New evidence from deep-sea sediment cores in the subpolar North Atlantic demonstrates that a significant component of sub-Milankovitch climate variability occurs in distinct 1-2 kyr cycles. We have traced that cyclicity from the present to within marine isotope stage 5, an interval spanning more than 80 kyrs. The most robust indicators of the cycle are repeated increases in the percentages of two petrologic tracers, Icelandic glass and hematite-stained grains. Both are sensitive measures of ice rafting episodes associated with ocean surface coolings. The petrologic tracers exhibit a consistent relation to Heinrich events, implying that mechanisms forcing Heinrich events were closely linked to those forcing the cyclicity. Our records further suggest that Dansgaard/Oeschger events may be amplifications of the cycle brought about by the impact of iceberg (fresh water) discharges on North Atlantic thermohaline circulation. The tendency of thermohaline circulation to undergo threshold behavior only when fresh water input is relatively large may explain the absence of Dansgaard/Oeschger events in the Holocene and their long pacings (thousands of years) in the early part of the glaciation. Finally, evidence from cores near Newfoundland confirms previous suggestions that the Little Ice Age was the most recent cold phase of the 1-2 kyr cycle and that the North Atlantic tended to oscillate in a muted Dansgaard/Oeschger-like mode during the Holocene.

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues.Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. Many coasts feature sequences of Quaternary and Neogene shorelines that are shaped by a combination of sealevel oscillations and tectonics. We compiled a global synthesis of sea-level changes for the following highstands: MIS 1, MIS 3, MIS 5e and MIS 11. Also, we date the apparent onset of sequences of paleoshorelines either from published data or tentatively extrapolating an age for the uppermost, purported oldest shoreline in each sequence. Including the most documented MIS 5e benchmark, we identify 926 sequences out of which 185 also feature Holocene shorelines. Six areas are identified where elevations of the MIS 3 shorelines are known, and 31 feature elevation data for MIS 11 shorelines. Genetic relationships to regional geodynamics are further explored based on the elevations of the MIS 5e benchmark. Mean apparent uplift rates range from 0.01 ± 0.01 mm/yr (hotspots) to 1.47 ± 0.08 mm/yr (continental collision). Passive margins appear as ubiquitously uplifting, while tectonic segmentation is more important on active margins. From the literature and our extrapolations, we infer ages for the onset of formation for~180 coastal sequences. Sea level fingerprinting on coastal sequences started at least during mid Miocene and locally as early as Eocene. Whether due to the changes in the bulk volume of seawater or to the temporal variations in the shape of ocean basins, estimates of eustasy fail to explain the magnitude of the apparent sea level drop. Thus, vertical ground motion is invoked, and we interpret the longlasting development of those paleoshore sequences as the imprint of glacial cycles on globally uplifted margins in response to continental compression. The geomorphological expression of the sequences matches the amplitude and frequency of glacial cyclicity. From middle Pleistocene to present-day, moderately fast (100,000 yrs) oscillating sea levels favor the development of well identified strandlines that are distinct from one another. Pliocene and Lower Pleistocene strandlines associated with faster cyclicity (40,000 yrs) are more compact and easily merge into rasas, whereas older Cenozoic low-frequency eustatic changes generally led to widespread flat-lying coastal plains.