There is compelling evidence that episodic deposition of large volumes of freshwater into the oceans strongly influenced global ocean circulation and climate variability during glacial periods1,2. In the North Atlantic region, episodes of massive freshwater discharge to the North Atlantic Ocean were related to distinct cold periods known as Heinrich Stadials1,2,3. By contrast, the freshwater history of the North Pacific region remains unclear, giving rise to persistent debates about the existence and possible magnitude of climate links between the North Pacific and North Atlantic oceans during Heinrich Stadials4,5. Here we find that there was a strong connection between changes in North Atlantic circulation during Heinrich Stadials and injections of freshwater from the North American Cordilleran Ice Sheet to the northeastern North Pacific. Our record of diatom δ18O (a measure of the ratio of the stable oxygen isotopes 18O and 16O) over the past 50,000 years shows a decrease in surface seawater δ18O of two to three per thousand, corresponding to a decline in salinity of roughly two to four practical salinity units. This coincided with enhanced deposition of ice-rafted debris and a slight cooling of the sea surface in the northeastern North Pacific during Heinrich Stadials 1 and 4, but not during Heinrich Stadial 3. Furthermore, results from our isotope-enabled model6 suggest that warming of the eastern Equatorial Pacific during Heinrich Stadials was crucial for transmitting the North Atlantic signal to the northeastern North Pacific, where the associated subsurface warming resulted in a discernible freshwater discharge from the Cordilleran Ice Sheet during Heinrich Stadials 1 and 4. However, enhanced background cooling across the northern high latitudes during Heinrich Stadial 3—the coldest period in the past 50,000 years7—prevented subsurface warming of the northeastern North Pacific and thus increased freshwater discharge from the Cordilleran Ice Sheet. In combination, our results show that nonlinear ocean–atmosphere background interactions played a complex role in the dynamics linking the freshwater discharge responses of the North Atlantic and North Pacific during glacial periods.
Access optionsAccess options
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was largely part of the Innovative NOrth Pacific EXperiment (INOPEX), funded by the Bundesministerium für Bildung und Forschung. We also acknowledge funding by the Helmholtz Postdoc program (PD-301; to X.Z.), as well as Helmholtz funding through the Polar Regions and Coasts in the Changing Earth System (PACES) program of the Alfred Wegener Institute. Funding from the Qingdao National Laboratory for Marine Science and Technology (QNLM201703) is also acknowledged. We thank U. Böttjer, B. Glückselig and R. Cordelair for the thorough purification of diatom samples for isotope analyses; M. Warnkross for picking planktic foraminifera for stable-isotope analysis and radiocarbon dating; S. Steph and A. Mackensen for performing the foraminiferal oxygen-isotope analysis; and G. Knorr for helpful discussions.Reviewer information
Nature thanks S. Dee, A. Hu, K. Thirumalai and the other anonymous reviewer(s) for their contribution to the peer review of this work.
Extended data figures and tables
Left, LGM_NA; right LGM_NA+NP. Model results are presented as anomalies between the hosing simulations and the LGM state (see Methods). a, e, Total precipitation anomalies. b, f, Precipitation δ18O anomalies. c, g, δ18Odiat. anomalies. d, h, Subsurface δ18Osw anomalies (at depths of 120–180 m). The yellow star marks the location of core SO202-27-6.
a–d, SO202-27-6. a, Scheme and pictures describing the sediment core74. b, Linear sedimentation rate (LSR). c, Calcium intensity based on XRF analysis. d, Iron intensity based on XRF analysis. ka, thousands of years ago; kcps, thousands of counts per second. e, NGRIP δ18O record7. f, NGRIP dust concentration31, including age-control points for SO202-27-6. g, Mean summer insolation at 65° N (ref. 75).
a, Study area as shown in Fig. 1, including the modern extent of the CIS (black), the extent of the CIS during the LGM76,77 (black dashed line), the site of the studied core (yellow star), and the locations where precipitation (red dots) and modern glacier (light blue dots) parameters were taken for the SSS/δ18Osw mixing model. b, Study area from the North Pacific. Shaded white areas represent the extents of the LIS and CIS during the LGM78. c, SSS/δ18Osw mixing model assuming linear regression between SSS and δ18Osw. We used three low-salinity endmembers and one high-salinity endmember to estimate SSS changes at our core site between 43 kyr and 11 kyr ago (see Methods). The Northern Hemisphere map and the SSS map were created using Ocean Data View79.
a, b, Results obtained using COSMOS. a, LGM_NA experiment. b, 30kyr_NA experiment. c–f, Results obtained using ECHAM5. c, A_HS1 experiment. d–f, A_HS1 experiment, imposing SST fields on the Atlantic Basin (Atl) only (d), the Atlantic Basin and the east Equatorial Pacific (EEP; e), and the EEP only (f). The yellow star marks the location of studied Core SO202-27-6. Surface-wind anomalies (vectors) are presented in m s−1. Sea-level-pressure anomalies are shown with shading.
Extended Data Fig. 5 Results from freshwater hosing experiments LGM_NP, LGM_NA and LGM_NA02, presented as anomalies.
a–c, LGM_NP experiment (0.1 Sv). a, Global SST anomaly; b, North Pacific subsurface temperature anomaly (120–180 m). c, Temperature anomaly over the meridional transaction of the Atlantic basin (60° W to 15° W). d–g, LGM_NA experiment (0.15 Sv) (d, e) and LGM_NA02 experiment (0.2 Sv) (f, g). d, f, AMOC field anomalies. e, g, Subsurface temperature anomalies (120–180 m). The yellow star marks the location of core SO202-27-6.
Extended Data Fig. 6 Comparison of northwestern North Pacific eolian dust and iron intensity, as well as NGRIP dust concentration over the last deglaciation.
a, Eolian dust (terrestrial 4He concentration)32 and b, iron intensity from core SO202-07-6 (51.3° N, 167.7° E; 2,340 m water depth). c, NGRIP dust concentration31. Dust changes in the northwestern North Pacific and Greenland are synchronous32, and coincide with iron-intensity changes in the northwestern North Pacific. B/A, Bølling/Allerød interstadial; YD, Younger Dryas cold period. Red arrows mark chronological coincidence between the changes in 4He, iron intensity and NGRIP dust concentration; ncc STP g−1, nano-cubic centimetre per gram at standard temperature and pressure.
Extended Data Fig. 8 Diatom isotope sample composition, residual contamination with non-biogenic silicates and mass-balance-corrected δ18Odiat. (from core SO202-27-6).
a, Relative abundances of the following diatom species in the isotope samples: C. marginatus, C. oculus-iridis, and other diatom species. b, Contamination of purified diatom samples with non-biogenic silicates, estimated by inductively coupled plasma optical emission spectrometry (ICP-OES) and energy-dispersive X-ray spectrometry (EDS). c, Blue line, measured δ18Odiat. values (error bars indicate errors of replicate analyses or long-term reproducibility of standards (1σ)). Black dotted lines, δ18Odiat. values that have been mass-balance-corrected for contamination with non-biogenic silicates (estimated by EDS), and using one of two different δ18O values for non-biogenic silicate contamination (+2‰ or +30‰). Contamination values, δ18Odiat. values and mass-balance-corrected δ18Odiat. values younger than 25 kyr bp are taken from ref. 12.