Atmospheric composition 1 million years ago from blue ice in the Allan Hills, Antarctica
Edited by Mark H. Thiemens, University of California at San Diego, La Jolla, CA, and approved April 16, 2015 (received for review December 1, 2014)
Significance
Bubbles of ancient air trapped in ice cores permit the direct reconstruction of atmospheric composition and allow us to link greenhouse gases and global climate over the last 800 ky. Here, we present new ice core records of atmospheric composition roughly 1 Ma from a shallow ice core drilled in the Allan Hills blue ice area, Antarctica. These records confirm that interglacial CO2 concentrations decreased by 800 ka. They also show that the link between CO2 and Antarctic temperature extended into the warmer world of the mid-Pleistocene.
Abstract
Here, we present direct measurements of atmospheric composition and Antarctic climate from the mid-Pleistocene (∼1 Ma) from ice cores drilled in the Allan Hills blue ice area, Antarctica. The 1-Ma ice is dated from the deficit in 40Ar relative to the modern atmosphere and is present as a stratigraphically disturbed 12-m section at the base of a 126-m ice core. The 1-Ma ice appears to represent most of the amplitude of contemporaneous climate cycles and CO2 and CH4 concentrations in the ice range from 221 to 277 ppm and 411 to 569 parts per billion (ppb), respectively. These concentrations, together with measured δD of the ice, are at the warm end of the field for glacial–interglacial cycles of the last 800 ky and span only about one-half of the range. The highest CO2 values in the 1-Ma ice fall within the range of interglacial values of the last 400 ka but are up to 7 ppm higher than any interglacial values between 450 and 800 ka. The lowest CO2 values are 30 ppm higher than during any glacial period between 450 and 800 ka. This study shows that the coupling of Antarctic temperature and atmospheric CO2 extended into the mid-Pleistocene and demonstrates the feasibility of discontinuously extending the current ice core record beyond 800 ka by shallow coring in Antarctic blue ice areas.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
Ice cores serve as a critical archive of past environmental conditions, providing constraints on global atmospheric composition and the climate of polar regions (1). Reconstructions of atmospheric CO2 and CH4 from air trapped in ice cores dating as far back as 800 ka indicate a link between greenhouse gases and global climate in the form of 100-ky glacial cycles (Fig. 1). These climate cycles are recorded in proxy records from deep sea sediments reflecting variations in ocean temperature and continental ice volume (2). Deep sea records indicate that the 100-ky glacial cycle developed only ∼900,000 y ago [the mid-Pleistocene transition (MPT)] (3). Before this time and going back to 2.8 Ma, glacial cycles lasted, on average, 40 ky (4). The origins of both the 100- and 40-ky glacial cycles, their links to orbital forcing, and changes in atmospheric greenhouse gases are debated. Extending ice core records to earlier times would advance our understanding of links between greenhouse gases, climate, and causes of the MPT.
Fig. 1.
One archive for extending ice core records beyond 800 ky is blue ice areas (BIAs), outcrops of glacial ice brought to the surface by ice flow guided by bedrock topography (5). These records are likely to be stratigraphically complex because of deformation associated with ice transport but may also contain the oldest easily accessible ice on the planet. In the Allan Hills, Antarctica, the antiquity of shallow ice is documented by terrestrial ages of englacial meteorites exposed on the surface by ablation (6). These ages cluster between 100 and 400 ky, with a small number that extend to 1 Ma and a single meteorite yielding an age of 2.2 Ma (7).
Here, we present the first, to our knowledge, direct snapshots of atmospheric composition during the MPT from an ice core drilled at Site BIT-58 (8) in the Allan Hills BIA (Fig. S1). We date ice and trapped gases directly, taking advantage of the slow leak of 40Ar into the atmosphere from the decay of 40K in Earth’s interior (the 40Aratm geochronometer). The observed increase in the 40Ar/38Ar ratio of the atmosphere is small, resulting in uncertainties for a single sample of ±213 ky (9). Measurements of Ar isotope ratios date a 12-m section at the base of Site BIT-58 to ∼1 Ma (Fig. 2). We report on measurements of CO2 and CH4 concentrations, the δ18O of paleoatmospheric O2, δD, and deuterium excess (d) in the 1-Ma ice from Site BIT-58. This work provides a direct window into atmospheric composition and Antarctic climate during the mid-Pleistocene.
