Last Interglacial Sea Levels
I recently attended a talk by Professor Chris Turney, now at the University of Technology Sydney, on his research in West Antarctica. He spoke of work on the response of the Antarctic ice sheet to rising temperatures alluding to potential future behaviour of the ice sheet under global warming. With colleagues he has published a paper on how early Last Interglacial (LIG) ocean warming impacted on ice mass loss (PNAS, February 25, 2020, v.117, 3996-4006). What excited me was his reference in the talk to the extraordinary insights of John Mercer, a glaciologist, who in 1978 published a paper in Nature (v.271, 321-325) on “West Antarctic ice sheet and CO2 greenhouse effect: a threat of disaster”.
The LIG (Marine Isotope Stage MIS 5e) has long fascinated me as a result of field work in the 1960s and 70s in Australia and North America. LIG spans the period c.130000 to 116000 years ago. Recognition and dating of marine deposits at sites along the NSW coast indicated both a high sea level of c. 4m above present and warmer ocean temperatures. I discussed one site, Largs in the Hunter Valley, in blog 194, July 2021. Colin Murray-Wallace with whom I had the pleasure of working with at this site, in 1991 undertook a review of LIG sites around Australia (Murray-Wallace and Belperio, Quaternary Science Reviews, 10, 441-461). This work was updated in the excellent book that he and Colin Woodroffe published in 2014 on “Quaternary Sea-Level Changes” (Cambridge University Press, 484pp).
In 1978, John Chappell and I had a short note in Nature (v. 272, 809-810) that brought together our observations involving uranium series dating of corals from PNG and NSW. We were testing the hypothesis of a possible Antarctic ice surge during the LIG proposed by Wilson for East Antarctica who had extended earlier ideas attributed to Mercer in 1968 on a surge from West Antarctica. However, it was Mercer’s 1978 Nature paper mentioned in the talk by Chris Turney that stimulated me to look again at recent publications on the LIG. It is now likely that global sea levels could be as much as 6-9 m above present during the LIG, but were there two distinct maxima? It was clear this had become a hot topic given implications of what occurred in the LIG relative to present-day global warming conditions (e.g. Siddell et al., 2013, PAGES News 21 (1) 36-37, 22498). In the book cited above, the two Colins summarise literature up to 2014 in Section 6.9.3 teasingly titled “Global estimates of last interglacial sea levels—the sanctity of the 6m APSL datum?”. This section has many useful references on the topic.
What is at issue here are apparent differences in the interpretation of global sea level positions during the LIG based on coastal observations, ocean oxygen isotope studies, and modelling involving isostatic adjustment of coastal areas in various locations. The existence of differences is critical in understanding the behaviour of continental ice during interglacial conditions. It is not just a question of how much ice melted to raise sea level above present, but what were the major sources of melting, and whether at some stage during the LIG there was an ice sheet collapse in either west or east Antarctica as we were (following Mercer and others) speculating 50 plus years ago. Furthermore if the LIG peaked more than twice as suggested by some (e.g. Rohling et al., 2008, Nature Geoscience, 1, 38-42), then as noted by Siddell and colleagues more creative thinking in terms of mechanisms would be required to explain what is driving these oscillations.
Evidence for oscillations of the magnitude pointing to a dramatic ice sheet event is far from conclusive as outlined by Dutton and Lambeck in Science (2012, v.337, 216-219). They examined a global database of coral U-Th ages in the context of isostatic modelling and stratigraphic evidence. They are quite cautious in the interpretation of data for LIG sea levels indicating the need to improve uncertainty in global sea level estimates. From their work 2-4m of the higher sea level could come from Greenland, and +3.3m from West Antarctica. This would account for the lower estimate single peak of +5.5m. However, the upper limit of +9m “implies additional meltwater contributions from adjacent sections of East Antarctica”. Their modeling does not invoke any late LIG surge effect, a conclusion that is consistent in oxygen isotope records from deep-sea cores.
Two well documented studies from the Australian coast have added to the discussion of what was happening in a relatively stable tectonic context. O’Leary and colleagues in 2013 published in Nature Geoscience on “Ice sheet collapse following a prolonged period of stable sea level during the last interglacial”. They presented a database of well-dated corals from LIG shorelines along the WA coast between Cape Vlaming and Cape Leeuwin. Combined with geophysical modeling they were able to “robustly constrain the timing of ice sheet collapse during MIS 5e”. The conclusion reached was that between 127 and 119 kyr ago, the global sea level remained relatively stable at about 3-4m above present, but stratigraphically a younger set of corals with U-series ages of around c.118 kyr occur up to about 9m. From this study the view expressed was that “in the last few thousand years of the interglacial, a critical ice sheet instability threshold was crossed, resulting in the catastrophic collapse of polar ice sheets and substantial sea-level rise”.
The second well-documented LIG site is from northwest Tasmania. Ian Goodwin and colleagues have examined a magnificent strandplain using OSL as the dating tool (Goodwin et al., 2023, “Robbins Island: the index site for regional Last Interglacial sea level, wave climate and the subtropical ridge around Bass Strait, Australia”, Quaternary Science Reviews, 1 April 2023, v.305, 107996). Observed relative sea-level (RSL) history was compared with modelled RSL that accounted for the theoretical fall in RSL through the LIG due to glacio-isostatic forcing at this far-field location. They identified three phases of RSL change: (1) a fall between c. 129 to 126 kyr; (2) a stillstand between 126 and 119kyr at c. 5.75m above present mean sea level; and (3) another fall between 118 and 114kyr. It is concluded that the “stillstand departure from Glacio-Isostatic Adjustment (GIA) theory, points unambiguously to persistent polar meltwater contributions to sea level of c. 2m from 126 to 119kyr, where the component of RSL fall due to GIA was balanced by the RSL rise from meltwater”. From a climatological perspective their work on the LIG implies a 5-degree poleward shift of the Subtropical Ridge during this period.
Returning now to the contribution from Chris Turney and colleagues over concerns about the sensitivity of Antarctica to rising ocean temperatures. They have provided multiple lines of evidence for substantial mass loss of ice across the Weddell Sea during the LIG. It is recognised that testing the hypothesis of Mercer and others of this event has proved challenging. However, their ice-sheet modelling predicts that Antarctica may have contributed several metres to global sea level during the LIG . This suggests “that this ice sheet lies close to a “tipping point” under projected warming”.
Two final points: first I am saddened but not surprised that the contributions of the late John Chappell to this problem from his work on tectonically uplifted coral reefs on the Huon Peninsula of PNG get very infrequently cited these days. I must do a blog on this someday. Second, along the NSW coast there is scope for doing more work similar to that of Goodwin et al. on LIG strandplains with known morphostratigraphies but with limited geochronology.
Bruce Thom
Words by Prof Bruce Thom. Please respect the author’s thoughts and reference appropriately: (c) ACS, 2025. For correspondence about this blog post please email admin@australiancoastalsociety.org.au
#278