Subsurface ocean warming preceded Heinrich Events

Heinrich Events, marked by ice-rafted debris (IRD) layers in the North Atlantic, signify substantial freshwater release from the Laurentide Ice Sheet (LIS). While the climatic impact of these events is well-documented, the triggering mechanism remains debated. The "binge-purge" hypothesis suggests internal ice-sheet oscillations as the primary driver, while another prominent hypothesis emphasizes the role of weakened Atlantic Meridional Overturning Circulation (AMOC). Existing surface ocean temperature and IRD data show surface cooling preceding ice-rafting events, and AMOC weakening is observed before Heinrich Events. However, subsurface ocean temperature data near the LIS grounding line has been lacking, hindering the assessment of subsurface warming as a potential trigger. This study addresses this gap by analyzing high-resolution subsurface temperature and salinity reconstructions from sediment core GeoB18530-1 in the western subpolar North Atlantic to investigate the relationship between subsurface ocean conditions and the timing of Heinrich Events.

Previous research extensively documented the global climatic consequences of Heinrich Events, characterized by massive iceberg discharges into the North Atlantic. The "binge-purge" hypothesis posits that internal ice-sheet dynamics within the LIS drive these events, irrespective of external forcing. However, this is challenged by evidence of surface ocean cooling preceding Heinrich events, indicating that surface processes alone are insufficient to fully explain the phenomena. Other studies suggest that a weakened AMOC, leading to subsurface warming, plays a significant role. Numerical modeling simulations support the idea that a reduced AMOC causes subsurface warming, ultimately triggering rapid ice shelf retreat and LIS destabilization. While some studies showed mid-depth North Atlantic warming before Heinrich Events, the absence of dedicated subsurface temperature proxies near the Labrador Sea has limited the conclusive assessment of this mechanism's importance.

This research focused on sediment core GeoB18530-1, located east of Newfoundland. The core offers high temporal resolution (~250 years) for the last 27,000 years. Subsurface temperatures were reconstructed using the Mg/Ca ratio of the foraminifera *Neogloboquadrina pachyderma sinistral* (N. *pachyderma* sin.), a species inhabiting approximately 150 m water depth. The stable oxygen isotopic composition (δ¹⁸O) of N. *pachyderma* sin. was used to estimate salinity changes. The chronology was established using 20 accelerator mass spectrometer ¹⁴C ages. The study site's proximity to the Labrador Sea, within the IRD belt, and the presence of well-defined IRD layers in the core make it ideal for investigating the timing of ocean dynamics relative to Heinrich Events. IRD layers were further characterized using the Ca/Sr ratio from bulk sediments. The sensitivity of the Mg/Ca-derived temperatures to changes in habitat depth and seasonal cycle were carefully assessed using modern hydrographic data. Finally, the calculated oxygen isotopic composition of seawater (δ¹⁸Osw) was adjusted for ice volume changes to obtain the regional ice-volume-corrected δ¹⁸O record (δ¹⁸Oic). X-ray fluorescence core scanning was also used to analyze elemental ratios in the sediment core.

The analysis revealed a recurring pattern of significant subsurface ocean warming (subSSTMg/Ca) in the western subpolar North Atlantic, consistently preceding the deposition of Heinrich layers. The timing of the warming peaks, constrained by radiocarbon ages, showed a clear precursory relationship to Heinrich Events. For example, warmest subsurface ocean temperatures (8.4 °C and 12.5 °C) were synchronous with the onset of IRD deposition during Heinrich Events 2 and 1, respectively. These temperatures were considerably higher than the modern subsurface temperature (~7 °C) at a comparable depth. Following the onset of Heinrich Events, rapid cooling and freshening occurred, reflecting known meltwater intrusions. A subsequent rise in temperature and salinity was observed in the later phase of Heinrich Events, possibly corresponding to multiple ice-stream advances. The study found no such temperature increases prior to Heinrich Events in the NGRIP atmospheric temperature record or in North Atlantic sea surface temperatures. Rather, these surface temperatures showed opposite trends to the subsurface temperatures, suggesting strong isolation between subsurface and surface waters during these periods. Early subsurface warming was also observed in the eastern subpolar North Atlantic, suggesting large-scale heat storage in the subpolar North Atlantic. Finally, the study linked subsurface warming with AMOC weakenings, observing a 1-2 kyrs lead time for subsurface warming before each Heinrich Event.

The findings strongly support the hypothesis that subsurface ocean warming is a critical trigger for Heinrich Events. The consistent pattern of subsurface warming preceding IRD deposition suggests a causal relationship. The accumulation of heat near the ice-shelf grounding line likely destabilized the ice sheet, leading to rapid ice-margin retreat and iceberg discharge. This is consistent with existing ice-sheet modeling studies that highlighted the importance of subsurface ocean warming in initiating Heinrich Events. The close correlation between subsurface warming and AMOC weakening points to a significant role for ocean circulation changes in setting the stage for these events. The opposing temperature trends between the surface and subsurface waters emphasizes the crucial role of subsurface processes in driving Heinrich Events.

This study provides the first robust evidence that subsurface ocean warming in the western subpolar North Atlantic preceded Heinrich Events. The consistent pattern of warming before each event, coupled with its correlation with AMOC weakening, points to a critical role for subsurface ocean dynamics in triggering ice-sheet instabilities. The results have significant implications for understanding future climate change, as projected AMOC weakening could lead to increased interior-ocean warming, threatening the stability of modern marine-terminating Arctic glaciers and altering the North Atlantic freshwater budget. Future research should explore the detailed mechanisms of heat transport and storage within the subpolar Atlantic and further refine ice-sheet models to incorporate the observed subsurface warming dynamics.

The study relies on proxy data, which always have inherent uncertainties. While efforts were made to assess the sensitivity of the Mg/Ca-based temperature reconstructions to changes in habitat depth and seasonal cycle, some uncertainties remain. The spatial extent of subsurface warming is also constrained by the single core location; thus, the findings might not represent the whole subpolar Atlantic. The temporal resolution, although high compared to other studies, could be further improved to capture potentially finer-scale dynamics.