,呈现出西浅东深的"耳状"分布特征,在冬春季节,南北(经向)窄东西(纬向)宽,呈纬向带状分布;在夏秋季节,WP-WP明显向北扩展。平均深度最大中心位于(5°S,180°)附近。由WPWP区域不同深度的异常海温变化与Niño3指数的相关分析可知,WPWP次表层异常海温变化与Niño3指数呈显著的负相关关系,而与表层的异常海温的关系并不密切,这一结果进一步证明了西太平洋暖池对ENSO的贡献是来自次表层异常海温的东传。     

海洋的季节变化及风应力异常对海洋影响的数值试验

对一个6层5°×4°网格的全球海洋模式作了一些改进,建立了10层5°×4°网格的全球海洋模式,进行了季节变化数值模拟,积分250a,取得稳定的结果.除了高纬度海洋外,模拟的季节变化与实际观测十分接近.在此基础上,作了热带太平洋海温场对热带季风异常响应的3组敏感性实验,第1组为赤道西太平洋异常西风向东传播的试验;第2组为整个赤道太平洋风应力振荡异常试验;第3组为赤道西太平洋异常西风、东风交替向东传播的敏感性试验.模拟结果表明:(1)第1组风应力敏感性实验结果揭示出,西太平洋西风异常的向东传播的风应力异常可以产生类似厄尔尼诺的赤道东太平洋变暖;(2)第2组试验结果表明,热带太平洋风应力的局地振荡首先在中太平洋东西部激发出海温扰动,然后海温扰动分别向东太平洋和西太平洋传播,从而引起东、西太平洋海温的异常;(3)第3组试验验证风应力QBO可以产生海洋中类似的QBO振荡.     

全球百年海表面温度年际和年代际变化特征分析

利用100 a(1903—2002年)HadiSST的逐月资料,将全球海表面温度异常(SSTA)作为整体进行经验正交函数分解(EOF分解),提取了控制各大洋SSTA的主导模态和各大洋之间的联合模态,分析了相应的空间分布和时间序列。研究结果表明:SSTA变化最剧烈的海区是赤道中东太平洋、西北大西洋湾流海区和北太平洋黑潮延伸体海区。热带太平洋厄尔尼诺和南方涛动是主导模态并具有2~7 a周期的年际变化;SSTA变化第二主模态和第三主模态都是以大约70 a为周期的年代际变化为主的跨大洋联合模态。第二主模态的空间分布主要表现为中纬度北太平洋和北大西洋反位相、热带太平洋与大西洋反位相的太平洋-大西洋双偶极子型分布。SSTA变化的第三模态主要呈现南北半球海洋反位相的特征,为北太平洋-北大西洋-南大洋联合模态。第四模态基本上是反映各个不同海域特有的局地海洋-大气相互作用模态,该模态的时间序列具有1~4 a周期的年际变化和约9 a周期的年代际变化。     

热带太平洋-南海-印度洋海面风与海面温度年际变化整体耦合的主模态

利用COADS资料和相关分析、经验正交函数分解、奇异值分解和小波分析等方法对热带太平洋-南海-印度洋的海表面温度(SST)及海面风场(SSW)进行了研究,确定了热带太平洋、南海以及印度洋作为1个整体时海面风场及海面温度场相互作用的主模态。在把热带太平洋、南海和印度洋作为整体的情况下,研究了该模态的时空分布特征。各种方法研究均表明,该模态在热带太平洋上为ENSO模态;在热带印度洋上为对应于ENSO态的印度洋“单极型”模态;在南海则表现为整个海盆尺度上的一致型。     

西太平洋暖池变异及其对西太平洋次表层海温场的影响

应用热带太平洋上层XBT温度资料,分析研究了西太平洋暖池区(0°～16°N,125°～145°E)上层海洋的变化特征以及与西太平洋次表层海温场之间的关系.研究表明,西太平洋暖池区的垂向温度存在显著的年际变化,尤其在次表层(120～200m)的变化最为明显.西太平洋暖池区的次表层冷暖信号明显早于西太平洋次表层的海温异常.分析发现,西太平洋暖池区的海温异常是导致整个西太平洋次表层海温场变异的关键区,当西太平洋暖池区的次表层冷暖信号加强时,3～4个月后西太平洋海温场出现大范围的冷暖异常.     