Fig. 2.
Climate Records from Ice Cores in the Allan Hills BIA
Previous research using surface samples and vertical ice cores from the Allan Hills Main Ice Field (MIF) has shown that shallow coring (<200 m) can yield high-quality records of Antarctic climate and atmospheric composition that are continuous over long time intervals (10). In particular, results from surface samples and cores drilled along the MIF flow line reveal a continuous climate record spanning marine isotope stage (MIS) 5/6 (Fig. 1). It is exposed both at the surface along a 5-km transect and in the long core (225 m) at Site 27 (10). The ice was dated to 85–250 ka by combining 40Aratm with the stratigraphy of both δD of the ice [isotopic temperature (11)] and δ18O of paleoatmospheric O2 [δ18Oatm (12)]. Measurements of CO2 and CH4 at Site 27 are also consistent with records from other Antarctic ice cores for this time period (Fig. S2). The climate record from Site 27 across MIS 5/6 provides an important reference frame for comparison with ice of greater antiquity found in the Allan Hills BIA, because environmental conditions in the source regions are expected to be similar.
Site BIT-58 sits near the crest of the northwest to southeast trending ice ridge off the MIF in an area of Allan Hills where terrestrial meteorite ages suggest ice older than 1 Ma (6, 7). Although directions and magnitudes of ice flow at Site BIT-58 are not known, measurements from the nearby MIF are consistent with flow to the east or northeast. Bedrock topography, determined using ground penetrating radar, indicates that Site BIT-58 sits on a local bedrock high with a total ice thickness of ∼130 m. Although drilling came to within 5 m of bedrock, recovered ice was clean, and there was no evidence for contamination from bedrock from visual inspection or the chemistry of trapped gases (δO2/N2 and air content) (SI Text).
Measured 40Aratm ages from the 126-m core at Site BIT-58 are shown in Fig. 2. The core can be separated into two distinct sections. An upper unit extends from 25 to 113 m, with a weighted mean 40Aratm age of 320 ± 160 ky (1σ; n = 19) (Table S1). Below is a basal unit from 113 m to at least 126 m (the bottom of the core), with an average 40Aratm age of 990 ± 110 ky (1σ; n = 6). The transition between 320-ka and 1-Ma ice occurs between Ar-dated samples at 112.8 and 117.4 m. The shift in age occurs across an interval from 113- to 115-m depth that is also associated with relatively large changes in the δD of the ice and δ18O of paleoatmospheric O2 (Fig. 2 and Table S2). Within the 1-Ma ice, the 40Aratm ages from different depths are indistinguishable. As a result, although the ice below 117 m at Site BIT-58 likely reflects a range of ages around 1 Ma, the range is less than our analytical uncertainty (±213 ky; 1σ).
Two independent lines of evidence support an 40Aratm age >800 ky for the ice below 117.4 m at Site BIT-58. First, 3 of 15 CO2 measurements yield concentrations that are higher than any measured from European Project for Ice Coring in Antarctica (EPICA) Dome C between 450 and 800 ky (Fig. 3B and Table S3). Second, a subset of the paired measurements of CO2 and CH4 shows no overlap with known atmospheric values between 450 and 800 ky (Fig. 3B).
Fig. 3.
In addition, water isotopes show that there is an abrupt transition in ice age at ∼113 m. Records of deuterium excess (d) in the Site BIT-58 ice—a measure of temperature, relative humidity, seasonality of precipitation, and wind speed in the precipitation source region (13)—indicate two distinct populations above and below 113-m depth (Fig. S3 and Table S4). Measured d values for both populations range from +0.9‰ to −10.7‰. Parenthetically, these values are extremely depleted relative to those observed in the Antarctic interior [e.g., d ∼ +4‰ to +12‰ at EPICA Dome C (14)], consistent with a local source of precipitation for the 320-ka and 1-Ma ice in the Allan Hills BIA (13).
