Search

High-resolution ice core records from coastal Antarctica are particularly useful to inform our understanding of environmental changes and their drivers. Here, we present a decadally resolved record of sea-salt sodium (a proxy for open-ocean area) and non-sea salt calcium (a proxy for continental dust) from the well-dated Roosevelt Island Climate Evolution (RICE) core, focusing on the time period between 40-26 ka BP. The RICE dust record exhibits an abrupt shift towards a higher mean dust concentration at 32 ka BP. Investigating existing ice-core records, we find this shift is a prominent feature across Antarctica. We propose that this shift is linked to an equatorward displacement of Southern Hemisphere westerly winds. Subsequent to the wind shift, data suggest a weakening of Southern Ocean upwelling and a decline of atmospheric CO2 to lower glacial values, hence making this shift an important glacial climate event with potentially important insights for future projections.

Abstract. Commonly, three mechanisms of firn air transport are distinguished: molecular diffusion, advection, and near-surface convective mixing. Here we identify and describe a fourth mechanism, namely dispersion driven by synoptic-scale surface pressure variability (or barometric pumping). We use published gas chromatography experiments on firn samples to derive the along-flow dispersivity of firn, and combine this dispersivity with a dynamical air pressure propagation model forced by surface air pressure time series to estimate the magnitude of dispersive mixing in the firn. We show that dispersion dominates mixing within the firn lock-in zone. Trace gas concentrations measured in firn air samples from various polar sites confirm that dispersive mixing occurs. Including dispersive mixing in a firn air transport model suggests that our theoretical estimates have the correct order of magnitude, yet may overestimate the true dispersion. We further show that strong barometric pumping, such as at the Law Dome site, may reduce the gravitational enrichment of δ15N–N2 and other tracers below gravitational equilibrium, questioning the traditional definition of the lock-in depth as the depth where δ15N enrichment ceases. Last, we propose that 86Kr excess may act as a proxy for past synoptic activity (or paleo-storminess) at the site.

Ice cores from the northwestern Tibetan Plateau (NWTP) contain long records of regional climate variability, but refrozen meltwater and dust in these cores has hampered development of robust timescales. Here, we introduce an approach to dating the ice via the isotopic composition of atmospheric O2 in air bubbles (δ18Oatm), along with annual layer counting and radiocarbon dating. We provide a robust chronology for water isotope records (δ18Oice and d-excess) from three ice cores from the Guliya ice cap in the NWTP. The measurement of δ18Oatm, although common in polar ice core timescales, has rarely been used on ice cores from low-latitude, high-altitude glaciers due to (1) low air pressure, (2) the common presence of refrozen melt that adds dissolved gases and reduces the amount of air available for analysis, and (3) the respiratory consumption of molecular oxygen (O2) by micro-organisms in the ice, which fractionates the δ18O of O2 from the atmospheric value. Here, we make corrections for melt and respiration to address these complications. The resulting records of water isotopes from the Guliya ice cores reveal climatic variations over the last 15,000 y, the timings of which correspond to those observed in independently dated lake and speleothem records and confirm that the Guliya ice cap existed before the Holocene. The millennial-scale drivers of δ18Oice are complex and temporally variable; however, Guliya δ18Oice values since the mid-20th century are the highest since the beginning of the Holocene and have increased with regional air temperature.

Our understanding of climate dynamics during millennial-scale events is incomplete, partially due to the lack of their precise phase analyses under various boundary conditions. Here we present nine speleothem oxygen-isotope records from mid-to-low-latitude monsoon regimes with sub-centennial age precision and multi-annual resolution, spanning the Heinrich Stadial 2 (HS2) - a millennial-scale event that occurred at the Last Glacial Maximum. Our data suggests that the Greenland and Antarctic ice-core chronologies require +320- and +400-year adjustments, respectively, supported by extant volcanic evidence and radiocarbon ages. Our chronological framework shows a synchronous HS2 onset globally. Our records precisely characterize a centennial-scale abrupt "tropical atmospheric seesaw" superimposed on the conventional "bipolar seesaw" at the beginning of HS2, implying a unique response/feedback from low-latitude hydroclimate. Together with our observation of an early South American monsoon shift at the HS2 termination, we suggest a more active role of low-latitude hydroclimate dynamics underlying millennial events than previously thought.

Abstract. Gas isotope thermometry using the isotopes of molecular nitrogen and argonhas been used extensively to reconstruct past surface temperature changefrom Greenland ice cores. The gas isotope ratios δ15N andδ40Ar in the ice core are each set by the amount ofgravitational and thermal fractionation in the firn. The gravitationalcomponent of fractionation is proportional to the firn thickness, and thethermal component is proportional to the temperature difference between thetop and bottom of the firn column, which can be related to surfacetemperature change. Compared to Greenland, Antarctic climate change istypically more gradual and smaller in magnitude, which results in smallerthermal fractionation signals that are harder to detect. This has hamperedapplication of gas isotope thermometry to Antarctic ice cores. Here, we present an analytical method for measuring δ15N andδ40Ar with a precision of 0.002 ‰ per atomicmass unit, a two-fold improvement on previous work. This allows us toreconstruct changes in firn thickness and temperature difference at the SouthPole between 30 and 5 kyr BP. We find that variability in firn thickness iscontrolled in part by changes in snow accumulation rate, which is, in turn,influenced strongly by the along-flowline topography upstream of the icecore site. Variability in our firn temperature difference record cannot beexplained by annual-mean processes. We therefore propose that the ice coregas isotopes contain a seasonal bias due to rectification of seasonalsignals in the upper firn. The strength of the rectification also appears tobe linked to fluctuations in the upstream topography. As further evidencefor the existence of rectification, we present new data from the Dome Fujiice core that are also consistent with a seasonal bias throughout theHolocene. Our findings have important implications for the interpretation of ice coregas records. For example, we show that the effects of upstream topography onice core records can be significant at flank sites like the South Pole – theyare responsible for some of the largest signals in our record. Presumablyupstream signals impact other flank-flow ice cores such as EDML, Vostok, andEGRIP similarly. Additionally, future work is required to confirm theexistence of seasonal rectification in polar firn, to determine how spatiallyand temporally widespread rectifier effects are, and to incorporate therelevant physics into firn air models.

