Thermohaline instability and the formation of glacial North Atlantic super polynyas at the onset of Dansgaard‐Oeschger warming events

Abstract

Late Quaternary rapid warming events inferred on the basis of oxygen isotopic data from Greenland ice cores are the most prominent characteristic of millennial‐scale Dansgaard‐Oeschger oscillations. In a coupled climate model simulation which has accurately reproduced this oscillatory behavior for the first time, we show that formation of a glacial North Atlantic super polynya characterizes the initial stage of transition from cold stadial to warm interstadial conditions. The winter polynya forms within the otherwise sea ice‐covered North Atlantic as a consequence of the onset of a thermohaline convective instability beneath an extensive stadial sea ice lid. Early in the stadial period, the tendency for thermal convective instability of an extensive warm pool beneath the sea ice lid is strongly inhibited by a stabilizing vertical salinity gradient, which gradually diminishes until a thermohaline convective instability occurs that leads to polynya formation and the rapid retreat of North Atlantic sea ice cover.

Assimilating the ICE‐6G_C Reconstruction of the Latest Quaternary Ice Age Cycle Into Numerical Simulations of the Laurentide and Fennoscandian Ice Sheets

“…Looking beyond these linear responses of ice thickness to surface climate boundary conditions, the highlighted regions in Figure emphasize that there are also high amplitude and largely τ f ‐independent local anomalies that evidence our perturbative approach breaking down in a few isolated, small‐scale North American regions even as it remains valid elsewhere. The appearance of sensitive regions covering parts of the modern Hudson Strait and Baffin Bay is consistent with literature discussions about the dynamical origins of Heinrich Events and D‐O oscillations (e.g., Hemming, ; MacAyeal, ; Peltier & Vettoretti, ; Vettoretti & Peltier, , ), and the spatial structure of the anomalies in Figure is, moreover, consistent with patterns seen in more idealized PISM‐based analyses of ice streaming (Robel & Tziperman, ). The results are in this regard a very useful starting point both for improved, fully coupled ice dynamical simulations and for refinements of the pure GIA‐based I6G reconstruction.…”

Objective extraction and analysis of statistical features of Dansgaard–Oeschger events

“…Different drivers for these AMOC changes have been proposed, including North Atlantic freshwater forcing (Ganopolski and Rahmstorf, 2001;Timmermann et al, 2003;Kageyama et al, 2013), variations in ice sheet topography (Zhang et al, 2014), and atmospheric CO 2 (Zhang et al, 2017). On the other hand, unforced millennial-scale oscillations involving the AMOC have been reported in comprehensive climate models (Vettoretti and Peltier, 2016;Klockmann et al, 2018). In these oscillations, sea ice variability in ocean convection areas plays an important role, which has been proposed previously (Li et al, 2010;Dokken et al, 2013;Petersen et al, 2013) and is supported by recent proxy records (Sadatzki et al, 2019).…”

“…However, evidence for a different scenario might arise with new data or analyses, as in the studies by Rypdal (2016) and Boers (2018), who suggest a bifurcation in a fast climate subsystem before DO warmings, evidenced by increases in high-frequency variance of the ice core record prior to some events. If this is the case, it would mean that the transitions are not purely noise-induced but predictable to some degree and potentially part of a self-sustained oscillation such as in the model experiments by Vettoretti and Peltier (2016) and Klockmann et al (2018).…”

Equilibrium simulations of Marine Isotope Stage 3 climate

“…The ocean bathymetry is adapted based on an estimated sea level lowering of 70 m below present day (Waelbroeck et al, 2002). As a consequence, many shallow ocean grid points on the shelf turn into land, thereby modifying the land-sea mask (Fig.…”