Daniel Baggenstos’s research while affiliated with Australian Antarctic Division and other places

Using high-precision ice core measurements of CO2 , δ¹³C–CO2 , CH4 , and N2O, this study provides carbon isotope constraints on a sizeable, centennial-scale CO2 jump at the onset of Marine Isotope Stage 9 (MIS 9). The very end of the Heinrich stadial (HS) characterizing Termination IV (T-IV, ca. 343 to 333 ka ago) shows a 250-y-long jump in greenhouse gas concentrations, followed by a 1.3 ka gradual decline back to the initial concentration. During this so-called overshoot, CO2 and CH4 reach their highest levels (about 303 ppm and 800 ppb, respectively) over the past 800 ka prior to industrialization. The jump in CO2 is not accompanied by a change in δ¹³C–CO2, suggesting that multiple mechanisms contributed to the exceptionally elevated CO2 values. Following the jump, a slow 0.2‰ enrichment in δ¹³C–CO2 occurs. We propose that during the jump, the sudden resumption of deepwater formation in the North Atlantic (NA) triggered an amplified release of CO2 from the Southern Ocean (SO) by a northward shift of the Intertropical Convergence Zone (ITCZ) and the SO westerlies, potentially in combination with a rapid land carbon release. The latter is expected from temporally enhanced wildfire activity related to higher fuel load and regionally changing weather conditions in connection to the ITCZ shift. A combination of marine proxy records and box model simulation suggests that the δ¹³C–CO2 decrease expected from these processes is compensated by a net temperature increase in global sea surface temperature (SST) at the time of the AMOC resumption.

During deglaciations, Earth takes up vast amounts of energy, about half of which heats the global ocean. Thus, ocean heat content (OHC) is a key metric to assess Earth's energy budget. Recent modeling studies suggest that OHC changes not only in response to orbitally driven climate change but is also modulated on millennial timescales by the Atlantic Meridional Overturning Circulation (AMOC). Here, we present the first OHC record for the last four deglaciations using noble‐gas ratios in the EPICA Dome C ice core. The record reveals millennial‐scale OHC variability in all studied deglaciations, most prominently as OHC maxima at the end of Terminations II, III, and IV. These millennial‐scale OHC changes are anti‐correlated with AMOC strength, suggesting that the AMOC modulates OHC across different climate states. Furthermore, given the magnitude of the end‐of‐termination OHC maxima, AMOC‐induced OHC changes may be an important control of early interglacial atmospheric CO2, sea level, and climate.

Oxygen isotope ratios (δ¹⁸O) of foraminifera in marine sediment records have fundamentally shaped our understanding of the ice ages and global climate change. Interpretation of these records has, however, been challenging because they reflect contributions from both ocean temperature and ice volume. Here, instead of disentangling, we reconstruct global benthic foraminiferal δ¹⁸O across the last deglaciation (18–11.5 ka) with ice volume constraints from fossil corals and ocean temperature constraints from ice core noble gases. We demonstrate that, while ocean temperature and ice volume histories are distinct, their summed contributions to δ¹⁸O agree remarkably well with benthic δ¹⁸O records. Given the agreement between predicted and observed δ¹⁸O, we further build upon recent insight into global energy fluxes and introduce a framework to quantitively reconstruct top-of-atmosphere net radiative imbalance, or Earth’s energy imbalance, from δ¹⁸O. Finally, we reconstruct 150,000 years of energy imbalance, which broadly follows Northern Hemisphere summer insolation but shows millennial-scale energy gain during the cold intervals surrounding Heinrich events. This suggests that, in addition to external forcing, internal variability plays an important role in modifying the global energy budget on long (millennial-plus) timescales.

Plain Language Summary Most of the heat added to the climate system by anthropogenic climate change is taken up by the oceans. To better understand how the ocean responds to climate change over hundreds to thousands of years, an indirect measure for the mean ocean temperature (MOT) based on the temperature‐dependent solubility of noble gases has been developed. Noble gas concentrations of the past atmosphere are archived in air bubbles in polar ice cores, which have been used to reconstruct the MOT of the past 20,000 years when Earth's climate was propelled out of the last ice age. However, uncertainties remain regarding critical parameters that are required to derive the correct MOT of the past. Here we make use of an Earth system model that explicitly simulates the noble gases and allows us to assess these parameters in detail under modern and past climate conditions. We find that changes in wind, sea‐ice, and ocean circulation all play important roles in the partitioning of noble gases between the atmosphere and ocean. By taking these effects into account our model suggests a revised best‐estimate MOT cooling of the last ice age to −2.1 ± 0.7°C, which is about 0.5°C warmer than previous estimates.