Here we use high-precision carbon isotope data (δ13C-CO2) to show atmospheric CO2 during Marine Isotope Stage 4 (MIS 4, ~70.5-59 ka) was controlled by a succession of millennial-scale processes. Enriched δ13C-CO2 during peak glaciation suggests increased ocean carbon storage. Variations in δ13C-CO2 in early MIS 4 suggest multiple processes were active during CO2 drawdown, potentially including decreased land carbon and decreased Southern Ocean air-sea gas exchange superposed on increased ocean carbon storage. CO2 remained low during MIS 4 while δ13C-CO2 fluctuations suggest changes in Southern Ocean and North Atlantic air-sea gas exchange. A 7 ppm increase in CO2 at the onset of Dansgaard-Oeschger event 19 (72.1 ka) and 27 ppm increase in CO2 during late MIS 4 (Heinrich Stadial 6, ~63.5-60 ka) involved additions of isotopically light carbon to the atmosphere. The terrestrial biosphere and Southern Ocean air-sea gas exchange are possible sources, with the latter event also involving decreased ocean carbon storage.

The history of atmospheric oxygen (PO2) and the processes that act to regulate it remain enigmatic because of difficulties in quantitative reconstructions using indirect proxies. Here, we extend the ice-core record of PO2 using 1.5-million-year-old (Ma) discontinuous ice samples drilled from Allan Hills Blue Ice Area, East Antarctica. No statistically significant difference exists in PO2 between samples at 1.5 Ma and 810 thousand years (ka), suggesting that the Late-Pleistocene imbalance in O2 sources and sinks began around the time of the transition from 40- to 100-ka glacial cycles in the Mid-Pleistocene between ~1.2 Ma and 700 ka. The absence of a coeval secular increase in atmospheric CO2 over the past ~1 Ma requires negative feedback mechanisms such as Pco2-dependent silicate weathering. Fast processes must also act to suppress the immediate Pco2 increase because of the imbalance in O2 sinks over sources beginning in the Mid-Pleistocene.

The atmospheric history of molecular hydrogen (H2) from 1852 to 2003 was reconstructed from measurements of firn air collected at Megadunes, Antarctica. The reconstruction shows that H2 levels in the southern hemisphere were roughly constant near 330 parts per billion (ppb; nmol H2 mol-1 air) during the mid to late 1800s. Over the twentieth century, H2 levels rose by about 70% to 550 ppb. The reconstruction shows good agreement with the H2 atmospheric history based on firn air measurements from the South Pole. The broad trends in atmospheric H2 over the twentieth century can be explained by increased methane oxidation and anthropogenic emissions. The H2 rise shows no evidence of deceleration during the last quarter of the twentieth century despite an expected reduction in automotive emissions following more stringent regulations. During the late twentieth century, atmospheric CO levels decreased due to a reduction in automotive emissions. It is surprising that atmospheric H2 did not respond similarly as automotive exhaust is thought to be the dominant source of anthropogenic H2. The monotonic late twentieth century rise in H2 levels is consistent with late twentieth-century flask air measurements from high southern latitudes. An additional unknown source of H2 is needed to explain twentieth-century trends in atmospheric H2 and to resolve the discrepancy between bottom-up and top-down estimates of the anthropogenic source term. The firn air-based atmospheric history of H2 provides a baseline from which to assess human impact on the H2 cycle over the last 150 y and validate models that will be used to project future trends in atmospheric composition as H2 becomes a more common energy source.

AbstractRapid Access Ice Drill is a new drilling technology capable of quickly accessing the glacial bed of Antarctic ice sheets, retrieving ice core and rock core samples, and providing boreholes for downhole logging of physical properties. Scientific goals include searching for old ice near the glacial bed and sampling subglacial bedrock. During field trials near McMurdo Station on a piedmont glacier at Minna Bluff in the 2019–20 austral summer, we successfully completed a ‘top-to-bottom’ operational sequence in three boreholes by (1) augering through firn, (2) creating a borehole packer seal in non-porous ice, (3) establishing fluid circulation, (4) quickly drilling a borehole in ice at penetration rates up to 1.2 m min−1, (5) acquiring a short ice core at depth, (6) penetrating the glacial bed at a depth of ~677 m, (7) recovering a 3.2 m core of ice, basal till and subglacial bedrock, (8) optically logging the borehole on wireline, (9) testing hydrofracture potential by overpressuring the borehole fluid and (10) operating in an environmentally benign yet rapid field mode. Minna Bluff testing, therefore, demonstrates the effectiveness of this integrated system to drill rapidly through thick ice and penetrate across the glacial bed to take cores of bedrock.

Abrupt climate changes during the last deglaciation have been well preserved in proxy records across the globe. However, one long-standing puzzle is the apparent absence of the onset of the Heinrich Stadial 1 (HS1) cold event around 18 ka in Greenland ice core oxygen isotope δ18O records, inconsistent with other proxies. Here, combining proxy records with an isotope-enabled transient deglacial simulation, we propose that a substantial HS1 cooling onset did indeed occur over the Arctic in winter. However, this cooling signal in the depleted oxygen isotopic composition is completely compensated by the enrichment because of the loss of winter precipitation in response to sea ice expansion associated with AMOC slowdown during extreme glacial climate. In contrast, the Arctic summer warmed during HS1 and YD because of increased insolation and greenhouse gases, consistent with snowline reconstructions. Our work suggests that Greenland δ18O may substantially underestimate temperature variability during cold glacial conditions.

Water-stable isotopes in polar ice cores are a widely used temperature proxy in paleoclimate reconstruction, yet calibration remains challenging in East Antarctica. Here, we reconstruct the magnitude and spatial pattern of Last Glacial Maximum surface cooling in Antarctica using borehole thermometry and firn properties in seven ice cores. West Antarctic sites cooled ~10°C relative to the preindustrial period. East Antarctic sites show a range from ~4° to ~7°C cooling, which is consistent with the results of global climate models when the effects of topographic changes indicated with ice core air-content data are included, but less than those indicated with the use of water-stable isotopes calibrated against modern spatial gradients. An altered Antarctic temperature inversion during the glacial reconciles our estimates with water-isotope observations.

