Advanced Search
Article Contents

Dissimilarity among Ocean Reanalyses in Equatorial Pacific Upper-Ocean Heat Content and Its Relationship with ENSO

Fund Project:

This work is fully supported by the National Natural Science Foundation (Grant No. 42120104001)


doi: 10.1007/s00376-021-1109-8

  • This study focuses on the temporal variation of dissimilarity in heat content (HC) anomalies in the upper 300 m of ocean (HC300A) in the equatorial Pacific (±10°N) and its response to the El Niño-Southern Oscillation (ENSO). The HC300A anomalies are derived from four ocean reanalyses that are commonly used in ENSO studies and are compared using a simple differencing method. The dissimilarity in HC300A is found to vary closely with the magnitude of ENSO (regardless of phase), meaning that it tends to be greater during strong ENSO events. However, the dissimilarity among ocean reanalyses persists after the event decays. This effect is more pronounced after strong events. The persistence of the dissimilarity after ENSO events is a result of a late maturation of the ENSO signal, its persistence, and the interruption of the signal decay due to follow-up ENSO events. The combined effect of these three factors slows down the decay of HC300A in the region and hence results in the slow decay of dissimilarity. It is also found that areas with a significant spread in vertical temperature profiles collocate with the ENSO signal during warm ENSO phases. Thus, differences in subsurface process reconstruction are a significant factor in the dissimilarity among ocean reanalyses during warm ENSO events.
    摘要: 海洋再分析数据的准确性对研究和理解厄尔尼诺—南方涛动现象(ENSO)的机理十分重要。本文发现不同海洋再分析数据之间在赤道太平洋(±10°N)上层300米海洋热含量(HC300A)的差异随ENSO生命周期有显著的变化。在ENSO发展期,不同海洋再分析数据间HC300A差异会随ENSO增强而增大。在ENSO衰退期,HC300A的差异随ENSO强度衰减而减小。不同海洋再分析数据间的差异减小速率比增大速率小,存在不对称性。并且,我们进一步发现这种资料间差异变化速率的不对称性跟ENSO强度有关。强ENSO事件期间数据间差异的速率不对称性会比弱ENSO事件期间的更大。同海洋再分析数据间HC300A差异与赤道太平洋内海洋热含量信号的强度关系密切。赤道太平洋内海洋热含量信号强度愈强,数据间的差异愈大。海洋热含量信号在ENSO发展期快速增长,但ENSO事件衰退后信号强度缓慢减弱。在强ENSO事件衰退期后,信号强度衰减速率可能因为后续ENSO事件而变得更慢。赤道太平洋内海洋热含量信号跟ENSO生命周期的关系,导致不同数据间HC300A差异随ENSO生命周期发生显著的变化。
  • 加载中
  • Figure 1.  The top 350 meters of ocean temperature variability of different ocean reanalyses in the equatorial Pacific Ocean (140°E to 90°W, meridional average from 10°S to 10°N) A 7-year high-pass filter was applied for the whole study period. The magenta dashed line in each plot marks the 300-meter level. All ocean reanalyses show that ocean temperature varies the most in the top 300 meters of the ocean.

    Figure 2.  The three-month moving average of standardized HC300A in February 2008 derived using SODA. Despite the use of the three-month moving average, short waves can still be observed (here, in the off-equatorial central Pacific). These waves are irrelevant to ENSO and are filtered out to reduce their influence on the dissimilarity measurement.

    Figure 3.  Ranked HC300A in April 1998 after data pretreatments for each of the four reanalysis products. Signals irrelevant to ENSO (see the text) are filtered out.

    Figure 4.  (a) Time series of dissimilarity in HC300A in the equatorial Pacific derived from each ocean reanalysis to the others (colored solid lines) and the absolute value of Niño-3.4 SSTA (dashed black line). A nine-month moving average is applied to the all-time series. It is clear that the dissimilarity time series, regardless of which product, closely follows the absolute Niño-3.4 SSTA most of the time. Exceptions can be found after strong ENSO events, in which the dissimilarity persists. (b) Average dissimilarity in HC300A among reanalyses in the inner-equatorial region (solid red line), off-equatorial regions (solid blue line), and absolute Niño-3.4 SSTA (dashed black line). Note that the two dissimilarity time series are on different scales.

    Figure 5.  Rate of change of average dissimilarity among reanalyses (blue bars) and absolute value of Niño-3.4 SSTA (red bars). The average dissimilarity shows moderate positive skew (0.507), while the absolute value of Niño-3.4 SSTA shows negligible skew (–0.086). Dissimilarity tends to grow faster than it decays. ENSO state tends to grow and decay at a similar rate.

    Figure 6.  Dissimilarity (solid line) and signal strength (dashed line) in the equatorial region (magenta) and off-equatorial region (cyan).

    Figure 7.  Average dissimilarity among ocean reanalyses in HC300A in the equatorial Pacific throughout the study period. Average dissimilarity is lower in the inner-equatorial region and off-equatorial western Pacific. However, the dissimilarity in the off-equatorial central and eastern Pacific is higher.

    Figure 8.  Lagged-correlation maps of HC300A (shading) and standardized HC300A spread anomaly (dotted and meshed areas) to the absolute value of D(0)JF(1) Niño-3.4 SSTA. Here, only years with positive D(0)JF(1) Niño-3.4 SSTA (14 years in total) are used in the calculation of correlation. Red shading and dots signify areas with positive correlation. Blue shading and mesh signify areas with negative correlation. Areas with correlations less than the 90% confidence level are removed.

    Figure 9.  Same as Fig. 7 except only the years with negative D(0)JF(1) Niño-3.4 SSTA (14 years in total) are used in the calculation of correlation. The absolute value of Niño-3.4 SSTA is used in the calculation.

    Figure 10.  Lagged correlation of standardized ocean temperature anomaly (OTA) (shading) and spread among the reanalyses on OTA to ND(0)J(1) Niño-3.4 SSTA (dots and meshes). Red and blue shading denote the areas where the lagged correlation of OTA is positive and negative, respectively. Dots and meshes indicate the areas where the lagged correlation of the spread is positive and negative, respectively. Only the years that end with a warm event are considered. Figures on the left represent the evolution in the inner-equatorial region. Figures on the right represent the evolution in the off-equatorial region. Only the statistically significant signal are drawn (test level of 10%).

    Figure 11.  Continuum of Fig. 10. The years after a warm event are considered in the calculation of correlation.

