Isotopic constraints on marine and terrestrial N2O emissions during the last deglaciation

Journal name:
Nature
Volume:
516,
Pages:
234–237
Date published:
DOI:
doi:10.1038/nature13971
Received
Accepted
Published online

Nitrous oxide (N2O) is an important greenhouse gas and ozone-depleting substance that has anthropogenic as well as natural marine and terrestrial sources1. The tropospheric N2O concentrations have varied substantially in the past in concert with changing climate on glacial–interglacial and millennial timescales2, 3, 4, 5, 6, 7, 8. It is not well understood, however, how N2O emissions from marine and terrestrial sources change in response to varying environmental conditions. The distinct isotopic compositions of marine and terrestrial N2O sources can help disentangle the relative changes in marine and terrestrial N2O emissions during past climate variations4, 9, 10. Here we present N2O concentration and isotopic data for the last deglaciation, from 16,000 to 10,000 years before present, retrieved from air bubbles trapped in polar ice at Taylor Glacier, Antarctica. With the help of our data and a box model of the N2O cycle, we find a 30 per cent increase in total N2O emissions from the late glacial to the interglacial, with terrestrial and marine emissions contributing equally to the overall increase and generally evolving in parallel over the last deglaciation, even though there is no a priori connection between the drivers of the two sources. However, we find that terrestrial emissions dominated on centennial timescales, consistent with a state-of-the-art dynamic global vegetation and land surface process model that suggests that during the last deglaciation emission changes were strongly influenced by temperature and precipitation patterns over land surfaces. The results improve our understanding of the drivers of natural N2O emissions and are consistent with the idea that natural N2O emissions will probably increase in response to anthropogenic warming11.

At a glance

Figures

  1. Changes in tropospheric N2O and climate proxies during the last glacial-interglacial cycle and the last deglaciation.
    Figure 1: Changes in tropospheric N2O and climate proxies during the last glacial–interglacial cycle and the last deglaciation.

    a, The past 120 kyr on the AICC2012 timescale26: temperature proxies δ18Oice (δ18Oice = ((18O/16O)sample/(18O/16O)VSMOW − 1) × 1,000‰; VSMOW, Vienna Standard Mean Ocean Water) of Greenland (upper grey curve; North Greenland Ice Core Project27) and Antarctica (lower grey curve; EPICA Dronning Maud Land28), as well as tropospheric N2O (pink; EPICA Dome C29 and North Greenland Ice Core Project3, 5, 8). b, Detailed data from the last deglaciation from 16 to 10 kyr bp on an updated version of the WDC06A-7 timescale15 (Methods): Taylor Glacier CH4 (purple triangles) together with Talos Dome CH4 (grey circles30), Taylor Glacier N2O (pink), as well as δ15N (blue) and δ18O (green; relative to VSMOW) of N2O. Solid lines show splines with a cut-off period of 600 yr through the N2O concentration and isotopic composition data. Error bars indicate pooled standard deviations of replicates (±1σ, n = 10); grey shaded areas indicate ±1σ envelopes from the Monte Carlo approach (Methods). BA, Bølling–Allerød; YD, Younger Dryas; PB, Preboreal.

  2. N2O emissions during the last deglaciation.
    Figure 2: N2O emissions during the last deglaciation.

    a, Total N2O emissions. b, Marine (blue) and terrestrial (green) N2O emission changes relative to 16 kyr bp. Total, marine and terrestrial emissions were inversely calculated using the box model such that they recover the Taylor Glacier N2O and δ15N splines (solid lines in c and d, respectively) in a forward calculation. The uncertainty bands related to the emissions result from the Monte Carlo approach and indicate ±1σ of all solutions. The absolute changes in marine and terrestrial emissions depend on the initial marine fraction, which was set to 37% of the total emissions at 16 kyr bp (see Extended Data Fig. 5 for sensitivity studies). c, Taylor Glacier N2O, with ±1σ error bars. d, Taylor Glacier δ15N of N2O, with ±1σ error bars. The orange dashed line indicates pre-industrial (ad 1750) δ15N (ref. 17). The dashed pink (c) and blue (d) lines show N2O and δ15N calculated using the modelled marine and terrestrial emissions but assuming equilibrium with respect to the sink at any time (Methods); the differences between solid and dashed lines indicate the effect of atmospheric imbalances. e, AMOC changes estimated using the Bern3D Earth System Model (including the ±1σ uncertainty band), constrained by proxy data23. f, Terrestrial N2O emission changes independently inferred from LPX-Bern. g, TraCE-21ka temperature changes over land surfaces25 used to force LPX-Bern.