The magnitude of global cooling during the Last Glacial Maximum (LGM, the coldest multimillennial interval of the last glacial period) is an important constraint for evaluating estimates of Earth's climate sensitivity1,2. Reliable LGM temperatures come from high-latitude ice cores3,4, but substantial disagreement exists between proxy records in the low latitudes1,5-8, where quantitative low-elevation records on land are scarce. Filling this data gap, noble gases in ancient groundwater record past land surface temperatures through a direct physical relationship that is rooted in their temperature-dependent solubility in water9,10. Dissolved noble gases are suitable tracers of LGM temperature because of their complete insensitivity to biological and chemical processes and the ubiquity of LGM-aged groundwater around the globe11,12. However, although several individual noble gas studies have found substantial tropical LGM cooling13-16, they have used different methodologies and provide limited spatial coverage. Here we use noble gases in groundwater to show that the low-altitude, low-to-mid-latitude land surface (45 degrees south to 35 degrees north) cooled by 5.8 ± 0.6 degrees Celsius (mean ± 95% confidence interval) during the LGM. Our analysis includes four decades of groundwater noble gas data from six continents, along with new records from the tropics, all of which were interpreted using the same physical framework. Our land-based result broadly supports a recent reconstruction based on marine proxy data assimilation1 that suggested greater climate sensitivity than previous estimates5-7.

Abstract. Deglaciations are characterized by relatively fast and near-synchronous changes in ice sheet volume, ocean temperature, and atmospheric greenhouse gas concentrations, but glacial inceptions occur more gradually. Understanding the evolution of ice sheet, ocean, and atmospheric conditions from interglacial to glacial maximum provides important insight into the interplay of these components of our climate system. Using noble gas measurements in ancient ice samples, we reconstruct mean ocean temperature (MOT) from 74 to 59.5 ka BP, covering the Marine Isotope Stage (MIS) 5-4 boundary, MIS 4, and part of the MIS 4-3 transition. Comparing this MOT reconstruction to previously published MOT reconstructions from the last glacial cycle, we find that the majority of interglacial-glacial ocean cooling occurred across MIS 5, and MOT reached full glacial levels by MIS 4 (−2.7 ± 0.3 °C relative to the Holocene). Comparing MOT to contemporaneous records of CO2 and benthic ?18O, we find that ocean cooling and the solubility pump can explain most of the CO2 drawdown and increase in ?18O across MIS 5. The timing of ocean warming and cooling in our record indicates that millennial scale climate variability plays a crucial role in setting mean ocean temperature during this interval, as seen during other periods, such as the last deglaciation.

Abstract. Deglaciations are characterized by relatively fast andnear-synchronous changes in ice sheet volume, ocean temperature, andatmospheric greenhouse gas concentrations, but glacial inception occurs moregradually. Understanding the evolution of ice sheet, ocean, and atmosphereconditions from interglacial to glacial maximum provides insight into theinterplay of these components of the climate system. Using noble gasmeasurements in ancient ice samples, we reconstruct mean ocean temperature(MOT) from 74 to 59.7 ka, covering the Marine Isotope Stage (MIS) 5a–4boundary, MIS 4, and part of the MIS 4–3 transition. Comparing this MOTreconstruction to previously published MOT reconstructions from the last andpenultimate deglaciation, we find that the majority of the lastinterglacial–glacial ocean cooling must have occurred within MIS 5. MOTreached equally cold conditions in MIS 4 as in MIS 2 (−2.7 ± 0.3 ∘C relative to the Holocene, −0.1 ± 0.3 ∘Crelative to MIS 2). Using a carbon cycle model to quantify the CO2 solubility pump, we show that ocean cooling can explain most of theCO2 drawdown (32 ± 4 of 40 ppm) across MIS 5. Comparing MOT tocontemporaneous records of benthic δ18O, we find that ocean coolingcan also explain the majority of the δ18O increase across MIS 5 (0.7 ‰of 1.3 ‰). The timing of ocean warming and cooling inthe record and the comparison to coeval Antarctic isotope data suggest anintimate link between ocean heat content, Southern Hemisphere high-latitude climate,and ocean circulation on orbital and millennial timescales.

Abstract. The variations of δO2/N2 and δAr/N2 in the Dome Fuji ice core were measured from 112 m (bubbly ice) to 2001 m (clathrate hydrate ice). Our method, combined with the low storage temperature of the samples (−50 ∘C), successfully excludes post-coring gas-loss fractionation signals from our data. From the bubbly ice to the middle of the bubble–clathrate transition zone (BCTZ) (112–800 m) and below the BCTZ (>1200 m), the δO2/N2 and δAr/N2 data exhibit orbital-scale variations similar to local summer insolation. The data in the lower BCTZ (800–1200 m) have large scatter, which may be caused by millimeter-scale inhomogeneity of air composition combined with finite sample lengths. The insolation signal originally recorded at the bubble close-off remains through the BCTZ, and the insolation signal may be reconstructed by analyzing long ice samples (more than 50 cm for the Dome Fuji core). In the clathrate hydrate zone, the scatter around the orbital-scale variability decreases with depth, indicating diffusive smoothing of δO2/N2 and δAr/N2. A simple gas diffusion model was used to reproduce the smoothing and thus constrain their permeation coefficients. The relationship between δAr/N2 and δO2/N2 is markedly different for the datasets representing bubble close-off (slope ∼ 0.5), bubble–clathrate hydrate transformation (∼1), and post-coring gas loss (∼0.2), suggesting that the contributions of the mass-independent and mass-dependent fractionation processes are different for those cases. The method and data presented here may be useful for improving the orbital dating of deep ice cores over the multiple glacial cycles and further studying non-insolation-driven signals (e.g., atmospheric composition) of these gases.

Abstract. The atmospheric He/N2 ratio is expected to increase due to the emission of He associated with fossil fuels and isexpected to also vary in both space and time due to gravitational separationin the stratosphere. These signals may be useful indicators of fossil fuelexploitation and variability in stratospheric circulation, but directmeasurements of He/N2 ratio are lacking on all timescales. Here wepresent a high-precision custom inlet system for mass spectrometers thatcontinuously stabilizes the flow of gas during sample–standard comparisonand removes all non-noble gases from the gas stream. This enablesunprecedented accuracy in measurement of relative changes in the helium molefraction, which can be directly related to the 4He/N2 ratio usingsupplementary measurements of O2/N2, Ar/N2 and CO2.Repeat measurements of the same combination of high-pressure tanks using ourinlet system achieves a He/N2 reproducibility of∼ 10 per meg (i.e., 0.001 %) in 6–8 h analyses. This compares to interannual changesof gravitational enrichment at ∼ 35 km in the midlatitudestratosphere of order 300–400 per meg and an annual tropospheric increasefrom human fossil fuel activity of less than ∼ 30 per meg yr−1 (bounded by previous work on helium isotopes). The gettering andflow-stabilizing inlet may also be used for the analysis of other noble-gasisotopes and could resolve previously unobserved seasonal cycles inKr/N2 and Xe/N2.