    Table 1.  Ranks of signal strength and rank numbers

    Rank numberDescriptionRange of standardized HC300A
    −2Strong cold signalHC300A ≤ −2
    −1Cold signal−2 < HC300A ≤ −1
    0No signal−1 < HC300A < +1
    1Warm signal+1 ≤ HC300A < +2
    2Strong warm signalHC300A ≥ +2
    DownLoad: CSV
  • Balmaseda, M. A., K. Mogensen, and A. T. Weaver, 2013: Evaluation of the ECMWF ocean reanalysis system ORAS4. Quart. J. Roy. Meteor. Soc., 139, 1132−1161, https://doi.org/10.1002/qj.2063.
    Behringer, D. W., and Y. Xue, 2004: Evaluation of the global ocean data assimilation system at NCEP: The Pacific Ocean. Eighth Symposium on Integrated Observing and Assimilation Systems for Atmosphere, Oceans, and Land Surface, AMS 84th Annual Meeting, Washington State Convention and Trade Center, Seattle, Washington, 11−15.
    Carton, J. A., and B. S. Giese, 2008: A reanalysis of ocean climate using Simple Ocean Data Assimilation (SODA). Mon. Wea. Rev., 136, 2999−3017, https://doi.org/10.1175/2007MWR1978.1.
    Hu, D., and Coauthors, 2020: Review on observational studies of western tropical Pacific Ocean circulation and climate. Journal of Oceanology and Limnology, 38(4), 906−929, https://doi.org/10.1007/s00343-020-0240-1.
    Hu, Z. Z., A. Kumar, Y. Xue, and B. Jha, 2014: Why were some La Niñas followed by another La Niña. Climate Dyn., 42, 1029−1042, https://doi.org/10.1007/s00382-013-1917-3.
    Hua, L. J., and Y. Q. Yu, 2015: The asymmetry of subsurface temperature anomalies associated with ENSO in the equatorial western Pacific. Atmospheric and Oceanic Science Letters, 8, 23−28, https://doi.org/10.3878/AOSL20140067.
    Jin, F. F., 1997: An equatorial ocean recharge paradigm for ENSO. Part I: Conceptual model. J. Atmos. Sci., 54, 811−829, https://doi.org/10.1175/1520-0469(1997)054<0811:AEORPF>2.0.CO;2.
    McPhaden, M. J., A. J. Busalacchi, and D. L. T. Anderson, 2010: A TOGA retrospective. Oceanography, 23, 86−103, https://doi.org/10.5670/oceanog.2010.26.
    Suarez, M. J., and P. S. Schopf, 1988: A delayed action oscillator for ENSO. J. Atmos. Sci., 45, 3283−3287, https://doi.org/10.1175/1520-0469(1988)045<3283:ADAOFE>2.0.CO;2.
    Wang, B., R. G. Wu, and R. Lukas, 1999: Roles of the western North Pacific wind variation in thermocline adjustment and ENSO phase transition. J. Meteor. Soc. Japan. Ser. II, 77, 1−16, https://doi.org/10.2151/jmsj1965.77.1_1.
    Xue, Y., T. M. Smith, and R. W. Reynolds, 2003: Interdecadal changes of 30-yr SST normals during 1871−2000. J. Climate, 16, 1601−1612, https://doi.org/10.1175/1520-0442(2003)016<1601:ICOYSN>2.0.CO;2.
    Xue, Y., and Coauthors, 2012: A comparative analysis of upper-ocean heat content variability from an ensemble of operational ocean reanalyses. J. Climate, 25, 6905−6929, https://doi.org/10.1175/JCLI-D-11-00542.1.
    Zhang, S., M. J. Harrison, A. Rosati, and A. Wittenberg, 2007: System design and evaluation of coupled ensemble data assimilation for global oceanic climate studies. Mon. Wea. Rev., 135, 3541−3564, https://doi.org/10.1175/MWR3466.1.
    Zhou, W., and J. C. L. Chan, 2007: ENSO and the South China Sea summer monsoon onset. International Journal of Climatology, 27, 157−167, https://doi.org/10.1002/joc.1380.
  • [1] CHEN Xiao, YAN Youfang, CHENG Xuhua, QI Yiquan, 2013: Performances of Seven Datasets in Presenting the Upper Ocean Heat Content in the South China Sea, ADVANCES IN ATMOSPHERIC SCIENCES, 30, 1331-1342.  doi: 10.1007/s00376-013-2132-1
    [2] LIN Pengfei, LIU Hailong, ZHANG Xuehong, 2007: Sensitivity of the Upper Ocean Temperature and Circulation in the Equatorial Pacific to Solar Radiation Penetration Due to Phytoplankton, ADVANCES IN ATMOSPHERIC SCIENCES, 24, 765-780.  doi: 10.1007/s00376-007-0765-7
    [3] Xiaoxuan ZHAO, Riyu LU, 2020: Vertical Structure of Interannual Variability in Cross-Equatorial Flows over the Maritime Continent and Indian Ocean in Boreal Summer, ADVANCES IN ATMOSPHERIC SCIENCES, 37, 173-186.  doi: 10.1007/s00376-019-9103-0
    [4] Changyu LI, Jianping HUANG, Yongli HE, Dongdong LI, Lei DING, 2019: Atmospheric Warming Slowdown during 1998−2013 Associated with Increasing Ocean Heat Content, ADVANCES IN ATMOSPHERIC SCIENCES, 36, 1188-1202.  doi: 10.1007/s00376-019-8281-0
    [5] Yang AI, Ning JIANG, Weihong QIAN, Jeremy Cheuk-Hin LEUNG, Yanying CHEN, 2022: Strengthened Regulation of the Onset of the South China Sea Summer Monsoon by the Northwest Indian Ocean Warming in the Past Decade, ADVANCES IN ATMOSPHERIC SCIENCES, 39, 943-952.  doi: 10.1007/s00376-021-1364-8
    [6] LI Gang*, LI Chongyin, TAN Yanke, and BAI Tao, 2014: The Interdecadal Changes of South Pacific Sea Surface Temperature in the Mid-1990s and Their Connections with ENSO, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 66-84.  doi: 10.1007/s00376-013-2280-3
    [7] Xiaofei WU, Jiangyu MAO, 2019: Decadal Changes in Interannual Dependence of the Bay of Bengal Summer Monsoon Onset on ENSO Modulated by the Pacific Decadal Oscillation, ADVANCES IN ATMOSPHERIC SCIENCES, 36, 1404-1416.  doi: 10.1007/s00376-019-9043-8
    [8] Shang-Ping XIE, Yu KOSAKA, Yan DU, Kaiming HU, Jasti S. CHOWDARY, Gang HUANG, 2016: Indo-Western Pacific Ocean Capacitor and Coherent Climate Anomalies in Post-ENSO Summer: A Review, ADVANCES IN ATMOSPHERIC SCIENCES, 33, 411-432.  doi: 10.1007/s00376-015-5192-6
    [9] KANG Xianbiao, HUANG Ronghui, WANG Zhanggui, ZHANG Rong-Hua, 2014: Sensitivity of ENSO Variability to Pacific Freshwater Flux Adjustment in the Community Earth System Model, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 1009-1021.  doi: 10.1007/s00376-014-3232-2
    [10] ZHENG Fei, ZHANG Rong-Hua, ZHU Jiang, , 2014: Effects of Interannual Salinity Variability on the Barrier Layer in the Western-Central Equatorial Pacific: A Diagnostic Analysis from Argo, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 532-542.  doi: 10.1007/s00376-013-3061-8
    [11] ZHOU Qian, DUAN Wansuo, MU Mu, FENG Rong, 2015: Influence of Positive and Negative Indian Ocean Dipoles on ENSO via the Indonesian Throughflow: Results from Sensitivity Experiments, ADVANCES IN ATMOSPHERIC SCIENCES, 32, 783-793.  doi: 10.1007/s00376-014-4141-0
    [12] ZHANG Qiong, GUAN Yue, YANG Haijun, 2008: ENSO Amplitude Change in Observation and Coupled Models, ADVANCES IN ATMOSPHERIC SCIENCES, 25, 361-366.  doi: 10.1007/s00376-008-0361-5
    [13] ZHENG Weipeng, YU Yongqiang, 2007: ENSO Phase-Locking in an Ocean-tmosphere Coupled Model FGCM-1.0, ADVANCES IN ATMOSPHERIC SCIENCES, 24, 833-844.  doi: 10.1007/s00376-007-0833-z
    [14] FENG Junqiao, HU Dunxin, YU Lejiang, 2013: Role of Western Pacific Oceanic Variability in the Onset of the Bay of Bengal Summer Monsoon, ADVANCES IN ATMOSPHERIC SCIENCES, 30, 219-234.  doi: 10.1007/s00376-012-2040-9
    [15] WANG Zhiren, WU Dexing, CHEN Xue'en, QIAO Ran, 2013: ENSO Indices and Analyses, ADVANCES IN ATMOSPHERIC SCIENCES, 30, 1491-1506.  doi: 10.1007/s00376-012-2238-x
    [16] FU Weiwei, ZHOU Guangqing, WANG Huijun, 2006: Modeling the Tropical Pacific Ocean Using a Regional Coupled Climate Model, ADVANCES IN ATMOSPHERIC SCIENCES, 23, 625-638.  doi: 10.1007/s00376-006-0625-x
    [17] LI Chun, WU Lixin, 2013: Dynamic Linkage between the North Pacific and the Tropical Pacific: Atmosphere-Ocean Coupling, ADVANCES IN ATMOSPHERIC SCIENCES, 30, 306-314.  doi: 10.1007/s00376-012-2023-x
    [18] ZHANG Rong-Hua, 2015: A Hybrid Coupled Model for the Pacific Ocean-Atmosphere System. Part I: Description and Basic Performance, ADVANCES IN ATMOSPHERIC SCIENCES, 32, 301-318.  doi: 10.1007/s00376-014-3266-5
    [19] XIE Pinhua, LIU Wenqing, FU Qiang, WANG Ruibin, LIU Jianguo, WEI Qingnong, 2004: Intercomparison of Nox,SO2,O3,and Aromatic Hydrocarbons Measured by a Commercial DOAS System and Traditional Point Monitoring Techniques, ADVANCES IN ATMOSPHERIC SCIENCES, 21, 211-219.  doi: 10.1007/BF02915707
    [20] Yiran GUO, Jie CAO, Hui LI, Jian WANG, Yuchao DING, 2016: Simulation of the Interface between the Indian Summer Monsoon and the East Asian Summer Monsoon: Intercomparison between MPI-ESM and ECHAM5/MPI-OM, ADVANCES IN ATMOSPHERIC SCIENCES, 33, 294-308.  doi: 10.1007/s00376-015-5073-z