The variability in δD and δ18Oatm observed in the 320-ka unit at Site BIT-58 is comparable in magnitude with the variability associated with MIS 5/6 at nearby Site 27 (10). However, the abrupt transitions in δD and δ18Oatm suggest that the ice within the 320-ka unit at Site BIT-58 is stratigraphically disturbed. δDice varies on submeter scales within the 1-Ma ice (i.e., between 118 and 120 m) (Fig. 2), indicating that this unit is stratigraphically complex as well (Fig. 2 and Fig. S4). Stratigraphic disturbance associated with ice transport is not surprising in the Allan Hills BIA given the subglacial topography of the region, which includes large and abrupt changes in ice thickness [from >1 km to <200 m over a few kilometers (15)], and the distance over which the ice has been transported from its source region [∼20 km at present (16)]. As a result, the Allan Hills BIA is unlikely to preserve long (>100 ky) continuous records of ice >800 ky in age. Records may also be difficult to interpret, because layers can be overturned or oriented parallel to the drilling direction. However, deformation and folding also confer a benefit by increasing the likelihood that a given volume of ice samples a wider range of time periods and climate states. In addition, the shallow burial depths and cold temperatures that characterize ice core records from the Allan Hills BIA should minimize diffusive exchange of gases and the loss of paleoclimatic information in ice older than 1 My (17).
Given the evidence for stratigraphic disturbance, records of ice and gas chemistry in the 1-Ma unit at Site BIT-58 represent snapshots or short intervals of mid-Pleistocene climate that span some fraction of the full range of glacial–interglacial variability. Despite these limitations, our data represent the oldest direct measurements of atmospheric CO2, CH4, δ18O of O2, and Antarctic δDice and permit important preliminary conclusions to be drawn about atmospheric composition, Antarctic climate, and glacial cycles during the mid-Pleistocene.
Snapshots of Atmospheric Composition and Antarctic Climate at 1 Ma
Given the disturbed nature of the 1-Ma ice from Site BIT-58, we treat each of our data points as a snapshot of mid-Pleistocene climate. We then compare the field of values in the 1-Ma ice with those of younger ice in stratigraphically continuous cores at Allan Hills Site 27, Vostok, and EPICA Dome C. Exactly how many unique views exist in the 1-Ma ice is uncertain. Cross-plots of the data do not show correlations that would reflect simple mixing of two end members. To estimate the fraction of glacial–interglacial variability captured in our 1-Ma ice, we compare the range of variability in the 1-Ma ice and the MIS 5/6 ice from Site 27 with the range of variability in marine climate proxies [benthic foraminfera δ18O stack (4) and Mg/Ca records (18); Fig. 1] over the same time periods.
Considering only the benthic δ18O stack, the amplitudes of glacial cycles between MISs 19 and 38 average ∼1.1‰ or ∼60% of the amplitude of MIS 5/6 (1.9‰). In comparison, the range in δD values that we observe in the 1-Ma ice is 46–51% of the range for MIS 5/6 at Site 27, suggesting that we have recovered 80–88% of the full mid-Pleistocene glacial–interglacial variability. Taking into account the effects of changes in the partitioning of the δ18O signal between deep ocean temperature and ice volume using Mg/Ca-based temperature reconstructions (18) lowers this estimate to ∼75% of the full glacial–interglacial range. Therefore, the 1-Ma ice at Site BIT-58 represents most (75–88%) but probably not all of the glacial–interglacial climate range during the MPT. The missing ice most likely comes from glacial maxima, because these intervals are generally underrepresented in ice cores due to lower accumulation rates.