热带太平洋海平面低频变化

本文选取ECMWF ORAS4再分析数据对1959-2015年热带太平洋海平面的低频变化进行了分析。热带太平洋海平面年际变化第一模态反映了ENSO爆发阶段的海平面变化,热带东、西太平洋变化反相,其时间序列与Niño3.4指数高度相关。海平面第二模态则体现了El Niño爆发前后热带太平洋暖水的输运过程。El Niño爆发前热带西太平洋暖水聚集的位置,以及爆发后暖水向赤道外输运的位置在两类El Niño事件中均有所不同。此外,ENSO的周期在近半个世纪发生了显著的年代际变化,这一变化与热带太平洋的年代际变化有关。热带太平洋的年代际变化对海平面趋势变化也有着显著的影响。卫星高度计观测到的近20年海平面的快速上升（下降）正是由20世纪90年代后热带太平洋频繁的位相转换引起的。     

两类El Niño事件对中国近海及毗邻海域海表温度的影响

为研究东部型和中部型两类厄尔尼诺(El Ni?o)事件与中国近海海表温度(sea surface temperature,SST)变化间的联系,基于中国科学院大气物理研究所连续80年(1940—2019年)的SST再分析数据,采用EOF分解、合成分析等方法做了初步分析,发现中国近海及毗邻海域近80年SST变化与全球变暖密切相关。并且两类El Ni?o事件对中国近海SST变化的影响存在显著差异。东部型El Ni?o事件发展过程中,中国近海及毗邻海域SST在发展年主要为负异常,衰退期为正异常;中部型ElNi?o事件发展过程中, SST变化区域差异大,发展年日本附近海域为正异常, 28°N以南为弱的负异常。两类El Ni?o事件引发西太平洋风场反气旋涡的时间、位置与强度等的不同,是造成中国近海风场与海表温度异常(sea surface temperature anomaly, SSTA)差异的主要原因。     

全球气候模式中赤道东太平洋海表温度的季节循环：从CMIP5到CMIP6

The sea surface temperature(SST) seasonal cycle in the eastern equatorial Pacific(EEP) plays an important role in the El Ni?o–Southern Oscillation(ENSO) phenomenon. However, the reasonable simulation of SST seasonal cycle in the EEP is still a challenge for climate models. In this paper, we evaluated the performance of 17 CMIP6 climate models in simulating the seasonal cycle in the EEP and compared them with 43 CMIP5 climate models. In general, only CESM2 and SAM0-UNICON are able to successfully capture the annual mean SST characteristics,and the results showed that CMIP6 models have no fundamental improvement in the model annual mean bias.For the seasonal cycle, 14 out of 17 climate models are able to represent the major characteristics of the observed SST annual evolution. In spring, 12 models capture the 1–2 months leading the eastern equatorial Pacific region 1(EP1; 5°S–5°N, 110°–85°W) against the eastern equatorial Pacific region 2(EP2; 5°S–5°N, 140°–110°W). In autumn,only two models, GISS-E2-G and SAM0-UNICON, correctly show that the EP1 and EP2 SSTs vary in phase. For the CMIP6 MME SST simulation in EP1, both the cold bias along the equator in the warm phase and the warm bias in the cold phase lead to a weaker annual SST cycle in the CGCMs, which is similar to the CMIP5 results. However,both the seasonal cold bias and warm bias are considerably decreased for CMIP6, which leads the annual SST cycle to more closely reflect the observation. For the CMIP6 MME SST simulation in EP2, the amplitude is similar to the observed value due to the quasi-constant cold bias throughout the year, although the cold bias is clearly improved after August compared with CMIP5 models. Overall, although SAM0-UNICON successfully captured the seasonal cycle characteristics in the EEP and the improvement from CMIP5 to CMIP6 in simulating EEP SST is clear, the fundamental climate models simulated biases still exist.