Get Citation+

Export:  

Share Article

Manuscript History

Manuscript received: 15 March 2021
Manuscript revised: 10 July 2021
Manuscript accepted: 04 August 2021
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Dissimilarity among Ocean Reanalyses in Equatorial Pacific Upper-Ocean Heat Content and Its Relationship with ENSO

    Corresponding author: Wen ZHOU, wenzhou@cityu.edu.hk
  • 1. Guy-Carpenter Asia-Pacific Climate Impact Centre, School of Energy and Environment, City University of Hong Kong, Hong Kong, China
  • 2. Department of Atmospheric and Oceanic Sciences & Institute of Atmospheric Sciences, Fudan University, Shanghai 200438, China
  • 3. School of Marine Sciences, Sun Yat-sen University, Guangdong 510275, China

Abstract: This study focuses on the temporal variation of dissimilarity in heat content (HC) anomalies in the upper 300 m of ocean (HC300A) in the equatorial Pacific (±10°N) and its response to the El Niño-Southern Oscillation (ENSO). The HC300A anomalies are derived from four ocean reanalyses that are commonly used in ENSO studies and are compared using a simple differencing method. The dissimilarity in HC300A is found to vary closely with the magnitude of ENSO (regardless of phase), meaning that it tends to be greater during strong ENSO events. However, the dissimilarity among ocean reanalyses persists after the event decays. This effect is more pronounced after strong events. The persistence of the dissimilarity after ENSO events is a result of a late maturation of the ENSO signal, its persistence, and the interruption of the signal decay due to follow-up ENSO events. The combined effect of these three factors slows down the decay of HC300A in the region and hence results in the slow decay of dissimilarity. It is also found that areas with a significant spread in vertical temperature profiles collocate with the ENSO signal during warm ENSO phases. Thus, differences in subsurface process reconstruction are a significant factor in the dissimilarity among ocean reanalyses during warm ENSO events.