The ranges of measured CO2 and CH4 concentrations in the 1-Ma ice are also reduced compared with glacial cycles over the last 800 ky (Fig. 1). Measured CO2 concentrations range from 221 to 277 ppm or 45–50% of the average glacial cycle since 800 ka (1, 19, 20). Measured CH4 concentrations range from 411 to 569 parts per billion or 35–40% of the average glacial cycle since 800 ka (21). The reduction in the range of measured CO2 concentrations in the 1-Ma ice is comparable with that observed for δDice between the 1-Ma ice and MIS 5/6 at Site 27, whereas the reduction in the range of measured CH4 concentrations in the 1-Ma ice is slightly larger. Three of twelve measured CO2 concentrations in the 1-Ma ice are higher than any CO2 concentrations between 450 and 800 ka (Fig. 3B). Minimum CO2 concentrations are also ∼30 ppm higher than any glacial maxima values since 800 ka. Cross-plots of CO2 and δD for the 1-Ma ice indicate a relationship between East Antarctic temperature change and CO2 that is similar to that for MIS 5/6 at Site 27 and other Antarctic sites (Fig. 3A). Cross-plots of CH4 and CO2 show that 1-Ma samples fall within the envelope of glacial–interglacial variability over the last 800 ky. However, for a given CO2 concentration, CH4 concentrations in the 1-Ma ice populate the lower end of the range (Fig. 3B).
Implications for Glacial Cycles During the MPT
Our results support and reveal a number of features of mid-Pleistocene atmospheric composition and Antarctic climate. First, previous reconstructions of MPT CO2 levels from measurements of B isotopes in planktic foraminifera have shown that interglacial CO2 was similar to present values and that glacial CO2 was 30 ppm higher than 0- to 800-ka glacial values during the MPT (22). In agreement with these reconstructions, Site BIT-58 values indicate that maximum CO2 concentrations were slightly higher than interglacial values between 450 and 800 ka and comparable with values between 0 and 450 ka. Our records are also consistent with higher CO2 concentrations at glacial maxima during the MPT. However, the Site BIT-58 record is probably incomplete, and our minimum measured value of 221 ppm is an upper bound.
Second, all of the climate indicators—CO2, CH4, and δDice—exhibit smaller amplitude variability in the 1-Ma ice than in glacial cycles from 0 to 800 ka. Nevertheless, the relationships between climate properties are indistinguishable from those observed for glacial cycles over the last 800 ky. Some of the reduced variability in the 1-Ma ice may come from the fact that the record likely captures most but not all of the glacial–interglacial range during the MPT. Mean δDice and CO2 concentrations in the 1-Ma ice are also shifted toward higher (warmer) values, consistent with warmer climates during the MPT.
Third, measured CH4 concentrations in the 1-Ma ice are low relative to coeval CO2 concentrations in the context of the 800-ka ice core record (Fig. 1). The CH4 concentration range in the 1-Ma ice also accounts for a smaller fraction of the 0- to 800-ka range than any other climate property. One possible explanation is that variations in atmospheric CH4 during the MPT lacked the overshoots associated with rapid Northern Hemisphere warming events (23). Although ice rafting in the North Atlantic predated the MPT (24), millennial-scale climate variability was much stronger after the MPT (25). In the absence of these millennial events during the MPT, high CH4 concentrations would not be expected to co-occur with cooler Antarctic temperatures and lower CO2 concentrations.
Conclusions
Ice cores containing archives of 1-My air have been recovered from shallow depths in Antarctic BIAs, providing the first, to our knowledge, direct measurements of atmospheric composition during glacial cycles of the mid-Pleistocene. Although these records are stratigraphically complex and likely incomplete, our results are consistent with the prevailing views (4, 22) that the amplitude of glacial cycles was diminished during this time, climates were warmer, interglacial CO2 concentrations were greater than between 450 and 800 ka, and glacial CO2 concentrations, at most, were ∼30 ppm higher than between 0 and 800 ka. Our record also provides reconstructions of atmospheric CH4 >800 ka and shows that CH4 concentrations are both surprisingly low and exhibit less variability than the other climate indicators compared with glacial cycles over the last 800 ky.
Materials and Methods
40Aratm Geochronometer.