Abstract. Air in polar ice cores provides unique information on past climatic and atmospheric changes. We developed a new method combining wet extraction, gaschromatography and mass spectrometry for high-precision, simultaneous measurements of eight air components (CH4, N2O andCO2 concentrations; δ15N, δ18O, δO2∕N2 and δAr∕N2; and total aircontent) from an ice-core sample of ∼ 60 g. The ice sample is evacuated for ∼ 2 h and melted under vacuum, and thereleased air is continuously transferred into a sample tube at 10 K within 10 min. The air is homogenized in the sample tubeovernight at room temperature and split into two aliquots for mass spectrometric and gas chromatographic measurements. Care is taken to minimize(1) contamination of greenhouse gases by using a long evacuation time, (2) consumption of oxygen during sample storage by a passivation treatment onsample tubes, and (3) fractionation of isotopic ratios with a long homogenization time for splitting. Precision is assessed by analyzing standardgases with artificial ice and duplicate measurements of the Dome Fuji and NEEM ice cores. The overall reproducibility (1 SD) of duplicate ice-coreanalyses are 3.2 ppb, 2.2 ppb and 2.9 ppm for CH4, N2O and CO2 concentrations;0.006 ‰, 0.011 ‰, 0.09 ‰ and 0.12 ‰ for δ15N, δ18O, δO2∕N2and δAr∕N2; and 0.63 mLSTP kg−1 for total air content, respectively. Our new method successfully combines thehigh-precision, small-sample and multiple-species measurements, with a wide range of applications for ice-core paleoenvironmental studies.

Abstract. A new ice core drilled at the South Pole provides a54 000-year paleoenvironmental record including the composition of the pastatmosphere. This paper describes the SP19 chronology for the South Poleatmospheric gas record and complements a previous paper (Winski et al., 2019)describing the SP19 ice chronology. The gas chronology is based on adiscrete methane (CH4) record with 20- to 190-year resolution. Toconstruct the gas timescale, abrupt changes in atmospheric CH4 duringthe glacial period and centennial CH4 variability during the Holocenewere used to synchronize the South Pole gas record with analogous data fromthe West Antarctic Ice Sheet Divide ice core. Stratigraphic matching basedon visual optimization was verified using an automated matching algorithm.The South Pole ice core recovers all expected changes in CH4 based onprevious records. Gas transport in the firn results in smoothing of theatmospheric gas record with a smoothing function spectral width that rangesfrom 30 to 78 years, equal to 3 % of the gas-age–ice-age difference, orΔage. The new gas chronology, in combination with the existing iceage scale from Winski et al. (2019), allows a model-independentreconstruction of the gas-age–ice-age difference through the whole record,which will be useful for testing firn densification models.

Abstract. Accurate simulation of atmospheric circulation,particularly in the lower stratosphere, is challenging due to unresolvedwave–mean flow interactions and limited high-resolution observations for validation. Gravity-induced pressure gradients lead to a small butmeasurable separation of heavy and light gases by molecular diffusion in thestratosphere. Because the relative abundance of Ar to N2 is exclusivelycontrolled by physical transport, the argon-to-nitrogen ratio (Ar∕N2)provides an additional constraint on circulation and the age of air (AoA),i.e., the time elapsed since entry of an air parcel into the stratosphere. Here we use airborne measurements of N2O and Ar∕N2 from ninecampaigns with global coverage spanning 2008–2018 to calculate AoA and toquantify gravitational separation in the lowermost stratosphere. To thisend, we develop a new N2O–AoA relationship using a Markov chain Monte Carlo algorithm. We observe that gravitational separation increases systematically with increasing AoA for samples with AoA between 0 and 3 years. These observations are compared to a simulation of the TOMCAT/SLIMCAT 3-D chemical transport model, which has been updated to includegravitational fractionation of gases. We demonstrate that although AoA atold ages is slightly underestimated in the model, the relationship betweenAr∕N2 and AoA is robust and agrees with the observations. Thishighlights the potential of Ar∕N2 to become a new AoA tracer that is subject only to physical transport phenomena and can supplement the suite of available AoA indicators.

The Younger Dryas (YD), arguably the most widely studied millennial-scale extreme climate event, was characterized by diverse hydroclimate shifts globally and severe cooling at high northern latitudes that abruptly punctuated the warming trend from the last glacial to the present interglacial. To date, a precise understanding of its trigger, propagation, and termination remains elusive. Here, we present speleothem oxygen-isotope data that, in concert with other proxy records, allow us to quantify the timing of the YD onset and termination at an unprecedented subcentennial temporal precision across the North Atlantic, Asian Monsoon-Westerlies, and South American Monsoon regions. Our analysis suggests that the onsets of YD in the North Atlantic (12,870 ± 30 B.P.) and the Asian Monsoon-Westerlies region are essentially synchronous within a few decades and lead the onset in Antarctica, implying a north-to-south climate signal propagation via both atmospheric (decadal-time scale) and oceanic (centennial-time scale) processes, similar to the Dansgaard-Oeschger events during the last glacial period. In contrast, the YD termination may have started first in Antarctica at ∼11,900 B.P., or perhaps even earlier in the western tropical Pacific, followed by the North Atlantic between ∼11,700 ± 40 and 11,610 ± 40 B.P. These observations suggest that the initial YD termination might have originated in the Southern Hemisphere and/or the tropical Pacific, indicating a Southern Hemisphere/tropics to North Atlantic-Asian Monsoon-Westerlies directionality of climatic recovery.

Abstract. In 2013 an ice core was recovered from Roosevelt Island, an ice dome between two submarine troughs carved by paleo-ice-streams in the Ross Sea, Antarctica. The ice core is part of the Roosevelt Island Climate Evolution (RICE) project and provides new information about the past configuration of the West Antarctic Ice Sheet (WAIS) and its retreat during the last deglaciation. In this work we present the RICE17 chronology, which establishes the depth–age relationship for the top 754 m of the 763 m core. RICE17 is a composite chronology combining annual layer interpretations for 0–343 m (Winstrup et al., 2019) with new estimates for gas and ice ages based on synchronization of CH4 and δ18Oatm records to corresponding records from the WAIS Divide ice core and by modeling of the gas age–ice age difference. Novel aspects of this work include the following: (1) an automated algorithm for multiproxy stratigraphic synchronization of high-resolution gas records; (2) synchronization using centennial-scale variations in methane for pre-anthropogenic time periods (60–720 m, 1971 CE to 30 ka), a strategy applicable for future ice cores; and (3) the observation of a continuous climate record back to ∼65 ka providing evidence that the Roosevelt Island Ice Dome was a constant feature throughout the last glacial period.