摘要: 海洋再分析数据的准确性对研究和理解厄尔尼诺—南方涛动现象(ENSO)的机理十分重要。本文发现不同海洋再分析数据之间在赤道太平洋(±10°N)上层300米海洋热含量(HC300A)的差异随ENSO生命周期有显著的变化。在ENSO发展期,不同海洋再分析数据间HC300A差异会随ENSO增强而增大。在ENSO衰退期,HC300A的差异随ENSO强度衰减而减小。不同海洋再分析数据间的差异减小速率比增大速率小,存在不对称性。并且,我们进一步发现这种资料间差异变化速率的不对称性跟ENSO强度有关。强ENSO事件期间数据间差异的速率不对称性会比弱ENSO事件期间的更大。同海洋再分析数据间HC300A差异与赤道太平洋内海洋热含量信号的强度关系密切。赤道太平洋内海洋热含量信号强度愈强,数据间的差异愈大。海洋热含量信号在ENSO发展期快速增长,但ENSO事件衰退后信号强度缓慢减弱。在强ENSO事件衰退期后,信号强度衰减速率可能因为后续ENSO事件而变得更慢。赤道太平洋内海洋热含量信号跟ENSO生命周期的关系,导致不同数据间HC300A差异随ENSO生命周期发生显著的变化。

    • Ocean heat content (HC) is one of the most widely used ocean parameters in El Niño-Southern Oscillation (ENSO) studies, such as studies of ENSO phase transition and climate impacts (Wang et al., 1999; Zhou and Chan, 2007; Xue et al., 2012; Hu et al., 2014). It is defined as the vertical integration of ocean temperature from the surface to a certain depth [depending on purpose; Eq. (1)]. Since it includes subsurface temperature variability rather than only sea surface temperature variability, it can be used to highlight subsurface wave activity within the integrating column. Wave activity in the equatorial Pacific is an essential part of ENSO variability (Suarez and Schopf, 1988; Jin, 1997). Its physical significance in ENSO dynamics on an interannual time scale is emphasized in the Recharge-Discharge Oscillator Model (Jin, 1997). In this model, the ocean heat content in the Pacific basin is described as being transported in and out of the equatorial Pacific, leading to the ENSO cycle.

      Computation of HC requires vertical ocean temperature profiles within the range of depth in question. Computing HC distribution in the Pacific basin simply requires temperature profiles to be known throughout the whole basin. However, in-situ measurements are not always available and do not always cover the whole basin densely, especially prior to the completion of the Tropical Ocean-Global Atmosphere program [TOGA, completed in 1995 (McPhaden et al., 2010)]. Studying HC derived from measurements from only the pre-TOGA period may not be a reliable option. Ocean reanalyses provide temperature profiles that densely cover the whole Pacific basin after assimilating available data.

      Xue et al. (2012) investigated the difference between HC patterns derived from different ocean reanalyses. In their study, 10 near real-time ocean reanalyses were compared for their HC patterns in different ocean basins. It was found that the spread in the equatorial Pacific HC (±2°N, for ocean heat content from the surface to 300 meters below) were among those ocean reanalyses that varied from time to time. In particular, the ocean reanalyses diverge more than usual on HC during strong El Niño episodes in 1982−83 and 1997−98. However, they did not comment on the spread in HC other than these two episodes.

      While Xue et al. (2012) provided extensive insights into the consistency in equatorial Pacific Ocean HC among ocean reanalyses, for the sake of ENSO studies, more information concerning this topic is needed. In particular, it would be useful to know how the similarity among ocean reanalyses in off-equatorial HC changes over time. Off-equatorial processes are considered to be an important part of ENSO variability (Wang et al., 1999; Hu et al., 2014; Hua and Yu, 2015). Hu et al. (2014) studied the transitional properties of La Niña with HC covering ±6°N to show how a La Niña event can take place after another La Niña event. Hua and Yu (2015) reported that ENSO-related subsurface signals can be found up to 10°N. It is important to understand the similarity in reconstructed HC patterns among ocean reanalyses in off-equatorial regions as well.

      In this study, the intercomparison of HC anomalies in the Pacific Ocean is extended to ±10°N to cover off-equatorial processes. HC anomalies are computed for the upper 300 m (HC300A hereafter). Four ocean reanalysis products are chosen because of their frequent use in the field of study. To identify periods in which HC300A patterns are less similar, HC300A patterns derived from different ocean profiles are compared against each other every month. The variation of dissimilarity will also be explained.

      It should be emphasized that this study does not aim to judge the reliability of the chosen ocean reanalyses. Instead, it is intended to discover the scenarios in which the HC300A patterns in the equatorial Pacific from different ocean reanalyses tend to be less similar to each other. Second, this study will not explain the observed dissimilarity among ocean reanalysis products from a data assimilation point of view. Again, the main objective of this study is to highlight situations in which ocean reanalyses show greater divergence.

    2.   Data and methodology
    • Four ocean reanalyses are selected in this study. In alphabetical order, they are the Ocean Data Assimilation system of the Geophysical Fluid Dynamics Laboratory (GFDL) (Zhang et al., 2007), the Global Ocean Data Assimilation System (GODAS) of the National Centers for Environmental Prediction (Behringer and Xue, 2004), the Ocean Reanalysis System 4 (ORAs4) of the European Centre for Medium-Range Weather Forecasts (Balmaseda et al., 2013), and the Simple Ocean Data Assimilation (SODA) version 2.2.4 of the Department of Atmospheric and Oceanic Science at the University of Maryland and the Department of Oceanography at Texas A&M University (Carton and Giese, 2008). These reanalysis products are selected because of their frequent use in ENSO studies. Only monthly data are used. The NOAA Extended Reconstructed Sea Surface Temperature V3b (ERSST; Xue et al., 2003) is used to calculate sea surface temperature anomalies (SSTAs) in the Niño-3.4 region (5°S−5°N, 190°−240°E) as an index of ENSO magnitude.

      The study domain is (140°E to 90°W, 10°S to 10°N). As mentioned above, rather than focusing only on the narrow equatorial Pacific, this study also covers off-equatorial regions because of their role in ENSO variability (Hu et al., 2014; Hua and Yu, 2015). Subsequently, this study domain will be further divided into inner-equatorial (±5°N) and off-equatorial regions (5°–10°N and 5°–10°S). The study period is from January 1980 to December 2008, which is the common temporal coverage of selected ocean reanalyses at the time of this study.

