Research ArticleENVIRONMENTAL SCIENCE

Improved estimates of ocean heat content from 1960 to 2015

+ See all authors and affiliations

Science Advances  10 Mar 2017:
Vol. 3, no. 3, e1601545
DOI: 10.1126/sciadv.1601545
  • Fig. 1 Fractional coverage of the mapping method used in this study.

    (A) Averaged fraction (0 to 700 m) of the global ocean considered sampled for temperature data in each month for this study. (B) Same as (A) but for 700 to 2000 m. The mean fractional coverage of observations is shown in blue if the global ocean is divided into 1°-by-1° grids, and fractional coverage for NCEI yearly and the 5-year method is shown in green and orange, respectively. This figure starts from 1940. (C) Fractional coverage of the global ocean for layers within 0 to 700 m. (D) Same as (C) but for 700 to 2000 m.

  • Fig. 2 Observational correlations with distance.

    (A and C) Zonal-mean (Z) and (B and D) meridional-mean (M) correlation as a function of distance calculated using the ORAS4 reanalysis data. A linear trend and the mean seasonal cycle have been removed when calculating the correlation. The ORAS4 1° by 1° and monthly data were used.

  • Fig. 3 Global and basin-averaged sampling error compared with reconstructed temperature change.

    (A to G) The sampling errors for different truth fields subsampled by historical observation locations are shown as green dots, with the green solid line and the error bar for the mean and ±2σ, respectively. Blue stars denote the 16 different truth fields. The gray line is the reconstructed monthly temperature anomaly time series from 1960 to 2015 with ±2σ interval in gray shading, and the dark line is the time series after a 7-year low-pass filter (see Materials and Methods). The orange curve is the NCEI result, along with a 1-SE bar (dashed orange). (A) Global, (B) tropical/subtropical Pacific, (C) North Pacific, (D) Indian, (E) tropical/subtropical Atlantic, (F) North Atlantic, and (G) southern oceans. (H and I) S/N ratio of the temporal variability of our reconstruction on (H) decadal scales (solid lines) and (I) interannual scales (triangle lines) compared to the sampling error.

  • Fig. 4 Global OHC change time series.

    (A) OHC from 0 to 700 m (blue), 700 to 2000 m (red), and 0 to 2000 m (dark gray) from 1955 to 2015 as obtained by this study, with the uncertainty of the ±2σ interval shown in shading. All time series of the new analysis are smoothed by a 12-month running mean filter, relative to the 1997–2005 base period. (B) The new estimate is compared with an independent estimate from NCEI with its SE as dashed lines. Both OHC 0 to 700 m and OHC 700 to 2000 m are shown from 1957 to 2014. The baseline of the time series from NCEI is adjusted to the values of the current analysis within 2005–2014.

  • Fig. 5 OHC changes from 1960 to 2015 for different ocean basins.

    (A) For 0 to 2000 m, (B) 0 to 700 m, and (C) 700 to 2000 m. All the time series are relative to the 1997–1999 base period and smoothed by a 12-month running filter. The curves are additive, and the OHC changes in different ocean basins are shaded in different colors.

  • Fig. 6 Estimate of the ocean energy budget.

    The three major volcanic eruptions are marked. The energy budgets are relative to the 1958–1962 base period. The integrated net radiative imbalance from Allan et al. (65) estimated from the TOA is included in yellow and is multiplied by 0.93 to be comparable with the ocean energy budget. The TOA radiation is adjusted to the value of OHC within 2013–2014. The dashed gray lines encompass the 95% confidence interval.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/3/3/e1601545/DC1

    Supplementary Materials and Methods

    fig. S1. Illustration of the iterative EnOI-DE/CMIP5 method used in the current study.

    fig. S2. An example of the reconstructed fields after three iterative scans.

    fig. S3. Reconstruction of temperature field at 1200 m in August 2011 for the historical sampling (in February).

    fig. S4. Reconstruction of temperature field at 1200 m in August 2011 for the historical sampling (in September).

    fig. S5. Mean temperature error as a function of different choices of the influencing radii between the reconstructed and truth fields.

    fig. S6. Six major ocean basins defined in this study.

    fig. S7. Global and basin-averaged sampling error compared with reconstructed temperature change at 20 m.

    fig. S8. Global and basin-averaged sampling error compared with reconstructed temperature change at 300 m.

    fig. S9. Global and basin-averaged sampling error compared with reconstructed temperature change at 800 m.

    fig. S10. Global and basin-averaged sampling error compared with reconstructed temperature change at 1200 m.

    fig. S11. Geographical distribution of mean and 2σ sampling error for 0- to 2000-m average.

    fig. S12. Geographical distribution of mean and 2σ sampling error in 1°-by-1° grid at 20 and 1600 m.

    fig. S13. 2σ sampling error for different scans at different depths.

    fig. S14. Global OHC time series for the reconstructions after scan 1, scan 2, and scan 3.

    fig. S15. Frequency distribution of temperature anomalies for the years 1986 and 2015.

    fig. S16. Sampling error as calculated by two subsample methods.

    fig. S17. S/N ratio analysis for two methods of subsample test.

    fig. S18. Comparison between OHC and sea level change since 1993.

    fig. S19. Distribution of the ensemble anomalies.

    table S1. OHC trends obtained in this study for the 1960–1991 and 1992–2015 periods.

    table S2. Net OHC and EEI changes obtained in the current study compared with some independent estimates.

  • Supplementary Materials

    This PDF file includes:

    • Supplementary Materials and Methods
    • fig. S1. Illustration of the iterative EnOI-DE/CMIP5 method used in the current study.
    • fig. S2. An example of the reconstructed fields after three iterative scans.
    • fig. S3. Reconstruction of temperature field at 1200 m in August 2011 for the historical sampling (in February).
    • fig. S4. Reconstruction of temperature field at 1200 m in August 2011 for the historical sampling (in September).
    • fig. S5. Mean temperature error as a function of different choices of the influencing radii between the reconstructed and truth fields.
    • fig. S6. Six major ocean basins defined in this study.
    • fig. S7. Global and basin-averaged sampling error compared with reconstructed temperature change at 20 m.
    • fig. S8. Global and basin-averaged sampling error compared with reconstructed temperature change at 300 m.
    • fig. S9. Global and basin-averaged sampling error compared with reconstructed temperature change at 800 m.
    • fig. S10. Global and basin-averaged sampling error compared with reconstructed temperature change at 1200 m.
    • fig. S11. Geographical distribution of mean and 2σ sampling error for 0- to 2000-m average.
    • fig. S12. Geographical distribution of mean and 2σ sampling error in 1°-by-1° grid at 20 and 1600 m.
    • fig. S13. 2σ sampling error for different scans at different depths.
    • fig. S14. Global OHC time series for the reconstructions after scan 1, scan 2, and scan 3.
    • fig. S15. Frequency distribution of temperature anomalies for the years 1986 and 2015.
    • fig. S16. Sampling error as calculated by two subsample methods.
    • fig. S17. S/N ratio analysis for two methods of subsample test.
    • fig. S18. Comparison between OHC and sea level change since 1993.
    • fig. S19. Distribution of the ensemble anomalies.
    • table S1. OHC trends obtained in this study for the 1960–1991 and 1992–2015 periods.
    • table S2. Net OHC and EEI changes obtained in the current study compared with some independent estimates.

    Download PDF

    Files in this Data Supplement:

More Like This