Here we present a newly developed ice core gas-phase proxy that directly samples a component of the large-scale atmospheric circulation: synoptic-scale pressure variability. Surface pressure changes weakly disrupt gravitational isotopic settling in the firn layer, which is recorded in krypton-86 excess (86Krxs). The 86Krxs may therefore reflect the time-averaged synoptic pressure variability over several years (site “storminess”), but it likely cannot record individual synoptic events as ice core gas samples typically average over several years. We validate 86Krxs using late Holocene ice samples from 11 Antarctic ice cores and 1 Greenland ice core that collectively represent a wide range of surface pressure variability in the modern climate. We find a strong spatial correlation (r=-0.94, p<0.01) between site average 86Krxs and time-averaged synoptic variability from reanalysis data. The main uncertainties in the analysis are the corrections for gas loss and thermal fractionation and the relatively large scatter in the data. Limited scientific understanding of the firn physics and potential biases of 86Krxs require caution in interpreting this proxy at present. We show that Antarctic 86Krxs appears to be linked to the position of the Southern Hemisphere eddy-driven subpolar jet (SPJ), with a southern position enhancing pressure variability. We present a 86Krxs record covering the last 24 kyr from the West Antarctic Ice Sheet (WAIS) Divide ice core. Based on the empirical spatial correlation of synoptic activity and 86Krxs at various Antarctic sites, we interpret this record to show that West Antarctic synoptic activity is slightly below modern levels during the Last Glacial Maximum (LGM), increases during the Heinrich Stadial 1 and Younger Dryas North Atlantic cold periods, weakens abruptly at the Holocene onset, remains low during the early and mid-Holocene, and gradually increases to its modern value. The WAIS Divide 86Krxs record resembles records of monsoon intensity thought to reflect changes in the meridional position of the Intertropical Convergence Zone (ITCZ) on orbital and millennial timescales such that West Antarctic storminess is weaker when the ITCZ is displaced northward and stronger when it is displaced southward. We interpret variations in synoptic activity as reflecting movement of the South Pacific SPJ in parallel to the ITCZ migrations, which is the expected zonal mean response of the eddy-driven jet in models and proxy data. Past changes to Pacific climate and the El Niño–Southern Oscillation (ENSO) may amplify the signal of the SPJ migration. Our interpretation is broadly consistent with opal flux records from the Pacific Antarctic zone thought to reflect wind-driven upwelling. We emphasize that 86Krxs is a new proxy, and more work is called for to confirm, replicate, and better understand these results; until such time, our conclusions regarding past atmospheric dynamics remain speculative. Current scientific understanding of firn air transport and trapping is insufficient to explain all the observed variations in 86Krxs. A list of suggested future studies is provided.

Precision, accuracy, and temporal resolution are key to making full use of atmospheric trace gas records in ice cores. These aspects will become especially crucial for ice cores that aim to extend the ice core record to the last 1.5 Myr, i.e., across the Mid-Pleistocene Transition (as currently drilled within the European project Beyond EPICA – Oldest Ice Core (BE-OIC)). The ice from this period is expected to be close to bedrock and, due to glacier flow, extremely thinned with 15 000 years of climate history contained in only 1 m of ice. Accordingly, for a century-scale resolution, the sample vertical extent must be reduced to a few centimeters containing only about 1–2 mL air STP. We present a novel combined system for the extraction and the simultaneous measurement of CO2, CH4, and N2O concentrations, as well as δ13CO2, which achieves a vertical resolution of 1–2 cm (3.5×3.5 cm cross section) with precisions of 0.4 ppm, 3 ppb, 1 ppb, and 0.04 ‰, respectively, in sublimation tests with standard gas over gas-free ice. This is accomplished by employing a directional and continuous laser-induced sublimation followed by analysis of the sample gas by a quantum cascade laser absorption spectrometer (QCLAS). Besides the low sample volume requirements and the vertical resolution capabilities, the described method holds additional advantages over previous methods, including the immunity of the highly specific QCLAS analysis to drilling fluid contamination as well as the non-destructive nature of the spectroscopic gas analysis. The combined extraction and analysis system was extensively tested by sublimating gas-free ice with introduction of a standard gas to determine the accuracy and characterize potential artifacts. Moreover, Antarctic ice samples were measured to confirm the measurement performance, covering the range of variability expected in Pleistocene ice and highlighting the vertical resolution capabilities critical for its application within BE-OIC.