Abstract. Air in polar ice cores provides various information on past climatic and atmospheric changes. We developed a new method combining wet extraction, gas chromatography and mass spectrometry, for high-precision, simultaneous measurements of eight air components (CH4, N2O and CO2 concentrations, δ15N, δ18O, δO2/N2, δAr/N2 and total air content) from an ice core sample of ~60 g. The ice sample is evacuated for ~2 hours and melted under vacuum, and the released air is continuously transferred into a sample tube at 10 K within 10 minutes. The air is homogenized in the sample tube overnight at room temperature, and split into two aliquots for mass spectrometric and gas chromatographic measurements. Cares are taken to minimize contamination of greenhouse gases with long evacuation time, consumption of oxygen during sample storage by passivation treatment on sample tubes, and fractionation of isotopic ratios with long homogenization time for splitting. Precisions are assessed by analysing standard gases with artificial ice, and by duplicate measurements of the Dome Fuji and NEEM ice cores. The overall reproducibility (one standard deviation) from duplicate ice-core analyses are 3.2 ppb, 2.2 ppb and 3.1 ppm for CH4, N2O and CO2 concentrations, 0.006, 0.010, 0.09 and 0.12 ‰ for δ15N, δ18O, δO2/N2 and δAr/N2, and 0.67 mlSTP kg-1 for total air content, respectively. Our new method successfully combines the high-precision, small-sample and multiple-species measurements, with a wide range of applications for ice-core paleoenvironmental studies.

Abstract. A new ice core drilled at the South Pole provides a 54 000-year paleoenvironmental record including the composition of the past atmosphere. This paper describes the SP19 chronology for the South Pole atmospheric gas record and complements a previous paper (Winski et al., 2019) describing the SP19 ice chronology. The gas chronology is based on a discrete methane (CH4) record with 20- to 190-year resolution. To construct the gas time scale abrupt changes in atmospheric CH4 during the glacial period and centennial CH4 variability during the Holocene were used to synchronize the South Pole gas record with analogous data from the West Antarctic Ice Sheet Divide ice core. Stratigraphic matching based on visual optimization was verified using an automated matching algorithm. The South Pole ice core recovers all expected changes in CH4 based on previous records. Smoothing of the atmospheric record due to gas transport in the firn is evident but relatively minor, despite the deep lock-in depth in the modern South Pole firn column. The new gas chronology, in combination with the existing ice age scale from Winski et al. (2019), allows a model-independent reconstruction of the gas age-ice age difference through the whole record, which will be useful for testing firn densification models.

Gases preserved in ice cores provide a potential direct archive for atmospheric oxygen. Yet, oxygen-to-nitrogen ratios in ice cores (expressed as δO2/N2) are modified by a number of processes related to gas trapping and gas losses in the ice. Such complications have long hindered the use of ice core δO2/N2 to derive true atmospheric oxygen concentrations. Recently, a persistent decline in δO2/N2, observed in four different ice cores (GISP2, Vostok, Dome F, and EDC), is interpreted to reflect decreasing atmospheric O2 concentrations over the late Pleistocene (Stolper et al., 2016). The rate of δO2/N2 change is -8.4±0.2 ‰/Myr (1σ). Using new measurements made on EDC samples stored at -50 °C and therefore free from gas loss, Extier et al (2018) confirms the decrease in δO2/N2 with a slope of -7.0±0.6‰/Myr (1σ).Here, we present new δO2/N2 measurements made on 1.5-million-year-old blue ice cores from Allan Hills Blue Ice Areas, East Antarctica. We use argon-to-nitrogen ratios (δAr/N2) in the ice to correct for the fractionations during bubble close-off and gas losses. In those processes, δAr/N2 is fractionated in a fashion similar to δO2/N2 (Huber et al., 2006; Severinghaus and Battle, 2006). Paired δO2/N2-δAr/N2 values measured from the same sample were classified into three different time slices: 1.5 Ma (million years old), 950 ka, and 490 ka. Between 950 ka and 490 ka, we observe a decline in δO2/N2 similar to that observed in the aforementioned deep ice cores. This observation gives us confidence in the validity of the Allan Hills blue ice δO2/N2 records. Between 1.5 Ma and 950 ka, however, there is no statistically significant trend in ice core δO2/N2. Our results show a surprising lack of variability from 1.5 to 0.95 Ma; even during the past ~0.9 Ma, the rate of decline was very slow.

Abstract. Accurate simulation of atmospheric circulation, particularly in the lower stratosphere, is challenging due to unresolved wave-mean flow interactions and limited high-resolution observations for validation. Gravity-induced pressure gradients lead to a small but measurable separation of heavy and light gases by molecular diffusion in the stratosphere. Because the relative abundance of Ar to N2 is exclusively controlled by physical transport, the argon-to-nitrogen ratio (Ar/N2) provides an additional constraint on circulation and the age of air (AoA), i.e. the time elapsed since entry of an air parcel into the stratosphere. Here we use airborne measurements of N2O and Ar/N2 from nine campaigns with global coverage spanning 2008–2018 to calculate AoA and to quantify gravitational separation in the lowermost stratosphere. To this end, we develop a new N2O-AoA relationship using a Markov Chain Monte Carlo algorithm. We observe that gravitational separation increases systematically with increasing AoA for samples with AoA between 0 to 3 years. These observations are compared to a simulation of the TOMCAT/SLIMCAT 3-D chemical transport model, which has been updated to include gravitational fractionation of gases. We demonstrate that although AoA at old ages is slightly underestimated in the model, the relationship between Ar/N2 and AoA is robust and agrees with the observations. This highlights the potential of Ar/N2 to become a new AoA tracer that is subject only to physical transport phenomena and can supplement the suite of available AoA indicators.