In the solid earth, 40K decays to stable 40Ar. Because 40Ar slowly leaks into the atmosphere, its concentration increases with time. In contrast, because 38Ar and 36Ar are stable, primordial, and nonradiogenic, their atmospheric concentrations are constant. Thus, the 40Ar/38Ar ratio of the atmosphere rises with time (toward the future) and decreases with age (toward the past), providing a tool for dating. The term of merit is the paleoatmospheric 40Ar/38Ar ratio, which is defined as 40Aratm = δ40Ar/38Ar − 1.002δ38Ar/36Ar.
The latter term corrects for gravitational fractionation, so that δ40Aratm is a measure of the paleoatmospheric 40Ar/38Ar ratio. Studies of 40Aratm as a function of age in the Dome C and Vostok cores characterize its rate of change over the last 800 ka, enabling its use for dating ice (9). We linearly extrapolate the rate of change over the last 800 ka to older periods (i.e., 1–2 Ma), although even for ice of this age, the signal is small and uncertainties are large for a single measurement at 213–246 ky (Table S1). Reported errors (1σ) for each sample reflect both analytical errors and uncertainties in the calibration slope of the 40Aratm geochronometer.
Procedures for Ar analyses are a modification of the methods used in ref. 9. Trapped air was wet-extracted from an ∼500-g ice core sample. Two getters were used to sequentially purify the Ar by removing more than 99.9999% of N2, O2, and other nonnoble gases. Samples were then admitted to a Finnigan MAT 252 Mass Spectrometer, which simultaneously measures δ40Ar/38Ar and δ38Ar/36Ar. The SD of a single measurement of δ40Aratm in a sample of local air (Princeton Air) is ±0.0143‰ or ±213 ka. Reproducibility of natural ice core samples from repeated measurements of Holocene age ice from Antarctica from this study and the study in ref. 9 is −0.005 ± 0.0100‰ (n = 16), corresponding to an age uncertainty of ±149 ka. The reason for the difference is unclear at present, and as a result, we adopt the more conservative estimate of uncertainty (replicate measurements of Princeton Air or ±213 ka; 1σ) for this study. When possible, samples were measured in replicate from the same depth. Each ice sample was measured in conjunction with an aliquot of an in-house Ar standard and Princeton Air that had been processed through the same procedures as the ice core samples, with the exception of wet extraction. Each sample was analyzed for ∼3 h on the mass spectrometer. This long analysis period permits the precise measurement of δ40Aratm, despite the very low natural abundance of 38Ar. Measured 38/36Ar ratios in the ice samples deviate from the air standard because of gravitational fractionation in the firn, with the magnitude of the gravitational fractionation in 38/36Ar being roughly two times that expected for 15/14N.
δ18Oatm and δ15N.
Analyses for the O2/N2/Ar ratio, δ18O of O2, and δ15N of N2 were carried out as described in ref. 26. Briefly, ice was melted in vacuum, the dried headspace gases were collected by condensation at liquid helium temperatures, and the samples were admitted to the mass spectrometer (Thermo Finnegan Delta Plus XP) for elemental and isotope ratio analyses. External reproducibility is typically about ±0.02‰ for δ15N and ±0.03 ‰ for δ18Oatm. Samples with poor reproducibility in δ15N (>0.1‰) were omitted (n = 2).
CH4, Air Content, and CO2.
CH4 was analyzed using a melt–refreeze technique most recently described in ref. 27. Samples (∼60–70 g ice) were trimmed, melted under vacuum, and then, refrozen at about −70 °C. Methane concentrations in released air were measured using a gas chromatograph and referenced to air standards calibrated by National Oceanic and Atmospheric Administration (NOAA) Global Monitoring Division (GMD) on the NOAA04 scale. Precision was generally better than ±4 parts per billion. Measured CH4 concentrations were not corrected for gravitational fractionation, which had an effect that is minor given the small gravitational fractionation observed in δ15N at Site BIT-58. Individual sample uncertainties are reported in Table S3. The measurement also quantifies the total air content of the sample, an important parameter for evaluating sample quality and possibly, elevation of the deposition site.
