Research ArticleOCEANOGRAPHY

Deep-reaching acceleration of global mean ocean circulation over the past two decades

See allHide authors and affiliations

Science Advances  05 Feb 2020:
Vol. 6, no. 6, eaax7727
DOI: 10.1126/sciadv.aax7727
  • Fig. 1 TKE anomalies and linear trends in various datasets.

    (A to F) Thick (thin) color lines denote monthly TKE′ with (without) 61-month running mean in Estimating of Circulation and Climate of the Ocean, phase 2 (ECCO2), European Centre for Medium-Range Weather Forecasts (ECMWF) ORA-S3, Global Ocean Data Assimilation System (GODAS), AGVA, Geophysical Fluid Dynamics Laboratory Ensemble Coupled Data Assimilation (GFDL ECDA), and ECMWF ORA-S4 datasets, respectively. Black lines are the linear trends of the 61-month low pass–filtered series. (G) The ensemble-mean series of the TKE′ from the six reanalysis datasets. (H) Linear trends of global ocean mean KE anomaly before 1990 [Period A: ECMWF ORA-S3 and ORA-S4 1959–1990, GFDL ECDA 1960–1990, and National Oceanic and Atmospheric Administration (NOAA) GODAS 1980–1990] and after 1990 (Period B: ECMWF ORA-S3 1991–2011, ECCO2 1991–2013, GFDL ECDA 1991–2012, NOAA GODAS 1991–2011, AGVA 2005–2010, and ECMWF ORA-S4 1991–2013). SEs are shown in blue bar, and trends that are statistically significant at the 99% confidence level are denoted in blue circles, while nonsignificant trends are signposted with red “×”. SE of TKE trend in individual product is defined as the SD of the detrended time series, while that of ensemble mean is defined as the SD of trends from six reanalysis.

  • Fig. 2 Horizontal and vertical distribution of long-term KE change.

    (A) Linear trend of oceanic KE averaged over the upper 2000-m layer (shaded color, unit in 103 J m−2 decade−1) during 1991–2011 from the ensemble mean of the ECMWF ORA-S4, ORA-S3, ECCO2, GODAS, and GFDL ECDA. Area where statistical significance is above the 99% confidence level is highlighted by black dots. (B) Vertical distribution of linear trend (blue line) of ensemble and global mean KE′ and the linear trend in percentage (embedded red line, relative to the climatological KE at each depth). Black dots indicate the 99% confidence level, and shaded area denotes the error in the trend. (C) Time-depth plot of ensemble mean KE′ (color shading and black contour lines) integrated over the global ocean. KE′ is calculated by subtracting the time means over 1991–2011 from the monthly series and low-pass filtering with a cutoff period of 25 months. (D) As in (C), but for the AGVA during 2005–2010.

  • Fig. 3 Global mean sea surface wind speed and wind work.

    (A) Global mean wind speed (m s−1) at 10 m from various wind products. Each product is presented with a 13-month smoothed monthly time series superimposed with a 25-month smoothed time series. The thick blue line is the ensemble mean of the six wind products. The thick black line denotes the linear trend of the ensemble mean wind speed over the overlapping period of the six wind speed products (1985–2010) at the 99% confidence level. (B) Distribution of linear trend in wind work (10−3 W m−2 decade−1) during 1991–2011 from the ECMWF ORA-S4 assimilation. Area with confidence level higher than the 99% level is highlighted by black dots. (C) Comparison between globally integrated wind work (red) and ORA-S4 TKE′ (blue). All the time series are low pass–filtered (13-month running mean). Wind work is calculated using wind stress and 5-m current products from ECMWF ORA-S4 (see Materials and Methods for details).

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/6/6/eaax7727/DC1

    Supplementary Text

    Table S1. Features of various datasets and statistics of TKE′ from various data products.

    Fig. S1. Monthly TKE′ and corresponding linear trend in 12 CMIP5 model outputs.

    Fig. S2. KE′ in the tropical oceans and for individual basin-wide meridional sections.

    Fig. S3. KE′ in global and each ocean basin.

    Fig. S4. Horizontal distribution of long-term KE change in individual products.

    Fig. S5. Time-depth plot of global mean KE′ in individual products.

    Fig. S6. Time series of TKE′ with a focus on the upper 700-m layer.

    Fig. S7. Comparison between global mean wind, TKE′, and C(APE, KE).

    Fig. S8. Relationship between wind speed, PDO index, and TKE.

    Fig. S9. Global mean wind speed in CMIP5 models and comparison with reanalysis.

  • Supplementary Materials

    This PDF file includes:

    • Supplementary Text
    • Table S1. Features of various datasets and statistics of TKE′ from various data products.
    • Fig. S1. Monthly TKE′ and corresponding linear trend in 12 CMIP5 model outputs.
    • Fig. S2. KE′ in the tropical oceans and for individual basin-wide meridional sections.
    • Fig. S3. KE′ in global and each ocean basin.
    • Fig. S4. Horizontal distribution of long-term KE change in individual products.
    • Fig. S5. Time-depth plot of global mean KE′ in individual products.
    • Fig. S6. Time series of TKE′ with a focus on the upper 700-m layer.
    • Fig. S7. Comparison between global mean wind, TKE′, and C(APE, KE).
    • Fig. S8. Relationship between wind speed, PDO index, and TKE.
    • Fig. S9. Global mean wind speed in CMIP5 models and comparison with reanalysis.

    Download PDF

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article