  3. Isotopic composition of marine and terrestrial N2O sources.
    Extended Data Fig. 1: Isotopic composition of marine and terrestrial N2O sources.

    Field data of δ15N (relative to atmospheric N2) and δ18O (relative to VSMOW) of marine (blue triangles60, 61) and terrestrial (green crosses62, 63) N2O sources. Blue and green bars indicate the ranges as used in the box model (Extended Data Table 2). The mean tropospheric value of all Taylor Glacier data (orange diamond, with the orange box indicating the full range of the data) is enriched in heavy isotopes in both δ15N and δ18O relative to the approximate corresponding isotopic composition of the total source (black diamond) owing to the fractionation by the stratospheric sink.

  4. Comparison of Taylor Glacier and other data.
    Extended Data Fig. 2: Comparison of Taylor Glacier and other data.

    N2O concentration (diamonds), δ15N (triangles) and δ18O (crosses) data from Taylor Glacier (from the apparatus for N2O isotopes in grey, and from the apparatus for CO2 isotopes36 in orange) compared with a Taylor Glacier intercomparison measurement (red) and Talos Dome data (green) from the University of Bern, and with published data from Taylor Dome (blue4). Taylor Glacier and Talos Dome data from the University of Bern were corrected by +2.28 p.p.b. for the N2O concentration, −0.80‰ for δ15N and +0.36‰ for δ18O on the basis of intercalibration measurements made by Oregon State University and the University of Bern using firn air (Methods). The Taylor Glacier δ18O data from Oregon State University (grey crosses) were corrected by +1.38‰ on the basis of measurements with bubble-free ice and NOAA-1 standard gas (Extended Data Fig. 7c and Methods). Error bars indicate ±1σ.

  5. Effect of marine N2O cycle and inventory on tropospheric N2O concentration and [dgr]15N under changing emissions.
    Extended Data Fig. 3: Effect of marine N2O cycle and inventory on tropospheric N2O concentration and δ15N under changing emissions.

    Response of tropospheric N2O and δ15N to exponential increases in N2O emissions with timescales of 100, 200, 500 and 1,000 yr (from left to right; the maximum increase rate in the Taylor Glacier N2O data is indicated by the dotted grey line). Results from the two-box model (dashed lines show results without explicit representation of the ocean) are very similar to the results from the extended box model (with ocean, solid lines; Methods). The marine and terrestrial emissions are increased in parallel, that is, the marine fraction is always 37% of the total N2O emissions and the isotopic composition of emitted N2O remains constant. The decrease in δ15N is caused by imbalances between the sources and the stratospheric sink.

  6. Consistency of the calculated marine and terrestrial emissions with the Taylor Glacier [dgr]18O data.
    Extended Data Fig. 4: Consistency of the calculated marine and terrestrial emissions with the Taylor Glacier δ18O data.

    a, δ18O evolution for different initial marine fractions (red, 17%; purple, 37%) of the total emissions when calculated using the marine and terrestrial N2O emissions determined on the basis of the Taylor Glacier N2O concentration and δ15N data. In a Monte Carlo approach only scenarios with the same mean value as the Taylor Glacier δ18O data (green, with ±1σ error bars) were considered, which narrowed the possible δ18O isotopic composition of the sources. b, δ15N and δ18O of marine (triangles) and terrestrial (crosses) sources from modern field data (black, as in Extended Data Fig. 1), as well as the values needed to explain the Taylor Glacier data with different initial (at 16 kyr bp) marine fractions (red, 17%; purple, 37%). An initial marine fraction of 74% would require δ18O isotopic compositions outside the observed range (Extended Data Table 2), suggesting that such a high marine fraction is rather unlikely. Note that the Taylor Glacier data can be explained for an initial marine fraction of 74% when considering δ15N only, but only with rather extreme model parameters (Extended Data Fig. 5).

  7. Evolution of marine and terrestrial N2O emissions under different scenarios.
    Extended Data Fig. 5: Evolution of marine and terrestrial N2O emissions under different scenarios.