CO2 concentrations were measured using the dry extraction (crushing) method described in ref. 28; 8- to 15-g samples were crushed under vacuum, and the sample air was condensed in steel tubes at 11 K. CO2 concentrations were measured after equilibration to room temperature using gas chromatography and referenced to air standards calibrated by NOAA GMD on the World Meteorological Organization (WMO) scale. Generally, several replicate ice samples were analyzed for each depth, and results were averaged to obtain final CO2 concentration. Typical SEMs are <1 ppm for four to six replicates. Measured CO2 concentrations were also not corrected for gravitational fractionation. Individual sample uncertainties are reported in Table S3.
δDice and δ18Oice.
When possible, samples for Site BIT-58 were subsampled from cut slabs at 15-cm resolution in a −20 °C working freezer at the Climate Change Institute, University of Maine. Because some portions of the core were heavily fractured, continuous sections were not available at all depths. For δDice and δ18Oice measurements, ice samples were melted to liquid H2O at room temperature, and 1.6-mL aliquots were transferred to glass vials for analysis. Samples and laboratory standards, which span a wide range of naturally occurring isotopic values previously calibrated against standard mean ocean water, standard light Antarctic precipitation, and Greenland ice sheet precipitation, were measured by cavity ring-down spectroscopy using a Picarro Model L2130-i Ultra High-Precision Isotopic Water Analyzer coupled with a High-Precision Vaporizer and liquid autosampler module; δ18Oice and δDice were measured simultaneously with an internal precision at 2σ = ±0.05‰ for δ18O and ±0.10‰ for δDice.
Acknowledgments
We would like to acknowledge US Ice Drilling Design and Operations (IDDO), driller Mike Waszkiewicz, Kristin Schild, Melissa Rohde, and Ken Borek Air for assistance with the field work. Mike Kalk and Haun Marcott assisted with the CO2 measurements. This work was funded by National Science Foundation Grants ANT-0838843 (University of Maine) and ANT-0838849 (Princeton University).
Supporting Information
Supporting Information (PDF)
Supporting Information
- Download
- 617.50 KB
References
1
JR Petit, et al., Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 399, 429–436 (1999).
2
NJ Shackleton, et al., Oxygen isotope calibration of the onset of ice-rafting and history of glaciation in the North Atlantic region. Nature 307, 620–623 (1984).
3
PU Clark, et al., The middle Pleistocene transition: Characteristics, mechanisms, and implications for long-term changes in atmospheric pCO2. Quat Sci Rev 25, 3150–3184 (2006).
4
L Lisiecki, ME Raymo, A Pliocene-Pleistocene stack of 57 globally distributed benthic d18O records. Paleoceanography 20, PA1003 (2005).
5
R Bintanja, On the glaciological, meteorological, and climatological significance of Antarctic blue ice areas. Rev Geophys 37, 337–359 (1999).
6
K Nishiizumi, D Elmore, PW Kubik, Update on terrestrial ages of Antarctic meteorites. Earth Planet Sci Lett 93, 299–313 (1989).
7
P Scherer, et al., Allan Hills 88019: An Antarctic H-chondrite with a very long terrestrial age. Meteorit Planet Sci 32, 769–773 (1997).
8
RP Harvey, et al., Meteoritic event recorded in Antarctic ice. Geology 26, 607–610 (1998).
9
ML Bender, B Barnett, G Dreyfus, J Jouzel, D Porcelli, The contemporary degassing rate of 40Ar from the solid Earth. Proc Natl Acad Sci USA 105, 8232–8237 (2008).
10
NE Spaulding, et al., Climate archives from 80-250 ka in horizontal and vertical ice cores from the Allan Hills Blue Ice Area, Antarctica. Quat Res 80, 562–574 (2013).
11
J Jouzel, et al., Orbital and millennial Antarctic climate variability over the past 800,000 years. Science 317, 793–796 (2007).
12
M Bender, T Sowers, L Labeyrie, The Dole effect and its variations during the last 130,000 years as measured in the Vostok ice core. Global Biogeochem Cycles 8, 363–376 (1994).