Constraining the magnitude of past hydrological change may improve understanding and predictions of future shifts in water availability. Here we demonstrate that water-table depth, a sensitive indicator of hydroclimate, can be quantitatively reconstructed using Kr and Xe isotopes in groundwater. We present the first-ever measurements of these dissolved noble gas isotopes in groundwater at high precision (≤0.005‰ amu-1; 1σ), which reveal depth-proportional signals set by gravitational settling in soil air at the time of recharge. Analyses of California groundwater successfully reproduce modern groundwater levels and indicate a 17.9 ± 1.3 m (±1 SE) decline in water-table depth in Southern California during the last deglaciation. This hydroclimatic transition from the wetter glacial period to more arid Holocene accompanies a surface warming of 6.2 ± 0.6 °C (±1 SE). This new hydroclimate proxy builds upon an existing paleo-temperature application of noble gases and may identify regions prone to future hydrological change.

Over the past eight hundred thousand years, glacial-interglacial cycles oscillated with a period of one hundred thousand years ('100k world'1). Ice core and ocean sediment data have shown that atmospheric carbon dioxide, Antarctic temperature, deep ocean temperature, and global ice volume correlated strongly with each other in the 100k world2-6. Between about 2.8 and 1.2 million years ago, glacial cycles were smaller in magnitude and shorter in duration ('40k world'7). Proxy data from deep-sea sediments suggest that the variability of atmospheric carbon dioxide in the 40k world was also lower than in the 100k world8-10, but we do not have direct observations of atmospheric greenhouse gases from this period. Here we report the recovery of stratigraphically discontinuous ice more than two million years old from the Allan Hills Blue Ice Area, East Antarctica. Concentrations of carbon dioxide and methane in ice core samples older than two million years have been altered by respiration, but some younger samples are pristine. The recovered ice cores extend direct observations of atmospheric carbon dioxide, methane, and Antarctic temperature (based on the deuterium/hydrogen isotope ratio δDice, a proxy for regional temperature) into the 40k world. All climate properties before eight hundred thousand years ago fall within the envelope of observations from continuous deep Antarctic ice cores that characterize the 100k world. However, the lowest measured carbon dioxide and methane concentrations and Antarctic temperature in the 40k world are well above glacial values from the past eight hundred thousand years. Our results confirm that the amplitudes of glacial-interglacial variations in atmospheric greenhouse gases and Antarctic climate were reduced in the 40k world, and that the transition from the 40k to the 100k world was accompanied by a decline in minimum carbon dioxide concentrations during glacial maxima.

Abstract. We reconstruct atmospheric abundances of the potentgreenhouse gas c-C4F8 (perfluorocyclobutane, perfluorocarbonPFC-318) from measurements of in situ, archived, firn, and aircraft airsamples with precisions of ∼1 %–2 % reported on the SIO-14gravimetric calibration scale. Combined with inverse methods, we found near-zero atmospheric abundances from the early 1900s to the early 1960s, afterwhich they rose sharply, reaching 1.66 ppt (parts per trillion dry-air molefraction) in 2017. Global c-C4F8 emissions rose from near zero inthe 1960s to 1.2±0.1 (1σ) Gg yr−1 in the late 1970s tolate 1980s, then declined to 0.77±0.03 Gg yr−1 in the mid-1990sto early 2000s, followed by a rise since the early 2000s to 2.20±0.05 Gg yr−1 in 2017. These emissions are significantly larger thaninventory-based emission estimates. Estimated emissions from eastern Asiarose from 0.36 Gg yr−1 in 2010 to 0.73 Gg yr−1 in 2016 and 2017,31 % of global emissions, mostly from eastern China. We estimateemissions of 0.14 Gg yr−1 from northern and central India in 2016 andfind evidence for significant emissions from Russia. In contrast, recentemissions from northwestern Europe and Australia are estimated to be small(≤1 % each). We suggest that emissions from China, India, and Russiaare likely related to production of polytetrafluoroethylene (PTFE,“Teflon”) and other fluoropolymers and fluorochemicals that are based onthe pyrolysis of hydrochlorofluorocarbon HCFC-22 (CHClF2) in whichc-C4F8 is a known by-product. The semiconductor sector, wherec-C4F8 is used, is estimated to be a small source, at least inSouth Korea, Japan, Taiwan, and Europe. Without an obvious correlation withpopulation density, incineration of waste-containing fluoropolymers isprobably a minor source, and we find no evidence of emissions fromelectrolytic production of aluminum in Australia. While many possibleemissive uses of c-C4F8 are known and though we cannotcategorically exclude unknown sources, the start of significant emissionsmay well be related to the advent of commercial PTFE production in 1947.Process controls or abatement to reduce the c-C4F8 by-product wereprobably not in place in the early decades, explaining the increase inemissions in the 1960s and 1970s. With the advent of by-product reportingrequirements to the United Nations Framework Convention on Climate Change(UNFCCC) in the 1990s, concern about climate change and product stewardship,abatement, and perhaps the collection of c-C4F8 by-product for usein the semiconductor industry where it can be easily abated, it isconceivable that emissions in developed countries were stabilized and thenreduced, explaining the observed emission reduction in the 1980s and 1990s.Concurrently, production of PTFE in China began to increase rapidly. Withoutemission reduction requirements, it is plausible that global emissions todayare dominated by China and other developing countries. We predict thatc-C4F8 emissions will continue to rise and that c-C4F8will become the second most important emitted PFC in terms ofCO2-equivalent emissions within a year or two. The 2017 radiativeforcing of c-C4F8 (0.52 mW m−2) is small but emissions ofc-C4F8 and other PFCs, due to their very long atmosphericlifetimes, essentially permanently alter Earth's radiative budget and shouldbe reduced. Significant emissions inferred outside of the investigatedregions clearly show that observational capabilities and reportingrequirements need to be improved to understand global and country-scaleemissions of PFCs and other synthetic greenhouse gases and ozone-depletingsubstances.

The energy imbalance at the top of the atmosphere determines the temporal evolution of the global climate, and vice versa changes in the climate system can alter the planetary energy fluxes. This interplay is fundamental to our understanding of Earth's heat budget and the climate system. However, even today, the direct measurement of global radiative fluxes is difficult, such that most assessments are based on changes in the total energy content of the climate system. We apply the same approach to estimate the long-term evolution of Earth's radiative imbalance in the past. New measurements of noble gas-derived mean ocean temperature from the European Project for Ice Coring in Antarctica Dome C ice core covering the last 40,000 y, combined with recent results from the West Antarctic Ice Sheet Divide ice core and the sea-level record, allow us to quantitatively reconstruct the history of the climate system energy budget. The temporal derivative of this quantity must be equal to the planetary radiative imbalance. During the deglaciation, a positive imbalance of typically +0.2 W⋅m-2 is maintained for ∼10,000 y, however, with two distinct peaks that reach up to 0.4 W⋅m-2 during times of substantially reduced Atlantic Meridional Overturning Circulation. We conclude that these peaks are related to net changes in ocean heat uptake, likely due to rapid changes in North Atlantic deep-water formation and their impact on the global radiative balance, while changes in cloud coverage, albeit uncertain, may also factor into the picture.