    Sensitivity of marine (blue in large panels) and terrestrial (green in large panels) N2O emissions to initial marine fractions (red circles at 16 kyr bp) set to 17% (a), 37% (b) and 74% (c) in accordance with low, best and high estimates of the modern natural N2O budget1. The uncertainty bands related to the emissions (blue and green shaded areas) result from the Monte Carlo approach and indicate ±1σ of all solutions. For all scenarios, the maximum absolute changes in the marine fractions (black in large panels) over the last deglaciation are 7% or less. Dashed lines in the small panels show the distributions of the parameters as allowed for in the Monte Carlo approach (priors; Extended Data Table 2), and solid lines indicate the distributions of the parameters which allow for a reproduction of the Taylor Glacier N2O concentration and δ15N that respects the prescribed initial marine fractions (posteriors).

  8. Standard gases for [dgr]15N, [dgr]18O and N2O concentration.
    Extended Data Fig. 6: Standard gases for δ15N, δ18O and N2O concentration.

    The δ15N and δ18O values of 6.18‰ and 44.16‰ of the NOAA-1 standard gas collected at Niwot Ridge, Colorado, were assigned by linear extrapolation of the data from Cape Grim, Tasmania31, to the collection date of NOAA-1 (11 December 2008), on the basis of the assumption that N2O and its isotopes are well mixed in the troposphere owing to the rather long atmospheric lifetime. The N2O concentration of 322.32 ± 0.14 p.p.b. of the NOAA-1 standard gas was directly determined by the National Oceanic and Atmospheric Administration (NOAA-2006A scale; linear extrapolation of the Cape Grim data would lead to 320.9 p.p.b.). To test the calibrations for the N2O concentration and isotopic compositions, a second standard gas, NOAA-2, which was collected at Niwot Ridge on 5 October 1988, was measured against NOAA-1. The results (red crosses) were in good agreement with the Cape Grim data, in particular when taking into account the interannual scatter observed in the archived air.

  9. Stability in the course of the measurement series, characterization of the amount dependency of the measurement system, and tests with bubble-free ice.
    Extended Data Fig. 7: Stability in the course of the measurement series, characterization of the amount dependency of the measurement system, and tests with bubble-free ice.

    a, NOAA-3 standard gas measurements performed at the end of each measurement day. No significant drifts were observed in the course of the measurement series, and the standard deviations for δ15N and δ18O were respectively 0.14‰ and 0.32‰ (n = 31), as indicated by the grey areas (±1σ) around the means (dashed lines). b, NOAA-1 standard gas measurements resulting in similar peak areas to the preceding ice-sample measurement routinely performed throughout the measurement series. These measurements covering the full range of peak areas from ice samples did not reveal any significant amount dependency. The mean and standard deviation (dashed lines and grey areas) for δ15N were 6.24‰ ± 0.22‰ and those for δ18O were 44.18‰ ± 0.59‰ (±1σ, n = 58), in agreement with the expected values (solid lines). c, Measurements of different amounts of NOAA-1 standard gas which was stored in the extraction pots while pieces of bubble-free ice were grated. Dashed lines and grey areas indicate the means and standard deviations (±1σ, n = 10), and solid lines indicate the expected values. On the basis of these measurements with bubble-free ice and the results shown in b, N2O, δ15N and δ18O were not corrected for amount dependency. However, during the extraction of standard gas over ice, a −1.38‰ offset was introduced in δ18O. Accordingly, all Taylor Glacier δ18O values were corrected by +1.38‰.

Tables

  1. Equations forming the basis of the two-box model used to calculate marine and terrestrial N2O emissions
    Extended Data Table 1: Equations forming the basis of the two-box model used to calculate marine and terrestrial N2O emissions
  2. Parameters for the two-box model used to calculate marine and terrestrial N2O emissions
    Extended Data Table 2: Parameters for the two-box model used to calculate marine and terrestrial N2O emissions