13
P Ciais, JWC White, J Jouzel, JR Petit, The origin of present-day Antarctic precipitation from surface snow deuterium excess data. J Geophys Res Atmos 100, 18917–18927 (1995).
14
B Stenni, et al., The deuterium excess records of EPICA Dome C and Dronning Maud Land ice cores (East Antarctica). Quat Sci Rev 29, 146–159 (2010).
15
G Delisle, J Sievers, Sub-ice topography and meteorite finds near the Allan Hills and the Near Western Ice Field, Victoria Land, Antarctica. J Geophys Res 96, 15577–15587 (1991).
16
NE Spaulding, et al., Ice motion and mass balance at the Allan Hills blue-ice area, Antarctica, with implications for paleoclimate reconstructions. J Glaciol 58, 399–406 (2012).
17
B Bereiter, H Fischer, J Schwander, TF Stocker, Diffusive equilibration of N2, O2 and CO2 mixing ratios in a 1.5 million years old ice core. Cryosphere Discuss 8, 245–256 (2014).
18
H Elderfield, et al., Evolution of ocean temperature and ice volume through the mid-Pleistocene climate transition. Science 337, 704–709 (2012).
19
D Lüthi, et al., High-resolution carbon dioxide concentration record 650,000-800,000 years before present. Nature 453, 379–382 (2008).
20
B Bereiter, et al., Revision of the EPICA Dome C CO2 record from 800 to 600 kyr before present. Geophys Res Lett 42, 542–549 (2015).
21
L Loulergue, et al., Orbital and millennial-scale features of atmospheric CH4 over the past 800,000 years. Nature 453, 383–386 (2008).
22
B Hönisch, NG Hemming, D Archer, M Siddall, JF McManus, Atmospheric carbon dioxide concentration across the mid-Pleistocene transition. Science 324, 1551–1554 (2009).
23
T Blunier, EJ Brook, Timing of millennial-scale climate change in Antarctica and Greenland during the last glacial period. Science 291, 109–112 (2001).
24
ME Raymo, K Ganley, S Carter, DW Oppo, JF McManus, Millenial-scale climate instability during the early Pleistocene epoch. Nature 392, 699–702 (1998).
25
BDA Naafs, J Hefter, R Stein, Millennial-scale ice rafting events and Hudson Strait Heinrich(-like) events during the late Pliocene and Pleistocene: A review. Quat Sci Rev 80, 1–28 (2013).
26
GB Dreyfus, et al., Firn processes and d15N; potential for a gas-phase climate proxy. Quat Sci Rev 29, 28–42 (2010).
27
L Mitchell, E Brook, JE Lee, C Buizert, T Sowers, Constraints on the late holocene anthropogenic contribution to the atmospheric methane budget. Science 342, 964–966 (2013).
28
JH Ahn, EJ Brook, K Howell, A high-precision method for measurement of paleoatmospheric CO2 in small polar ice samples. J Glaciol 55, 499–506 (2009).
Information & Authors
Information
Published in
Classifications
Submission history
Published online: May 11, 2015
Published in issue: June 2, 2015
Keywords
Acknowledgments
We would like to acknowledge US Ice Drilling Design and Operations (IDDO), driller Mike Waszkiewicz, Kristin Schild, Melissa Rohde, and Ken Borek Air for assistance with the field work. Mike Kalk and Haun Marcott assisted with the CO2 measurements. This work was funded by National Science Foundation Grants ANT-0838843 (University of Maine) and ANT-0838849 (Princeton University).
Notes
This article is a PNAS Direct Submission.
Authors
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Altmetrics
Citations
Cite this article
Atmospheric composition 1 million years ago from blue ice in the Allan Hills, Antarctica, Proc. Natl. Acad. Sci. U.S.A.
112 (22) 6887-6891,
https://doi.org/10.1073/pnas.1420232112
(2015).
Copied!
Copying failed.
Export the article citation data by selecting a format from the list below and clicking Export.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.