Sea-level constraints on the amplitude and source distribution of Meltwater Pulse 1A

Jean Liu, Glenn A. Milne, Robert E. Kopp, Peter U. Clark, Ian Shennan

Research output: Contribution to journalArticlepeer-review

62 Citations (Scopus)

Abstract

During the last deglaciation, sea levels rose as ice sheets retreated. This climate transition was punctuated by periods of more intense melting; the largest and most rapid of these-Meltwater Pulse 1A-occurred about 14,500 years ago, with rates of sea-level rise reaching approximately 4 m per century. Such rates of rise suggest ice-sheet instability, but the meltwater sources are poorly constrained, thus limiting our understanding of the causes and impacts of the event. In particular, geophysical modelling studies constrained by tropical sea-level records suggest an Antarctic contribution of more than seven metres, whereas most reconstructions from Antarctica indicate no substantial change in ice-sheet volume around the time of Meltwater Pulse 1A. Here we use a glacial isostatic adjustment model to reinterpret tropical sea-level reconstructions from Barbados, the Sunda Shelf and Tahiti. According to our results, global mean sea-level rise during Meltwater Pulse 1A was between 8.6 and 14.6 m (95% probability). As for the melt partitioning, we find an allowable contribution from Antarctica of either 4.1 to 10.0 m or 0 to 6.9 m (95% probability), using two recent estimates of the contribution from the North American ice sheets. We conclude that with current geologic constraints, the method applied here is unable to support or refute the possibility of a significant Antarctic contribution to Meltwater Pulse 1A.

Original languageEnglish
Pages (from-to)130-134
Number of pages5
JournalNature Geoscience
Volume9
Issue number2
Early online date21 Dec 2015
DOIs
Publication statusPublished (in print/issue) - Feb 2016

Fingerprint

Dive into the research topics of 'Sea-level constraints on the amplitude and source distribution of Meltwater Pulse 1A'. Together they form a unique fingerprint.

Cite this