References

  1. Stocker, T. F. et al. (eds) Climate Change 2013: The Physical Science Basis (Cambridge Univ. Press, 2013)
  2. Flückiger, J. et al. Variations in atmospheric N2O concentration during abrupt climatic changes. Science 285, 227230 (1999)
  3. Schilt, A. et al. The response of atmospheric nitrous oxide to climate variations during the last glacial period. Geophys. Res. Lett. 40, 18881893 (2013)
  4. Sowers, T., Alley, R. B. & Jubenville, J. Ice core records of atmospheric N2O covering the last 106,000 years. Science 301, 945948 (2003)
  5. Schilt, A. et al. Atmospheric nitrous oxide during the last 140,000 years. Earth Planet. Sci. Lett. 300, 3343 (2010)
  6. Spahni, R. et al. Atmospheric methane and nitrous oxide of the late Pleistocene from Antarctic ice cores. Science 310, 13171321 (2005)
  7. Schilt, A. et al. Glacial-interglacial and millennial-scale variations in the atmospheric nitrous oxide concentration during the last 800,000 years. Quat. Sci. Rev. 29, 182192 (2010)
  8. Flückiger, J. et al. N2O and CH4 variations during the last glacial epoch: insight into global processes. Glob. Biogeochem. Cycles 18, GB1020 (2004)
  9. Ishijima, K. et al. Temporal variations of the atmospheric nitrous oxide concentration and its δ15N and δ18O for the latter half of the 20th century reconstructed from firn air analyses. J. Geophys. Res. 112, D03305 (2007)
  10. Röckmann, T., Kaiser, J. & Brenninkmeijer, C. A. M. The isotopic fingerprint of the pre-industrial and the anthropogenic N2O source. Atmos. Chem. Phys. 3, 315323 (2003)
  11. Stocker, B. D. et al. Multiple greenhouse-gas feedbacks from the land biosphere under future climate change scenarios. Nature Clim. Change 3, 666672 (2013)
  12. Prather, M. J., Holmes, C. D. & Hsu, J. Reactive greenhouse gas scenarios: systematic exploration of uncertainties and the role of atmospheric chemistry. Geophys. Res. Lett. 39, L09803 (2012)
  13. Crutzen, P. J. & Bruhl, C. A model study of atmospheric temperatures and the concentrations of ozone, hydroxyl, and some other photochemically active gases during the glacial, the preindustrial Holocene and the present. Geophys. Res. Lett. 20, 10471050 (1993)
  14. Martinerie, P., Brasseur, G. P. & Granier, C. The chemical composition of ancient atmospheres: a model study constrained by ice core data. J. Geophys. Res. 100, 1429114304 (1995)
  15. Wais Divide Project Members. Onset of deglacial warming in West Antarctica driven by local orbital forcing. Nature 500, 440444 (2013)
  16. Indermühle, A. et al. Atmospheric CO2 concentration from 60 to 20 kyr BP from the Taylor Dome Ice Core, Antarctica. Geophys. Res. Lett. 27, 735738 (2000)
  17. Bernard, S. et al. Constraints on N2O budget changes since pre-industrial time from new firn air and ice core isotope measurements. Atmos. Chem. Phys. 6, 493503 (2006)
  18. Galbraith, E. D. & Kienast, M. &. The NICOPP working group members. The acceleration of oceanic denitrification during deglacial warming. Nature Geosci. 6, 579584 (2013)
  19. McLauchlan, K. K., Williams, J. J., Craine, J. M. & Jeffers, E. S. Changes in global nitrogen cycling during the Holocene epoch. Nature 495, 352355 (2013)
  20. Hirsch, A. I. et al. Inverse modeling estimates of the global nitrous oxide surface flux from 1998–2001. Glob. Biogeochem. Cycles 20, GB1008 (2006)
  21. Goldstein, B., Joos, F. & Stocker, T. F. A modeling study of oceanic nitrous oxide during the Younger Dryas cold period. Geophys. Res. Lett. 30, 1092 (2003)
  22. Schmittner, A. & Galbraith, E. D. Glacial greenhouse-gas fluctuations controlled by ocean circulation changes. Nature 456, 373376 (2008)