Precision, accuracy, and temporal resolution are key to make full use of atmospheric trace gas records in ice cores. These aspects will become especially crucial for ice cores that aim to extend the ice core record to the last 1.5 Myr, i.e., across the Mid Pleistocene Transition (as currently drilled within the European project Beyond EPICA – Oldest Ice Core (BE-OIC)). The ice from this period is expected to be close to bedrock and, due to glacier flow, extremely thinned with 15,000 years of climate history contained in only one meter of ice. Accordingly, for a century-scale resolution, the sample vertical extent must be reduced to a few cm containing only about 1−2 mL air STP. We present a novel combined system for the extraction and the simultaneous measurement of CO2, CH4, and N2O concentrations, as well as δ13CO2, which achieves a vertical resolution of 1−2 cm with precisions of 0.4 ppm, 3 ppb, 1 ppb and 0.04 ‰, respectively. This is accomplished by employing a directional and continuous laser induced sublimation followed by analysis of the sample gas by quantum cascade laser absorption spectroscopy (QCLAS). Besides the low sample volume requirements and the vertical resolution capabilities, the described method holds additional advantages over previous methods, including the immunity of the highly specific QCLAS analysis to drilling fluid contamination as well as the non-destructive nature of the spectroscopic gas analysis. The combined extraction and analysis system was extensively tested by sublimating gas-free ice with introduction of a standard gas to determine the accuracy and characterize potential artefacts. Moreover, Antarctic ice samples were measured to confirm the measurement performance, covering the range of variability expected in Pleistocene ice and to highlight the vertical resolution capabilities critical for its application within BE-OIC.

Here we present a newly developed ice core gas-phase proxy that directly samples a component of the large-scale atmospheric circulation: synoptic-scale pressure variability. Surface pressure variability weakly disrupts gravitational isotopic settling in the firn layer, which is recorded in krypton-86 excess (86Krxs). We validate 86Krxs using late Holocene ice samples from eleven Antarctic and one Greenland ice core that collectively represent a wide range of surface pressure variability in the modern climate. We find a strong correlation (r = -0.94, p < 0.01) between site-average 86Krxs and site synoptic variability from reanalysis data. The main uncertainties in the method are the corrections for gas loss and thermal fractionation, and the relatively large scatter in the data. We show 86Krxs is linked to the position of the eddy-driven subpolar jet (SPJ), with a southern position enhancing pressure variability. We present a 86Krxs record covering the last 24 ka from the WAIS Divide ice core. West Antarctic synoptic activity is slightly below modern levels during the last glacial maximum (LGM); increases during the Heinrich Stadial 1 and Younger Dryas North Atlantic cold periods; weakens abruptly at the Holocene onset; remains low during the early and mid-Holocene, and gradually increases to its modern value. The WAIS Divide 86Krxs record resembles records of monsoon intensity thought to reflect changes in the meridional position of the intertropical convergence zone (ITCZ) on orbital and millennial timescales, such that West Antarctic storminess is weaker when the ITCZ is displaced northward, and stronger when it is displaced southward. We interpret variations in synoptic activity as reflecting movement of the South Pacific SPJ in parallel to the ITCZ migrations, which is the expected zonal-mean response of the eddy-driven jet in models and proxy data. Past changes to Pacific climate and the El Niño Southern Oscillation (ENSO) may amplify the signal of the SPJ migration. Our interpretation is broadly consistent with opal flux records from the Pacific Antarctic zone thought to reflect wind-driven upwelling. We emphasize that 86Krxs is a new proxy, and more work is called for to confirm, replicate and better understand these results; until such time, our conclusions regarding past atmospheric dynamics remain tentative. Current scientific understanding of firn air transport and trapping is insufficient to explain all the observed variations in 86Krxs.