      In this study, ocean heat content (HC) is defined by the following equation:

      where h is the depth below the sea surface where the vertical integration ends, ρ is the density of seawater, cp is the specific heat capacity of seawater, and T is the temperature of seawater. The ending depth of the integration (h) is chosen to be 300 m below the sea surface, which is the same as that used by Xue et al. (2012) and Hu et al. (2014). Other studies have used different integrating depths for HC ranging from 350 m to 400 m or more (e.g., Wang et al., 1999; Zhou and Chan, 2007). However, as HC anomalies are dominated by temperature variations along the thermocline between 50 m and 250 m below the sea surface (Fig. 1), adding extra integrating depth below the thermocline in the computation will not significantly affect the findings. Different ocean reanalyses have different climatological mean states as well as variability in ocean temperature (Xue et al., 2012). In order to account for these differences in the comparison, standardized anomalies are used. This standardization is done in each grid cell. Let x denote the raw values of ocean heat content in a certain grid cell, which itself is a time series with n time steps. The standardized anomaly of ocean heat content, denoted by $\left\langle{x}\right\rangle $, is

      Figure 1.  The top 350 meters of ocean temperature variability of different ocean reanalyses in the equatorial Pacific Ocean (140°E to 90°W, meridional average from 10°S to 10°N) A 7-year high-pass filter was applied for the whole study period. The magenta dashed line in each plot marks the 300-meter level. All ocean reanalyses show that ocean temperature varies the most in the top 300 meters of the ocean.

      where

      Very short, high-amplitude waves are found in some ocean reanalysis products. Figure 2 shows a snapshot of HC300A averaged from January to March 2008 from SODA. In this figure, short waves can still be observed in the off-equatorial central Pacific even though a three-month moving average has already been applied. These waves are considered not irrelevant to ENSO. To provide information relevant to ENSO studies, a set of data pretreatment procedures is employed to filter out such signals from the raw data. Details of the filtering procedure are as follows:

      Figure 2.  The three-month moving average of standardized HC300A in February 2008 derived using SODA. Despite the use of the three-month moving average, short waves can still be observed (here, in the off-equatorial central Pacific). These waves are irrelevant to ENSO and are filtered out to reduce their influence on the dissimilarity measurement.

      First, to remove high-frequency (sub-seasonal) signals, a three-month moving average is performed. Second, to remove the strong short waves seen in the raw data, an area filtering technique is applied. The HC300A field is linearly re-gridded into a coarse 2° (latitude) × 5° (longitude) data grid. Individual signals with a size of less than four grid boxes are removed.

      The HC300A field is then ranked according to the magnitude in each grid (Table 1). The ranking is made to reduce the influence of the magnitude difference in the re-created HC300A among reanalyses on the measure of dissimilarity. As a result, the analysis will be influenced to a greater extent by differences in the HC300A signal distribution, which is of interest in this study.

      Rank numberDescriptionRange of standardized HC300A
      −2Strong cold signalHC300A ≤ −2
      −1Cold signal−2 < HC300A ≤ −1
      0No signal−1 < HC300A < +1
      1Warm signal+1 ≤ HC300A < +2
      2Strong warm signalHC300A ≥ +2

      Table 1.  Ranks of signal strength and rank numbers

      Figure 3 shows a snapshot of HC300A after the treatment in April 1998, the decay phase of the 1997−98 El Niño. In general, a large area of positive HC300A (anomalously deep thermocline) can be observed in the east. In the west, the thermocline is anomalously shallow, shown as a large patch of negative HC300A on the plot. It is evident that the use of different ocean reanalyses to generate HC300A can yield different results.

      Figure 3.  Ranked HC300A in April 1998 after data pretreatments for each of the four reanalysis products. Signals irrelevant to ENSO (see the text) are filtered out.

      The dissimilarity in HC300A of one ocean reanalysis to the others is defined as the average of the sum of the absolute difference in the ranked HC300A of the reanalysis to the others as shown in Eq. (5). For instance, the dissimilarity in HC300A between GODAS and the other three reanalysis products at a certain time step is computed according to the following methods: First, compute the sum of the absolute difference in the ranked HC300A between GODAS and the others in all grids within the domain. Second, average the three sums to get the dissimilarity of GODAS for the time step. Repeat these processes for all four reanalyses at all time steps to acquire four dissimilarity time series. Equation (5) shows the definition of the dissimilarity of reanalysis Aj to the others at time t:

      where HCA is the ranked HC300A field and x and y are the coordinates in the field. The dissimilarity measuring technique is a simple differencing technique similar to root mean square differencing, except for the use of a ranking system and absolute value.

      There are many ways to measure dissimilarity between two patterns. These methods can be loosely categorized into two groups, based on either simple differencing or relative difference. Pattern correlation is an example of the latter and is not suitable for this purpose. It is noted that there are periods with very weak signals in the study domain. In such periods, the relative difference–based dissimilarity-measuring techniques yield a very high dissimilarity for even a few small differences between two patterns. Such subtle differences are likely to be ignored in qualitative studies. To avoid this undesirable property, a simple differencing technique is employed. However, this technique is also sensitive to differences in magnitude rather than simply the distribution of signals. The ranking system reduces this sensitivity in the measurement so that the results align with our objectives.

      The three-month average of Niño-3.4 SSTA from December, January, and February (DJF) in subsequent years is used to classify warm and cold years at the end of the developing year. Periods of time in the developing year are marked by (0), while (1) signifies periods of time in the subsequent year. The pointwise spread of HC300A is defined as the range of HC300A among the ocean reanalysis data sets (that is, the maximum minus the minimum). Growth and decay rate are measured by their rate of change, which is defined as the slope of six-month running least square linear regression. The six-month running window is selected because it approximates the typical time required for an ENSO event to grow and decay. Changing the length of the running window to three months does not affect the result significantly.