  23. Ritz, S. P. et al. Estimated strength of the Atlantic overturning circulation during the last deglaciation. Nature Geosci. 6, 208212 (2013)
  24. Jaccard, S. L. & Galbraith, E. D. Large climate-driven changes of oceanic oxygen concentrations during the last deglaciation. Nature Geosci. 5, 151156 (2012)
  25. Liu, Z. et al. Transient simulation of last deglaciation with a new mechanism for Bølling-Allerød warming. Science 325, 310314 (2009)
  26. Veres, D. et al. The Antarctic ice core chronology (AICC2012): an optimized multi-parameter and multi-site dating approach for the last 120 thousand years. Clim. Past 9, 17331748 (2013)
  27. NGRIP Members. High-resolution record of Northern Hemisphere climate extending into the last interglacial period. Nature 431, 147151 (2004)
  28. EPICA Community Members. One-to-one coupling of glacial climate variability in Greenland and Antarctica. Nature 444, 195198 (2006)
  29. Flückiger, J. et al. High-resolution Holocene N2O ice core record and its relationship with CH4 and CO2. Glob. Biogeochem. Cycles 16, 1010 (2002)
  30. Stenni, B. et al. Expression of the bipolar see-saw in Antarctic climate records during the last deglaciation. Nature Geosci. 4, 4649 (2011)
  31. Park, S. et al. Trends and seasonal cycles in the isotopic composition of nitrous oxide since 1940. Nature Geosci. 5, 261265 (2012)
  32. Schmitt, J., Seth, B., Bock, M. & Fischer, H. Online technique for isotope and mixing ratios of CH4, N2O, Xe and mixing ratios of organic trace gases on a single ice core sample. Atmos. Meas. Tech. 7, 26452665 (2014)
  33. Sowers, T. N2O record spanning the penultimate deglaciation from the Vostok ice core. J. Geophys. Res. 106, 3190331914 (2001)
  34. Sapart, C. J. et al. Simultaneous stable isotope analysis of methane and nitrous oxide on ice core samples. Atmos. Meas. Tech. 4, 26072618 (2011)
  35. Sperlich, P. et al. An automated GC-C-GC-IRMS setup to measure palaeoatmospheric δ13C-CH4, δ15N-N2O and δ18O-N2O in one ice core sample. Atmos. Meas. Tech. 6, 20272041 (2013)
  36. Bauska, T. K., Brook, E. J., Mix, A. C. & Ross, A. High-precision dual-inlet IRMS measurements of the stable isotopes of CO2 and the N2O/CO2 ratio from polar ice core samples. Atmos. Meas. Tech. 7, 38253837 (2014)
  37. Kaiser, J. Stable Isotope Investigations of Atmospheric Nitrous Oxide 1721. PhD thesis, Johannes Gutenberg Univ. Mainz. (2002)
  38. Lisiecki, L. E. & Lisiecki, P. A. Application of dynamic programming to the correlation of paleoclimate records. Paleoceanography 17, 1049 (2002)
  39. Buizert, C., Sowers, T. & Blunier, T. Assessment of diffusive isotopic fractionation in polar firn, and application to ice core trace gas records. Earth Planet. Sci. Lett. 361, 110119 (2013)
  40. Tans, P. P. A note on isotopic ratios and the global atmospheric methane budget. Glob. Biogeochem. Cycles 11, 7781 (1997)
  41. Sowers, T., Rodebaugh, A., Yoshida, N. & Toyoda, S. Extending records of the isotopic composition of atmospheric N2O back to 1800 AD from air trapped in snow at the South Pole and the Greenland Ice Sheet Project II ice core. Glob. Biogeochem. Cycles 16, 1129 (2002)
  42. Roth, R. Modeling Forcings and Responses in the Global Carbon Cycle-Climate System: Past, Present and Future. PhD thesis, Univ. Bern. (2013)
  43. Spahni, R. et al. Transient simulations of the carbon and nitrogen dynamics in northern peatlands: from the Last Glacial Maximum to the 21st century. Clim. Past 9, 12871308 (2013)
  44. Sitch, S. et al. Evaluation of ecosystem dynamics, plant geography and terrestrial carbon cycling in the LPJ dynamic global vegetation model. Glob. Change Biol. 9, 161185 (2003)
