2023 KIRK BRYAN AWARD: THE IMPORTANCE OF FOLLOWING THE UNEXPECTED: PALEOCLIMATE INSIGHTS FROM BENEATH BAFFIN ISLAND ICE

Here we present a case in which unexpected discoveries and seemingly conflicting evidence led to profound insights into the magnitude and character of modern climate change. Thin, cold-based ice caps on Baffin Island in Arctic Canada are retreating rapidly under current warming, exposing in situ tundra vegetation killed and preserved by earlier ice expansion. While the plant radiocarbon ages constrain the time of their death by ice cap expansion, their preservation (and ephemeral nature after exposure) also suggests continuous ice coverage since the initial ice advance.

Newly exposed preserved vegetation was initially pursued to reconstruct Baffin ice cap activity during Neoglaciation, and by extension Holocene climate evolution. However, during fieldwork, preserved plants with radiocarbon ages >40 ka were discovered at 30 ice caps, leading to a new hypothesis that some landscapes being exposed today have been ice covered since the most recent warm time prior to ~40 ka: almost certainly the last interglacial period, ~115 ka. We measured and modeled in situ cosmogenic ¹⁴C produced in un-eroded bedrock adjacent to the radiocarbon-dead plants to test this hypothesis. Together, the radiocarbon-dead vegetation ages, the low in situ ¹⁴C concentrations, and the modeled ice cover histories confirm continuous ice cover for >40 ka, suggesting that present summer warmth is greater than at any time since the last interglacial.

A second unexpected discovery came in a suite of ~9 ka vegetation ¹⁴C ages from 8 Baffin Island ice caps, documenting an early Holocene ice advance as peak Holocene summer insolation declined. How can we reconcile concurrent exposure of Neoglacial, early Holocene, and radiocarbon-dead vegetation across Baffin Island at a range of elevations (~500-1600 m asl)? Some ice caps mantle low-relief, low-elevation areas where the ice is not topographically constrained, while others mantle low-relief, high-elevation areas (‘pedestals’) between deeply incised fjords. The maximum extent and thickness of these pedestal ice caps is determined by the pedestal area and ice rheology. We propose that a combination of the asymmetrical growth/demise of ice caps and the recent rapid, large magnitude rise in ELA is prompting melt across all elevations, simultaneously exposing preserved vegetation of diverse ages.