    3.   Temporal variation of the dissimilarity in equatorial HC300A and its response to ENSO
    • Figure 4a shows the time series of the dissimilarity of each ocean reanalysis to the others. It is clear that they closely follow the absolute value of Niño-3.4 SSTA. Note that a nine-month moving average is applied to smooth the original spiky dissimilarity time series. After that, correlation coefficients between the absolute value of Niño-3.4 SSTA and the dissimilarity of each ocean reanalysis range from 0.60 to 0.67. The correlation coefficients are maximized when the absolute Niño-3.4 SSTA leads by one month (the range of correlation coefficients becomes 0.63 to 0.70). Hence, the dissimilarity among ocean reanalyses in HC300A generally lags slightly behind ENSO variability. The delay in the peak of dissimilarity varies from event to event. For instance, dissimilarity peaks four months after the mature phase of the 1997−98 El Niño. The large (slightly lagged) correlation between the two implies that the dissimilarity among ocean reanalyses of HC300A increases with the magnitude of ENSO events (both warm and cold phases). This result is consistent with the findings of Xue et al. (2012) that during the strong El Niño events of 1982−83 and 1997−98, the reanalyses diverge wider on HC300A than usual. The dissimilarity of ORAs4 to the rest of the reanalyses is significantly smaller than the average of all reanalyses (greater than 99% confidence level). As illustrated in Fig. 4a, ORAs4 has the lowest dissimilarity most of the time. Hence, HC300A from ORAs4 would be the closest to the ensemble mean of the selected reanalyses.

      Figure 4.  (a) Time series of dissimilarity in HC300A in the equatorial Pacific derived from each ocean reanalysis to the others (colored solid lines) and the absolute value of Niño-3.4 SSTA (dashed black line). A nine-month moving average is applied to the all-time series. It is clear that the dissimilarity time series, regardless of which product, closely follows the absolute Niño-3.4 SSTA most of the time. Exceptions can be found after strong ENSO events, in which the dissimilarity persists. (b) Average dissimilarity in HC300A among reanalyses in the inner-equatorial region (solid red line), off-equatorial regions (solid blue line), and absolute Niño-3.4 SSTA (dashed black line). Note that the two dissimilarity time series are on different scales.

      However, the coherence between the dissimilarity and the magnitude of ENSO (absolute value of Niño-3.4 SSTA) is broken after the maturity of ENSO events, especially after strong events. After the 1982−83 El Niño, for example, the dissimilarity among ocean reanalyses decays slowly from its maximum. As a result, the dissimilarity among ocean reanalyses remains high throughout the decay phase of the El Niño episode and lasts until late 1984. Persisting dissimilarity, although weaker, can also be observed in the decay phase of the 1987−88, 1993−94, and 1997−98 El Niño events, as well as the 1988−89 and 1999−2001 La Niña events.

      Higher dissimilarity after an ENSO event is a result of its slow decay rate (relative to its growth rate) and a delay in the decay of dissimilarity. The rate of change in the dissimilarity among ocean reanalyses shows a moderate positive skew of 0.507 (Fig. 5). Hence, the dissimilarity among ocean reanalyses in HC300A tends to grow faster than it can decay. On the other hand, the absolute value of Niño-3.4 SSTA shows much weaker skewness (–0.086), suggesting symmetry in the growth and decay rates. Also, after a strong ENSO event, dissimilarity starts to fall from a higher level and thus takes a longer time to decay. Dissimilarity among ocean reanalyses seems to be much larger after a strong ENSO event than a moderate one. As mentioned above, the decay of dissimilarity generally lags behind the peak of an ENSO event by months. Hence, dissimilarity starts to fall slowly one to four months after the peak of an ENSO event. This gives the impression that the dissimilarity in HC300A among ocean reanalyses persists throughout the decay phase of an ENSO event.

      Figure 5.  Rate of change of average dissimilarity among reanalyses (blue bars) and absolute value of Niño-3.4 SSTA (red bars). The average dissimilarity shows moderate positive skew (0.507), while the absolute value of Niño-3.4 SSTA shows negligible skew (–0.086). Dissimilarity tends to grow faster than it decays. ENSO state tends to grow and decay at a similar rate.

      Both inner-equatorial (±5°N) and off-equatorial regions (5°–10°N and 5°–10°S) contribute to dissimilarity across the whole equatorial Pacific. Figure 4b shows the variation in the average dissimilarity of HC300A among reanalyses in the two regions. The dissimilarity in the inner-equatorial region follows the ENSO state closely. This dissimilarity lags behind that of the ENSO state by a month, and it decays quickly after the event (except after the 1982−83 El Niño). In contrast, dissimilarity in the off-equatorial regions shows extra persistence after almost all major ENSO events in the study period (1986−87, 1987−88, 1988−89, 1991−92, 1997−98, and 1998−99) and lags behind the magnitude of ENSO by three months. The differences in variation suggest that the sources of dissimilarity in the two regions are different.

      The variations in the dissimilarity of the two regions are also found to closely follow the signal strength in the respective regions (Fig. 6). Here, the signal strength in a region is measured by the sum of the absolute rank number of all grid boxes within the region. With that, the correlation coefficient between the dissimilarity in off-equatorial regions (inner-equatorial regions) and the signal strength in the regions is 0.930 (0.848). Also, no lead-lag relationship is found between signal strength and dissimilarity in either region. Hence, the total dissimilarity is controlled mainly by signal strength across the whole equatorial Pacific. In other words, dissimilarity tends to be greater when signal strength is higher.

      Figure 6.  Dissimilarity (solid line) and signal strength (dashed line) in the equatorial region (magenta) and off-equatorial region (cyan).

    • The persistence of the dissimilarity after an ENSO event is related to the persistence of the signal strength in the basin. Here, the nature by which the equatorial Pacific signal strength varies is described in order to explain the persistence of the dissimilarity after an ENSO event. Note that the ocean heat content signal in the basin is related to the exchange of ocean heat content between the two equatorial bands. This will also be discussed in detail in section 5.

      The signal strengths in both the inner and off-equatorial regions peak around one to three months after the peak of the ENSO state. As the peak of the signal strength is delayed, its decay is also delayed. Thus, decay in the signal strength lags the decay phase of the ENSO event.

      Also, the signal strength tends to decay at a slower rate than it grows (this is indicated by the skewness of the rate of change of the signal strength, which is 0.387 and 0.674 in the equatorial region and off-equatorial region, respectively). These signals can persist for more than 18 months (observed subjectively); hence the same signal strength is maintained in the whole equatorial Pacific region. Furthermore, the signal persistence is observed to be greater after strong ENSO events. On the other hand, the initial signal that leads to an ENSO event is primarily generated less than a year before. Hence, signal strength can grow within a year but take more time to decay. As a result, the rate of change of the signal strength is positively skewed.

      The combined effect of delayed signal decay and slow signal decay can cause the signal strength in the equatorial Pacific basin to persist after the peak of an ENSO event. Considering that there is a high, positive instantaneous correlation between the signal strength and the dissimilarity among ocean reanalyses, the dissimilarity also persists after an ENSO event.