  45. Joos, F. et al. Transient simulations of Holocene atmospheric carbon dioxide and terrestrial carbon since the Last Glacial Maximum. Glob. Biogeochem. Cycles 18, GB2002 (2004)
  46. Gerten, D. et al. Terrestrial vegetation and water balance - hydrological evaluation of a dynamic global vegetation model. J. Hydrol. (Amst.) 286, 249270 (2004)
  47. Wania, R., Ross, I. & Prentice, I. C. Integrating peatlands and permafrost into a dynamic global vegetation model: 1. Evaluation and sensitivity of physical land surface processes. Glob. Biogeochem. Cycles 23, GB3014 (2009)
  48. Wania, R., Ross, I. & Prentice, I. C. Integrating peatlands and permafrost into a dynamic global vegetation model: 2. Evaluation and sensitivity of vegetation and carbon cycle processes. Glob. Biogeochem. Cycles 23, GB3015 (2009)
  49. Xu-Ri & Prentice, I. C. Terrestrial nitrogen cycle simulation with a dynamic global vegetation model. Glob. Change Biol. 14, 17451764 (2008)
  50. Xu-Ri, I. C., Spahni, R. & Niu, H. S. Modelling terrestrial nitrous oxide emissions and implications for climate feedback. New Phytol. 196, 472488 (2012)
  51. Wania, R., Ross, I. & Prentice, I. C. Implementation and evaluation of a new methane model within a dynamic global vegetation model: LPJ-WHyMe v1.3.1. Geosci. Model Dev. 3, 565584 (2010)
  52. Spahni, R. et al. Constraining global methane emissions and uptake by ecosystems. Biogeosciences 8, 16431665 (2011)
  53. Zürcher, S. et al. Impact of an abrupt cooling event on interglacial methane emissions in northern peatlands. Biogeosciences 10, 19631981 (2013)
  54. Pfeiffer, M., van Leeuwen, J., van der Knaap, W. O. & Kaplan, J. O. The effect of abrupt climatic warming on biogeochemical cycling and N2O emissions in a terrestrial ecosystem. Palaeogeogr. Palaeoclimatol. Palaeoecol. 391, 7483 (2013)
  55. He, F. Simulating Transient Climate Evolution of the Last Deglaciation with CCSM3. PhD thesis, Univ. Wisconsin-Madison. (2011)
  56. Mitchell, T. D. & Jones, P. D. An improved method of constructing a database of monthly climate observations and associated high-resolution grids. Int. J. Climatol. 25, 693712 (2005)
  57. Joos, F. & Spahni, R. Rates of change in natural and anthropogenic radiative forcing over the past 20,000 years. Proc. Natl Acad. Sci. USA 105, 14251430 (2008)
  58. Berger, A. L. Long-term variations of daily insolation and quaternary climatic changes. J. Atmos. Sci. 35, 23622367 (1978)
  59. Peltier, W. R. Global glacial isostasy and the surface of the ice-age Earth: The ICE-5G (VM2) model and GRACE. Annu. Rev. Earth Planet. Sci. 32, 111149 (2004)
  60. Kim, K.-R. & Craig, H. Two-isotope characterization of N2O in the Pacific Ocean and constraints on its origin in deep water. Nature 347, 5861 (1990)
  61. Yoshinari, T. et al. Nitrogen and oxygen isotopic composition of N2O from suboxic waters of the eastern tropical North Pacific and the Arabian Sea—measurement by continuous-flow isotope-ratio monitoring. Mar. Chem. 56, 253264 (1997)
  62. Kim, K.-R. & Craig, H. Nitrogen-15 and oxygen-18 characteristics of nitrous oxide: a global perspective. Science 262, 18551857 (1993)
  63. Pérez, T. et al. Isotopic variability of N2O emissions from tropical forest soils. Glob. Biogeochem. Cycles 14, 525535 (2000)
  64. Rahn, T. & Wahlen, M. Stable isotope enrichment in stratospheric nitrous oxide. Science 278, 17761778 (1997)
  65. Röckmann, T. et al. Isotopic enrichment of nitrous oxide (15N14NO, 14N15NO, 14N14N18O) in the stratosphere and in the laboratory. J. Geophys. Res. 106, 1040310410 (2001)