New ice cores retrieved from the Taylor Glacier (Antarctica) blue ice area contain ice and air spanning the Marine Isotope Stage (MIS) 5-4 transition, a period of global cooling and ice sheet expansion. We determine chronologies for the ice and air bubbles in the new ice cores by visually matching variations in gas- and ice-phase tracers to preexisting ice core records. The chronologies reveal an ice age-gas age difference (Δage) approaching 10 ka during MIS 4, implying very low snow accumulation in the Taylor Glacier accumulation zone. A revised chronology for the analogous section of the Taylor Dome ice core (84 to 55 ka), located to the south of the Taylor Glacier accumulation zone, shows that Δage did not exceed 3 ka. The difference in Δage between the two records during MIS 4 is similar in magnitude but opposite in direction to what is observed at the Last Glacial Maximum. This relationship implies that a spatial gradient in snow accumulation existed across the Taylor Dome region during MIS 4 that was oriented in the opposite direction of the accumulation gradient during the Last Glacial Maximum.

The South Pole Ice Core (SPICEcore) was drilled in 2014-2016 to provide a detailed multi-proxy archive of paleoclimate conditions in East Antarctica during the Holocene and late Pleistocene. Interpretation of these records requires an accurate depth-age relationship. Here, we present the SPICEcore (SP19) timescale for the age of the ice of SPICEcore. SP19 is synchronized to the WD2014 chronology from the West Antarctic Ice Sheet Divide (WAIS Divide) ice core using stratigraphic matching of 251 volcanic events. These events indicate an age of 54 302±519 BP (years before 1950) at the bottom of SPICEcore. Annual layers identified in sodium and magnesium ions to 11 341 BP were used to interpolate between stratigraphic volcanic tie points, yielding an annually resolved chronology through the Holocene. Estimated timescale uncertainty during the Holocene is less than 18 years relative to WD2014, with the exception of the interval between 1800 to 3100 BP when uncertainty estimates reach ±25 years due to widely spaced volcanic tie points. Prior to the Holocene, uncertainties remain within 124 years relative to WD2014. Results show an average Holocene accumulation rate of 7.4 cm yr-1 (water equivalent). The time variability of accumulation rate is consistent with expectations for steady-state ice flow through the modern spatial pattern of accumulation rate. Time variations in nitrate concentration, nitrate seasonal amplitude and δ15N of N2 in turn are as expected for the accumulation rate variations. The highly variable yet well-constrained Holocene accumulation history at the site can help improve scientific understanding of deposition-sensitive climate proxies such as δ15N of N2 and photolyzed chemical compounds.

Abstract. In 2013, an ice core was recovered from Roosevelt Island in the Ross Sea, Antarctica, as part of the Roosevelt Island Climate Evolution (RICE) project. Roosevelt Island is located between two submarine troughs carved by paleo-ice-streams. The RICE ice core provides new important information about the past configuration of the West Antarctic Ice Sheet and its retreat during the most recent deglaciation. In this work, we present the RICE17 chronology and discuss preliminary observations from the new records of methane, the isotopic composition of atmospheric molecular oxygen (δ18O-Oatm), the isotopic composition of atmospheric molecular nitrogen (δ15N-N2) and total air content (TAC). RICE17 is a composite chronology combining annual layer interpretations, gas synchronization, and firn modeling strategies in different sections of the core. An automated matching algorithm is developed for synchronizing the high-resolution section of the RICE gas records (60–720 m, 1971 CE to 30 ka) to corresponding records from the WAIS Divide ice core, while deeper sections are manually matched. Ice age for the top 343 m (2635 yr BP, before 1950 C.E.) is derived from annual layer interpretations and described in the accompanying paper by Winstrup et al. (2017). For deeper sections, the RICE17 ice age scale is based on the gas age constraints and the ice age-gas age offset estimated by a firn densification model. Novel aspects of this work include: 1) stratigraphic matching of centennial-scale variations in methane for pre-anthropogenic time periods, a strategy which will be applicable for developing precise chronologies for future ice cores, 2) the observation of centennial-scale variability in methane throughout the Holocene which suggests that similar variations during the late preindustrial period need not be anthropogenic, and 3) the observation of continuous climate records dating back to ∼ 65 ka which provide evidence that the Roosevelt Island Ice Dome was a constant feature throughout the last glacial period.

The global ocean constitutes the largest heat buffer in the global climate system, but little is known about its past changes. The isotopic and elemental ratios of heavy noble gases (krypton and xenon), together with argon and nitrogen in trapped air from ice cores, can be used to reconstruct past mean ocean temperatures (MOTs). Here we introduce two successively developed methods to measure these parameters with a sufficient precision to provide new constraints on past changes in MOT.The air from an 800-g ice sample - containing roughly 80 mL STP air - is extracted and processed to be analyzed on two independent dual-inlet isotope ratio mass spectrometers. The primary isotope ratios (δ15 N, δ40 Ar and δ86 Kr values) are obtained with precisions in the range of 1 per meg (0.001‰) per mass unit. The three elemental ratio values δKr/N2 , δXe/N2 and δXe/Kr are obtained using sequential (non-simultaneous) peak-jumping, reaching precisions in the range of 0.1-0.3‰.The latest version of the method achieves a 30% to 50% better precision on the elemental ratios and a twofold better sample throughput than the previous one. The method development uncovered an unexpected source of artefactual gas fractionation in a closed system that is caused by adiabatic cooling and warming of gases (termed adiabatic fractionation) - a potential source of measurement artifacts in other methods.The precisions of the three elemental ratios δKr/N2 , δXe/N2 and δXe/Kr - which all contain the same MOT information - suggest smaller uncertainties for reconstructed MOTs (±0.3-0.1°C) than previous studies have attained. Due to different sensitivities of the noble gases to changes in MOT, δXe/N2 provides the best constraints on the MOT under the given precisions followed by δXe/Kr, and δKr/N2 ; however, using all of them helps to detect methodological artifacts and issues with ice quality.