      In addition, extra persistence of the dissimilarity after strong ENSO events may also be caused by an interruption of signal decay due to a follow-up ENSO event. The dissimilarity apparently decays at a slower rate after the 1982−83 and 1991−92 El Niño events and in the strong ENSO activity period from 1997 to 2000 (Fig. 4a). A closer look into the evolution of the dissimilarity in the inner-equatorial region (Fig. 4b) reveals that the decay of the dissimilarity is interrupted (characterized by a short period of growth between the two decay periods). These interruptions coincide with a follow-up event after the original strong ENSO event. For example, the interruption at the end of 1998 is associated with the 1998−99 La Niña, which is related to the 1997−98 El Niño before it. Similar situations can also be observed after other strong ENSO events. The magnitude of the follow-up event (if one occurs), and the signal strength, tend to be stronger after a strong ENSO event since it leads to a stronger anti-phase equatorial signal (e.g., 1987−88 El Niño and the follow-up 1988−89 La Niña). Because of this, the interruption in the decay of the dissimilarity will be greater after a strong ENSO event, and it further slows the dissimilarity decay rate.

    4.   Horizontal distribution of dissimilarity and its relationship with different ENSO states
    • As mentioned in Xue et al. (2012), dissimilarity in HC300A among ocean reanalyses is not evenly distributed across the equatorial Pacific Ocean. Figure 4b has already shown that dissimilarity in the off-equatorial regions is higher than that in inner-equatorial regions most of the time. From the average dissimilarity across the equatorial Pacific Ocean (Fig. 7), it can be observed that the four ocean reanalyses diverge the most in the off-equatorial central and eastern Pacific. Lack of observations in the off-equatorial eastern Pacific in the past [see Figs. 5d, 5e in Xue et al. (2012)] may have caused the large dissimilarity in the region. In contrast, the dissimilarity in the western Pacific is noticeably lower than in the rest of the equatorial Pacific, which could be the result of the abundance of in-situ observations during the study period (Hu et al., 2020).

      Figure 7.  Average dissimilarity among ocean reanalyses in HC300A in the equatorial Pacific throughout the study period. Average dissimilarity is lower in the inner-equatorial region and off-equatorial western Pacific. However, the dissimilarity in the off-equatorial central and eastern Pacific is higher.

      There is a clear spatial structure of the spread in HC300A among ocean reanalyses during warm ENSO events. Figure 8 illustrates how the spread structure evolves in different stages of warm events. The selected stages include August, September, October (ASO), November, December, January (NDJ), February, March, April (FMA), and May, June, July (MJJ). Areas in which the spread of HC300A has a significantly positive (negative) lagged correlation (significant level at 10% tested with Student-t test) with D(0)JF(1) Niño-3.4 SSTA are marked by black dots (black meshes) in Fig. 8. In the calculation of correlation, only the 14 years with positive D(0)JF(1) Niño-3.4 SSTA are used. D(0)JF(1) Niño-3.4 SSTA is used to approximate the peak magnitude of the ENSO state each year. Lagged correlation between HC300A and D(0)JF(1) Niño-3.4 SSTA is also shown on the plot to represent the distribution of signals. It can be observed that the spatial pattern of the spread “propagates” along with the typical ENSO signals. This is clearest in MJJ(1), in which part of the spread may be associated with the spearhead of the advancing equatorial negative signal in the east-central Pacific. Also, in ND(0)J(1), the spread may be associated with meridional separation of the equatorial positive signal in the eastern Pacific.

      Figure 8.  Lagged-correlation maps of HC300A (shading) and standardized HC300A spread anomaly (dotted and meshed areas) to the absolute value of D(0)JF(1) Niño-3.4 SSTA. Here, only years with positive D(0)JF(1) Niño-3.4 SSTA (14 years in total) are used in the calculation of correlation. Red shading and dots signify areas with positive correlation. Blue shading and mesh signify areas with negative correlation. Areas with correlations less than the 90% confidence level are removed.

      However, the spatial structure of the spread during cold ENSO years is not as clear as that which is observed in warm years. Applying the same correlation analysis for the 14 years with D(0)JF(1) Niño-3.4 SSTA shows no propagating area of significant spread (Fig. 9). Note that the absolute value of D(0)JF(1) Niño-3.4 SSTA is used in the calculation of correlation. There is no area with a significant spread in HC300A in ASO(0) and MJJ(1). In ND(0)J(1) and FMA(1), areas with a significant negative correlation between the spread and absolute Niño-3.4 SSTA are mostly confined to the east-central Pacific. In the inner-equatorial region, the correlation between the spread and D(0)JF(1) is negative. This means the spread is reduced in these areas when the cold event is stronger. Compared to that of warm ENSO years, the spread pattern of cold ENSO years does not evolve clearly and is mostly confined to ND(0)J(1). Hence, the relationship between the spread pattern in HC300A during cold ENSO events is far less clear than that of warm events. However, this does not mean that there is no relationship between the dissimilarity in HC300A among ocean reanalyses and cold ENSO events, nor that the dissimilarity decreases with the magnitude of a cold event.

      Figure 9.  Same as Fig. 7 except only the years with negative D(0)JF(1) Niño-3.4 SSTA (14 years in total) are used in the calculation of correlation. The absolute value of Niño-3.4 SSTA is used in the calculation.

    5.   Possible mechanism of dissimilarity variability in a warm ENSO phase
    • As discovered in the previous sections, the dissimilarities in both equatorial and off-equatorial regions of the Pacific Ocean follow closely with the signal strength in the respective region (Fig. 6). Here, we discuss the mechanism of the variation of signal strength in order to explain the variation of the dissimilarity during a warm ENSO event.

      In the recharge-discharge oscillator theory (Jin, 1997), a prominent thermocline tilting is observed along the equatorial Pacific during the mature phase of a warm ENSO event. The thermocline is anomalously deep (shallow) in the eastern (western) Pacific. As a result, a strong signal in (absolute) HC300A can be found in both the equatorial eastern and western Pacific during the mature phase of a warm ENSO event.

      Poleward Sverdrup transport along the equator is also maximized during the mature phase of a warm ENSO event. This poleward transport discharges ocean heat content away from the inner equatorial Pacific region. Such discharges have two effects: In the equatorial region, the basin mean ocean heat content becomes anomalously low. The off-equatorial region receives ocean heat content from the equatorial region. Both effects cause an increase in signal strength in the respective region.

