Antarctic contribution to Last Interglacial sea level from observations in northwest Europe

The Last Interglacial (LIG; approx. 115 - 130 ka) can provide an important testing ground for how ice sheets respond to warming. However, the magnitude, sources, and timing of sea-level change during this period remain debated. Here we present new sea-level observations with optically dated numerical age constraints from southwest Britain and combine them with published LIG sea level data from the UK, northern France, and Denmark. These locations are a key target for studying LIG sea level because of their low sensitivity to Greenland mass loss, which allows us to constrain the Antarctic contribution to Last Interglacial global mean sea level (GMSL).

We correct the sea-level observations for long-term uplift and combine them with a suite of 576 deglacial GIA models in which we vary both Earth structure and the size and timing of Eurasian ice sheet collapse during the penultimate deglaciation. We use Bayesian inference to identify those GIA models that are most consistent with the spatial and temporal trends of the observations and obtain a posterior GIA-corrected sea level. Next, we use 8 different published scenarios of Greenland ice sheet evolution over the LIG that vary in their maximum ice sheet contribution to GMSL from 0.5 m to 5 m. We calculate the respective time-varying fingerprint in northwest Europe and remove it from our posterior GIA corrected sea level along with an estimate of the contribution from mountain glaciers and thermal expansion. This results in an estimate of Antarctic ice sheet melt.

We find that the Antarctic contribution to LIG global mean sea level peaked early in the deglaciation (likely prior to 126 ka) with a most likely peak contribution of 5.6m (3.3 - 8.8 m; 68% conf. interval). The Antarctic contribution gradually declines over the course of the LIG to around 1 m by 118 ka. Our results suggest an asynchronous melt history of an early southern hemisphere contribution followed by later Greenland ice-sheet mass loss. Further work will be needed to combine these results with records from other locations to provide accurate estimates for both Antarctic and Greenland ice melt.