Abstract. A new ice core retrieved from the Taylor Glacier blue ice area contains ice and air spanning the Marine Isotope Stage (MIS) 5/4 transition (74 to 65 ka), a period of global cooling and glacial inception. Dating the ice and air bubbles in the new ice core reveals an ice age-gas age difference (Δage) approaching 10 ka during MIS 4, implying very low accumulation at the Taylor Glacier accumulation zone on the northern flank of Taylor Dome. A revised chronology for the Taylor Dome ice core (80 to 55 ka), situated to the south of the Taylor Glacier accumulation zone, shows that Δage did not exceed 2.5 ka at that location. The difference in Δage between the new Taylor Glacier ice core and the Taylor Dome ice core implies a spatial gradient in snow accumulation across Taylor Dome that intensified during the last glacial inception and through MIS 4.

Abstract. High-resolution, well-dated climate archives provide anopportunity to investigate the dynamic interactions of climate patternsrelevant for future projections. Here, we present data from a new, annuallydated ice core record from the eastern Ross Sea, named the Roosevelt IslandClimate Evolution (RICE) ice core. Comparison of this record with climatereanalysis data for the 1979–2012 interval shows that RICE reliably capturestemperature and snow precipitation variability in the region. Trends over thepast 2700 years in RICE are shown to be distinct from those in WestAntarctica and the western Ross Sea captured by other ice cores. For most ofthis interval, the eastern Ross Sea was warming (or showing isotopicenrichment for other reasons), with increased snow accumulation and perhapsdecreased sea ice concentration. However, West Antarctica cooled and thewestern Ross Sea showed no significant isotope temperature trend. Thispattern here is referred to as the Ross Sea Dipole. Notably, during theLittle Ice Age, West Antarctica and the western Ross Sea experienced colderthan average temperatures, while the eastern Ross Sea underwent a period ofwarming or increased isotopic enrichment. From the 17th century onwards, thisdipole relationship changed. All three regions show current warming, withsnow accumulation declining in West Antarctica and the eastern Ross Sea butincreasing in the western Ross Sea. We interpret this pattern as reflectingan increase in sea ice in the eastern Ross Sea with perhaps the establishmentof a modern Roosevelt Island polynya as a local moisture source for RICE.

Abstract. Ancient air trapped in ice core bubbles has been paramount to developing our understanding of past climate and atmospheric composition. Before air bubbles become isolated in ice, the atmospheric signal is altered in the firn column by transport processes such as advection and diffusion. However, the influence of low-permeability layers and barometric pumping (driven by surface pressure variability) on firn air transport is not well understood and is not readily captured in conventional one-dimensional (1-D) firn air models. Here we present a two-dimensional (2-D) trace gas advection–diffusion–dispersion model that accounts for discontinuous horizontal layers of reduced permeability. We find that layering or barometric pumping individually yields too small a reduction in gravitational settling to match observations. In contrast, when both effects are active, the model's gravitational fractionation is suppressed as observed. Layering focuses airflows in certain regions in the 2-D model, which acts to amplify the dispersive mixing resulting from barometric pumping. Hence, the representation of both factors is needed to obtain a realistic emergence of the lock-in zone. In contrast to expectations, we find that the addition of barometric pumping in the layered 2-D model does not substantially change the differential kinetic fractionation of fast- and slow-diffusing trace gases. Like 1-D models, the 2-D model substantially underestimates the amount of differential kinetic fractionation seen in actual observations, suggesting that further subgrid-scale processes may be missing in the current generation of firn air transport models. However, we find robust scaling relationships between kinetic isotope fractionation of different noble gas isotope and elemental ratios. These relationships may be used to correct for kinetic fractionation in future high-precision ice core studies and can amount to a bias of up to 0.45 °C in noble-gas-based mean ocean temperature reconstructions at WAIS Divide, Antarctica.

Abstract. Ancient air trapped in ice core bubbles has been paramount to developing our understanding of past climate and atmospheric composition. Before air bubbles become isolated in ice, the atmospheric signal is altered in the firn column by transport processes such as advection and diffusion. However, the influence of impermeable layers and barometric pumping (driven by surface pressure variability) on firn air transport is not well understood and cannot be captured in conventional 1-dimensional firn air models. Here we present a 2-dimensional (2D) trace gas advection-diffusion-dispersion model that accounts for discontinuous horizontal layers of reduced permeability. We find that layering and barometric pumping individually yield too small a reduction in gravitational settling to match observations. In contrast, a combination of both effects more strongly suppresses gravitational fractionation. Layering locally focuses airflows in the 2D model and thus amplifies the dispersive mixing resulting from barometric pumping. Hence, the representation of both factors is needed to obtain a more natural emergence of the lock-in zone. Moreover, we find that barometric pumping in the layered 2D model does not substantially change the differential kinetic fractionation of fast and slow diffusing trace gases, which is observed in nature. This suggests that further subgrid-scale physics may be missing in the current generation of firn air transport models. However, we find robust scaling relationships between kinetic isotope fractionation of different noble gas isotope and elemental ratios. These relationships may be used to correct for kinetic fractionation in future high precision ice core studies.

Glacial-state greenhouse gas concentrations and Southern Hemisphere climate conditions persisted until ∼17.7 ka, when a nearly synchronous acceleration in deglaciation was recorded in paleoclimate proxies in large parts of the Southern Hemisphere, with many changes ascribed to a sudden poleward shift in the Southern Hemisphere westerlies and subsequent climate impacts. We used high-resolution chemical measurements in the West Antarctic Ice Sheet Divide, Byrd, and other ice cores to document a unique, ∼192-y series of halogen-rich volcanic eruptions exactly at the start of accelerated deglaciation, with tephra identifying the nearby Mount Takahe volcano as the source. Extensive fallout from these massive eruptions has been found >2,800 km from Mount Takahe. Sulfur isotope anomalies and marked decreases in ice core bromine consistent with increased surface UV radiation indicate that the eruptions led to stratospheric ozone depletion. Rather than a highly improbable coincidence, circulation and climate changes extending from the Antarctic Peninsula to the subtropics-similar to those associated with modern stratospheric ozone depletion over Antarctica-plausibly link the Mount Takahe eruptions to the onset of accelerated Southern Hemisphere deglaciation ∼17.7 ka.