      As the ENSO event decays, the thermocline tilting also decays, which should result in a sharp reduction in signal strength in both regions. However, the increase in signal strength due to anomalously low ocean heat content in the equatorial region and the flux of ocean heat content leaving in the off-equatorial region partly cancels the reduction rate of signal strength. Consequently, the signal strength decays slowly as the ENSO event rapidly decays.

      After a weak warm ENSO event, the residual signal in the equatorial region may be too weak to trigger a follow-up event (a follow-up cold ENSO event). On the other hand, after a strong warm event, the residual, anomalously low ocean heat content in the equatorial region may lead to surface cooling in the eastern Pacific. The cooling then grows through Bjerknes feedback. The signal in the equatorial region can also be amplified due to the ensuing thermocline tilting. As a result, the signal strength after a strong warm ENSO event will decay even more slowly than a weak warm ENSO event.

    • As ocean heat content is essentially a vertical integration of ocean water temperature, spread in ocean temperature is the main factor of spread in ocean heat content among ocean reanalyses. Figures 10 and 11 show the lagged correlation of ND(0)J(1) Niño-3.4 SSTA to the standardized ocean temperature anomaly (OTA), which is used to trace the propagation of the ENSO signal, and spread among the reanalyses on OTA at different depths in the Pacific Ocean. Warm ENSO signals in the inner-equatorial regions (shading in the figures) can already be seen half a year before the maturation of a warm event. Cold signals can be seen three months later to the west. These two signals evolve and propagate to the east. The spread of OTA among ocean reanalyses largely collocates with these ENSO signals. This may suggest that the spread in HC300A during warm ENSO phases is likely to be caused by the differences in the reconstruction of subsurface dynamics among reanalyses.

      Figure 10.  Lagged correlation of standardized ocean temperature anomaly (OTA) (shading) and spread among the reanalyses on OTA to ND(0)J(1) Niño-3.4 SSTA (dots and meshes). Red and blue shading denote the areas where the lagged correlation of OTA is positive and negative, respectively. Dots and meshes indicate the areas where the lagged correlation of the spread is positive and negative, respectively. Only the years that end with a warm event are considered. Figures on the left represent the evolution in the inner-equatorial region. Figures on the right represent the evolution in the off-equatorial region. Only the statistically significant signal are drawn (test level of 10%).

      Figure 11.  Continuum of Fig. 10. The years after a warm event are considered in the calculation of correlation.

      The spread among the data sets is not as well-structured in the off-equatorial region as in the inner-equatorial region. A possible explanation for this centers around the fact that the spread in the off-equatorial region is naturally larger than that in the equatorial region, as suggested in Fig. 7. As a result, the spread in response to warm ENSO events (i.e., the anomaly of the spread among the data sets) stands out less from the noisy background spread. Nevertheless, the spread among the data is also mostly collocated with the ENSO signals. This suggests that the differences in the reproduced surface processes are the cause of dissimilarity in the off-equatorial region.

      The unclear spread pattern of HC300A among ocean reanalyses during cold ENSO years does not lead us to the relationship between subsurface temperature variation and variability of dissimilarity for these periods. In fact, the structure of spread in OTA among ocean reanalyses is unclear during cold ENSO phases (not shown). Further study is required to understand the mechanism behind the variability of the dissimilarity during cold ENSO events.

    6.   Discussion and summary
    • To study how the dissimilarity among ocean reanalyses in HC300A in the equatorial Pacific Ocean (±10°N) changes with time, HC300A is derived from the ocean reanalyses of GFDL, GODAS, ORAs4, and SODA. They are compared against one another using a simple differencing method.

      It is found that the dissimilarity among reanalyses in HC300A in the equatorial Pacific Ocean closely follows the ENSO state, which is represented by the absolute value of Niño-3.4 SSTA. Ocean reanalyses tend to split more when ENSO is strong. Also, the rate of change of the dissimilarity shows a moderate, positive skewness. The dissimilarity tends to decay more slowly than it grows, so it persists after ENSO events. It remains at a high magnitude after the mature phase of ENSO events, especially after strong events.

      The dissimilarity appears to be controlled mainly by the signal strength of the HC in the region of concern. The persistence of the dissimilarity after ENSO events is then caused by the delay in signal strength and the persistence of the signal. The persistence of the signal strength in the decay phase of ENSO may in turn be caused by the ocean heat content discharge.

      Thermocline tilting and basin mean thermocline depth contributes to the signal strength of HC300A in the equatorial Pacific. As a warm ENSO matures, the thermocline tilting maximizes and consequently results in high signal strength (and therefore, high dissimilarity). During the subsequent ENSO decay, ocean heat content discharge from the equatorial region to the off-equatorial region leads to anomalously shallow (deep) mean thermocline depth in the equatorial (off-equatorial) region. This keeps the signal strength high in both regions. Although the thermocline tilting decays with ENSO, the signal strength, due to ocean heat content discharge, reduces the decay rate of the signal strength. This decay rate will be even slower after a strong ENSO event due to the genesis of a follow-up ENSO event. Since the dissimilarity among the reanalyses is directly related to the signal strength in a region, the decay rate of the dissimilarity is also slowed, and it persists longer after ENSO events, especially after strong ENSO events.

      The spread of patterns in vertical ocean temperature in both the east and west equatorial regions show a clear relationship with propagating ENSO signals. The areas with significant spread collocate mostly with ENSO signals. Therefore, it is proposed that the source of the dissimilarity is related to the differences in the reproduced subsurface dynamics among reanalyses.

      The structure of the spread pattern in HC300A during cold ENSO events is not as clear. As a result, it is more difficult to directly relate the spread pattern with subsurface ENSO signals. However, as dissimilarity in HC300A is largely related to signal strength without a lead-lag relationship, and signal strength variation is caused by passing subsurface ENSO signals, the dissimilarity in HC300A during cold ENSO events should also be related to the variation in the reproduced subsurface dynamics, though further study is required to confirm this.

      Given that ocean reanalyses diverge to a greater extent after an ENSO event, especially after strong events, studies of ENSO decay and transition behaviors with ocean reanalyses are less reliable than studies of the developing phase. Sensitivity tests concerning the choice of ocean reanalysis products are more important in these kinds of studies. Such precaution is more important for analyses done over the off-equatorial eastern Pacific.

      Acknowledgements. This work is fully supported by the International Cooperation and Exchange Programme of the National Natural Science Foundation (Grant No. 42120104001).

Reference

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return