Research ArticleGEOLOGY

Pacific warm pool subsurface heat sequestration modulated Walker circulation and ENSO activity during the Holocene

See allHide authors and affiliations

Science Advances  14 Oct 2020:
Vol. 6, no. 42, eabc0402
DOI: 10.1126/sciadv.abc0402
  • Fig. 1 Time series of thermocline and SST anomalies in the IPWP compared to global climate indices during the past 25,000 years.

    (A) Site locations of paired SST and thermocline water temperature (TWT) records (white circles) and SST-only records (blue triangles) (table S1). Shadings indicate temperatures at 120-m water depth. (B) Mean TWTA (red) of the IPWP records. Solid black arrows mark the two major warming phases of TWTA between 22 to 19 ka and 13 to 11 ka, respectively. (C) Precession (dashed purple) and obliquity (orange) parameters (47). (D) Atmospheric pCO2 derived from West Antarctic Ice Sheet Divide ice core [gray dots; (26)]. ppmv, parts per million by volume. (E) Mean SSTA (blue) of the IPWP records and the global mean SST anomaly [ΔT, dark gray line; (48)]. (F) Mean IPWP G. ruber δ18OG anomaly (δ18OG-A, green) and LR04 benthic δ18O stack [gray line and symbols; (49)]. Shadings of proxy records show the 1σ SD. Vertical dashed lines denote the timing of the deglacial onset of SST (~19 ka, blue) and TWT (~22 ka, red), the onset of the second deglacial warming step (gray), and the Early Holocene peak of TWT (EH-peak, ~10.8 ka, red). Dotted red arrow denotes the Middle Holocene peak of TWT (MH-peak, 7 ka). B.P., before the present.

  • Fig. 2 The two types of thermocline temperature anomaly (TWTA) records in the IPWP since the LGM.

    (A) Average TWTA (brown) and the original TWTA records of the open-ocean sites. (B) Same as (A) but for the near-equator sites in the Maritime Continent waters. The TWT records in (A) and (B) are defined as (EH-peak and MH-peak types), respectively. (C) First (blue) and second (red) principal components (PCs) of all the TWTA records. PC1 and PC2 explain 62 and 16% of the total variance, respectively. (D and E) Loadings of PC1 (D) and PC2 (E) for each site. (F) Linear combinations of PC1 and PC2 that resemble the Early Holocene peak type (PC1 − PC2) and Middle Holocene peak type (PC1 + PC2), respectively.

  • Fig. 3 Precession-forced Early Holocene TWTA peak.

    (A) Mean TWTA of the Early Holocene peak type (brown), precession (red dashed line), and obliquity [gray dashed line; (47)]. (B) Meridional SSTA gradient between southwest Pacific [SWP; site MD97-2120, from 45.5°S, 174.9°E; (3132)] and the IPWP. Holocene, last deglaciation, and LGM are separated by dashed vertical lines. CESM-simulated responses of Pacific subsurface temperature to June insolation in the Early Holocene are shown in (C to F): horizontal temperature anomaly distributions of upper-thermocline [at 120 m in (C)] and deeper thermocline [160- to 180-m water depth average in (D)] and meridional upper-water temperature anomaly profiles in the open Pacific [140°E to 140°W in (E)] and the Maritime Continent waters [100°E to 140°E in (F)]. Temperature anomalies in (C) to (F) are shown as regression coefficients against the standardized time series of the June insolation at precessional band in experiment CESM_GHG. White shadings mask insignificant results below 95% confidence level (t test). EQ, equator.

  • Fig. 4 Time series and simulated temperature and rainfall anomalies in the IPWP since the LGM.

    (A) Mean TWTA of Middle Holocene peak type (green) and the September 21st insolation at the equator (dashed red line) and obliquity [gray dotted line; (47)]. (B) Zonal temperature gradients at subsurface (Δsub-TA, in gray) and at sea surface [ΔSSTA, in violet; (42)] shown as the difference between the western equatorial Pacific and eastern equatorial Pacific. (C) δ18O records of northern Borneo stalagmites (15, 16) (dark and light green) and simulated annual mean rainfall (millimeters per day) over Borneo (dark gray, this study). PDB, Pee Dee belemnite. (D) Mean anomaly of seawater δ18O of IPWP (Δδ18Osw, dark gray, shading shows the 1σ error of the records). Shadings, vertical bars, and dashed lines are as in Fig. 3. Simulated response of the Pacific subsurface temperature and atmospheric variables to September insolation maximum are shown in (E to H): (E) Annual mean TWTA at 120-m water depth. (F) Depth profile of the annual mean temperature anomaly across the Pacific between 5°S and 5°N. (G) Late-autumn (October to December) anomalies of mean rainfall (colors, in millimeters per day) and horizontal winds at 850 hPa (arrows, in meters per second, reference arrow on top right). (H) Late-autumn mean Walker circulation anomalies between 5°S and 5°N across the Pacific, as indicated by anomalies in wind (arrows, in meters per second, reference arrow on top right) and in vertical velocity (colors, in pascal per second). Positive values in red indicate upward motion, and negative values in blue indicate downward motion. These anomalies in (E) to (H) are shown as regression coefficients against the standardized time series of the September insolation at precessional band in experiment CESM_GHG. SMOW, standard mean ocean water.

Supplementary Materials

  • Supplementary Materials

    Pacific warm pool subsurface heat sequestration modulated Walker circulation and ENSO activity during the Holocene

    Haowen Dang, Zhimin Jian, Yue Wang, Mahyar Mohtadi, Yair Rosenthal, Liming Ye, Franck Bassinot, Wolfgang Kuhnt

    Download Supplement

    The PDF file includes:

    • Sections S1 and S2
    • Figs. S1 to S4
    • Tables S1 to S3
    • Legend for data file S1
    • References

    Other Supplementary Material for this manuscript includes the following:

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article