Download references

Author information

Affiliations

  1. College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis, Oregon 97331, USA

    • Adrian Schilt,
    • Edward J. Brook &
    • Thomas K. Bauska
  2. Climate and Environmental Physics, Physics Institute, and Oeschger Centre for Climate Change Research, University of Bern, 3012 Bern, Switzerland

    • Adrian Schilt,
    • Hubertus Fischer,
    • Fortunat Joos,
    • Jochen Schmitt,
    • Renato Spahni &
    • Thomas F. Stocker
  3. Scripps Institution of Oceanography, University of California, San Diego, California 92037, USA

    • Daniel Baggenstos &
    • Jeffrey P. Severinghaus
  4. Department of Earth and Environmental Sciences, University of Rochester, Rochester, New York 14627, USA

    • Vasilii V. Petrenko
  5. National Institute of Water and Atmospheric Research, Wellington 6021, New Zealand

    • Hinrich Schaefer

Contributions

J.P.S., V.V.P. and E.J.B. initiated and led the Taylor Glacier project. T.K.B., E.J.B., D.B., V.V.P., H.S., A.S. and J.P.S. performed fieldwork. A.S. set up the experimental apparatus for N2O isotopes and carried out the measurements, with great support from T.K.B., E.J.B. and J.S. Intercomparison measurements were performed by J.S. at the University of Bern and by T.K.B. at Oregon State University. D.B. and T.K.B. developed the timescale for Taylor Glacier. A.S. and F.J. performed the box modelling. R.S. and F.J. produced and interpreted the LPX-Bern model results. All authors discussed the results and contributed to the manuscript, which was written by A.S.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Extended data figures and tables

Extended Data Figures

  1. Extended Data Figure 1: Isotopic composition of marine and terrestrial N2O sources. (186 KB)

    Field data of δ15N (relative to atmospheric N2) and δ18O (relative to VSMOW) of marine (blue triangles60, 61) and terrestrial (green crosses62, 63) N2O sources. Blue and green bars indicate the ranges as used in the box model (Extended Data Table 2). The mean tropospheric value of all Taylor Glacier data (orange diamond, with the orange box indicating the full range of the data) is enriched in heavy isotopes in both δ15N and δ18O relative to the approximate corresponding isotopic composition of the total source (black diamond) owing to the fractionation by the stratospheric sink.

  2. Extended Data Figure 2: Comparison of Taylor Glacier and other data. (301 KB)

    N2O concentration (diamonds), δ15N (triangles) and δ18O (crosses) data from Taylor Glacier (from the apparatus for N2O isotopes in grey, and from the apparatus for CO2 isotopes36 in orange) compared with a Taylor Glacier intercomparison measurement (red) and Talos Dome data (green) from the University of Bern, and with published data from Taylor Dome (blue4). Taylor Glacier and Talos Dome data from the University of Bern were corrected by +2.28 p.p.b. for the N2O concentration, −0.80‰ for δ15N and +0.36‰ for δ18O on the basis of intercalibration measurements made by Oregon State University and the University of Bern using firn air (Methods). The Taylor Glacier δ18O data from Oregon State University (grey crosses) were corrected by +1.38‰ on the basis of measurements with bubble-free ice and NOAA-1 standard gas (Extended Data Fig. 7c and Methods). Error bars indicate ±1σ.

  3. Extended Data Figure 3: Effect of marine N2O cycle and inventory on tropospheric N2O concentration and δ15N under changing emissions. (167 KB)

    Response of tropospheric N2O and δ15N to exponential increases in N2O emissions with timescales of 100, 200, 500 and 1,000 yr (from left to right; the maximum increase rate in the Taylor Glacier N2O data is indicated by the dotted grey line). Results from the two-box model (dashed lines show results without explicit representation of the ocean) are very similar to the results from the extended box model (with ocean, solid lines; Methods). The marine and terrestrial emissions are increased in parallel, that is, the marine fraction is always 37% of the total N2O emissions and the isotopic composition of emitted N2O remains constant. The decrease in δ15N is caused by imbalances between the sources and the stratospheric sink.

  4. Extended Data Figure 4: Consistency of the calculated marine and terrestrial emissions with the Taylor Glacier δ18O data. (212 KB)

    a, δ18O evolution for different initial marine fractions (red, 17%; purple, 37%) of the total emissions when calculated using the marine and terrestrial N2O emissions determined on the basis of the Taylor Glacier N2O concentration and δ15N data. In a Monte Carlo approach only scenarios with the same mean value as the Taylor Glacier δ18O data (green, with ±1σ error bars) were considered, which narrowed the possible δ18O isotopic composition of the sources. b, δ15N and δ18O of marine (triangles) and terrestrial (crosses) sources from modern field data (black, as in Extended Data Fig. 1), as well as the values needed to explain the Taylor Glacier data with different initial (at 16 kyr bp) marine fractions (red, 17%; purple, 37%). An initial marine fraction of 74% would require δ18O isotopic compositions outside the observed range (Extended Data Table 2), suggesting that such a high marine fraction is rather unlikely. Note that the Taylor Glacier data can be explained for an initial marine fraction of 74% when considering δ15N only, but only with rather extreme model parameters (Extended Data Fig. 5).

  5. Extended Data Figure 5: Evolution of marine and terrestrial N2O emissions under different scenarios. (499 KB)

    Sensitivity of marine (blue in large panels) and terrestrial (green in large panels) N2O emissions to initial marine fractions (red circles at 16 kyr bp) set to 17% (a), 37% (b) and 74% (c) in accordance with low, best and high estimates of the modern natural N2O budget1. The uncertainty bands related to the emissions (blue and green shaded areas) result from the Monte Carlo approach and indicate ±1σ of all solutions. For all scenarios, the maximum absolute changes in the marine fractions (black in large panels) over the last deglaciation are 7% or less. Dashed lines in the small panels show the distributions of the parameters as allowed for in the Monte Carlo approach (priors; Extended Data Table 2), and solid lines indicate the distributions of the parameters which allow for a reproduction of the Taylor Glacier N2O concentration and δ15N that respects the prescribed initial marine fractions (posteriors).

  6. Extended Data Figure 6: Standard gases for δ15N, δ18O and N2O concentration. (308 KB)

    The δ15N and δ18O values of 6.18‰ and 44.16‰ of the NOAA-1 standard gas collected at Niwot Ridge, Colorado, were assigned by linear extrapolation of the data from Cape Grim, Tasmania31, to the collection date of NOAA-1 (11 December 2008), on the basis of the assumption that N2O and its isotopes are well mixed in the troposphere owing to the rather long atmospheric lifetime. The N2O concentration of 322.32 ± 0.14 p.p.b. of the NOAA-1 standard gas was directly determined by the National Oceanic and Atmospheric Administration (NOAA-2006A scale; linear extrapolation of the Cape Grim data would lead to 320.9 p.p.b.). To test the calibrations for the N2O concentration and isotopic compositions, a second standard gas, NOAA-2, which was collected at Niwot Ridge on 5 October 1988, was measured against NOAA-1. The results (red crosses) were in good agreement with the Cape Grim data, in particular when taking into account the interannual scatter observed in the archived air.

  7. Extended Data Figure 7: Stability in the course of the measurement series, characterization of the amount dependency of the measurement system, and tests with bubble-free ice. (485 KB)

    a, NOAA-3 standard gas measurements performed at the end of each measurement day. No significant drifts were observed in the course of the measurement series, and the standard deviations for δ15N and δ18O were respectively 0.14‰ and 0.32‰ (n = 31), as indicated by the grey areas (±1σ) around the means (dashed lines). b, NOAA-1 standard gas measurements resulting in similar peak areas to the preceding ice-sample measurement routinely performed throughout the measurement series. These measurements covering the full range of peak areas from ice samples did not reveal any significant amount dependency. The mean and standard deviation (dashed lines and grey areas) for δ15N were 6.24‰ ± 0.22‰ and those for δ18O were 44.18‰ ± 0.59‰ (±1σ, n = 58), in agreement with the expected values (solid lines). c, Measurements of different amounts of NOAA-1 standard gas which was stored in the extraction pots while pieces of bubble-free ice were grated. Dashed lines and grey areas indicate the means and standard deviations (±1σ, n = 10), and solid lines indicate the expected values. On the basis of these measurements with bubble-free ice and the results shown in b, N2O, δ15N and δ18O were not corrected for amount dependency. However, during the extraction of standard gas over ice, a −1.38‰ offset was introduced in δ18O. Accordingly, all Taylor Glacier δ18O values were corrected by +1.38‰.

Extended Data Tables

  1. Extended Data Table 1: Equations forming the basis of the two-box model used to calculate marine and terrestrial N2O emissions (107 KB)
  2. Extended Data Table 2: Parameters for the two-box model used to calculate marine and terrestrial N2O emissions (126 KB)

Additional data