Reconstructed changes in Arctic sea ice over the past 1,450 years

Journal name:
Date published:
Published online

Arctic sea ice extent is now more than two million square kilometres less than it was in the late twentieth century, with important consequences for the climate, the ocean and traditional lifestyles in the Arctic1, 2. Although observations show a more or less continuous decline for the past four or five decades3, 4, there are few long-term records with which to assess natural sea ice variability. Until now, the question of whether or not current trends are potentially anomalous5 has therefore remained unanswerable. Here we use a network of high-resolution terrestrial proxies from the circum-Arctic region to reconstruct past extents of summer sea ice, and show that—although extensive uncertainties remain, especially before the sixteenth century—both the duration and magnitude of the current decline in sea ice seem to be unprecedented for the past 1,450 years. Enhanced advection of warm Atlantic water to the Arctic6 seems to be the main factor driving the decline of sea ice extent on multidecadal timescales, and may result from nonlinear feedbacks between sea ice and the Atlantic meridional overturning circulation. These results reinforce the assertion that sea ice is an active component of Arctic climate variability and that the recent decrease in summer Arctic sea ice is consistent with anthropogenically forced warming.

At a glance


  1. Map of the Arctic showing the location and type of proxies used in the reconstruction and ocean sediment cores used for comparison.
    Figure 1: Map of the Arctic showing the location and type of proxies used in the reconstruction and ocean sediment cores used for comparison.

    Red and blue contours respectively delineate the ice edge in August 2007 and 1951, the years of minimum and maximum ice extent from gridded historical sea ice data3.

  2. Validation statistics for the reconstruction.
    Figure 2: Validation statistics for the reconstruction.

    a, Shaded grey/black, Reconstruction of Error (RE) statistic; continuous and dashed red lines, 99% and 95% confidence levels on RE from Monte Carlo noise simulations, respectively; blue line, number of potential proxy predictors; dashed blue line, number of proxies retained for prediction; dashed black line, calibration period 1870–1995. b, Multiscale validation: the black contour delineates significant RE values (P<0.01); the inset shows the mean RE 99% threshold for significance by period. c, Observed ice extent record (black) over the 1870–1995 calibration interval and ice extent predicted by the reduced calibration sets during cross-validation, for four distinct reconstruction steps (coloured points).

  3. Comparison between reconstructed late-summer Arctic ice extent and other Arctic sea ice, climate and oceanic proxy records.
    Figure 3: Comparison between reconstructed late-summer Arctic ice extent and other Arctic sea ice, climate and oceanic proxy records.

    a, Forty-year smoothed reconstructed late-summer Arctic sea ice extent with 95% confidence interval. b, Chukchi Sea ice cover duration reconstructed from core 92-B5 (ref. 9). c, Fram Strait sea ice cover duration reconstructed from core JM04 (ref. 20). d, Normalized IP25 flux in the BASICC-8 sediment core, a proxy for springtime sea ice occurrence in the western Barents Sea. The timescale used is the average of the two published timescales, corresponding to high and low sedimentation rate alternatives23. e, Relative abundance of IP25 in core MD99-2275, a proxy for springtime sea ice occurrence along the northern coast of Iceland8. f, Reconstructed Arctic surface air temperature anomalies24. Our sea ice reconstruction includes 9 of the 23 proxies used in this reconstruction. g, Two reconstructions of the NAO index25, 26. The Cook02 record mainly reconstructs interannual variations26; its variance was matched to the Trouet09 record, which reflects decadal-scale changes25. h, Reconstructed sea surface temperature (SST) and salinity in Fram Strait20. All records with a time resolution smaller than 40years were interpolated to 1year and then smoothed with a 40-year lowpass filter. The dashed lines in a, b, c and f are from modern observations. The dotted line in c is interpolated between modern observations and the reconstruction. The shaded grey areas T1–T3 delineate the periods of reduced ice extent discussed in the text.


  1. Serreze, M. C., Holland, M. M. & Stroeve, J. Perspectives on the Arctic’s shrinking sea-ice cover. Science 315, 15331536 (2007)
  2. Perovich, D. & Richter-Menge, J. Loss of sea ice in the Arctic. Annu. Rev. Mar. Sci. 1, 417441 (2009)
  3. Walsh, J. E. & Chapman, W. L. 20th-century sea-ice variations from observational data. Ann. Glaciol. 33, 444448 (2001)
  4. Comiso, J. & Nishio, F. Trends in the sea ice cover using enhanced and compatible AMSR-E, SSM/I, and SMMR data. J. Geophys. Res. 113 doi:10.1029/2007JC004257 (2008)
  5. Polyak, L. et al. History of sea ice in the Arctic. Quat. Sci. Rev. 29, 17571778 (2010)
  6. Spielhagen, R. F. et al. Enhanced modern heat transfer to the Arctic by warm Atlantic water. Science 331, 450453 (2011)
  7. Wang, M. & Overland, J. A sea ice free summer Arctic within 30 years. Geophys. Res. Lett. 36 L07502 doi:10.1029/2009GL037820 (2009)
  8. Massé, G. et al. Abrupt climate changes for Iceland during the last millennium: evidence from high resolution sea ice reconstructions. Earth Planet. Sci. Lett. 269, 565569 (2008)
  9. de Vernal, A., Hillaire-Marcel, C., Solignac, S., Radi, T. & Rochon, A. in Arctic Sea Ice Decline: Observations, Projections, Mechanisms, and Implications (eds DeWeaver, E. T., Bitz, C. M. & Tremblay, L.-B.) (Geophysical Monograph Series Vol. 80) 2745 (American Geophysical Union, 2008)
  10. Macias Fauria, M. et al. Unprecedented low twentieth century winter sea ice extent in the Western Nordic Seas since AD 1200. Clim. Dyn. 34, 781795 (2010)
  11. Fisher, D. et al. Natural variability of Arctic sea ice over the Holocene. Eos 87, 273275 (2006)
  12. Grumet, N. S. et al. Variability of sea-ice extent in Baffin Bay over the last millennium. Clim. Change 49, 129145 (2001)
  13. Jones, P. D. et al. High-resolution palaeoclimatology of the last millennium: a review of current status and future prospects. Holocene 19, 349 (2009)
  14. Jacoby, G. C. & Ulan, L. D. Reconstruction of past ice conditions in a Hudson Bay estuary using tree rings. Nature 298, 637639 (1982)
  15. Wold, S., Sjöström, M. & Eriksson, L. PLS-regression: a basic tool of chemometrics. Chemom. Intell. Lab. Syst. 58, 109130 (2001)
  16. Polyakov, I. et al. Long-term ice variability in Arctic marginal seas. J. Clim. 16, 20782085 (2003)
  17. Mann, M. E. et al. Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia. Proc. Natl Acad. Sci. USA 105, 1325213257 (2008)
  18. Jones, P. & Mann, M. Climate over past millennia. Rev. Geophys. 42 RG2002 doi:10.1029/2003RG000143 (2004)
  19. Min, S.-K., Zhang, X., Zwiers, F. W. & Agnew, T. Human influence on Arctic sea ice detectable from early 1990s onwards. Geophys. Res. Lett. 35 doi:10.1029/2008GL035725 (2008)
  20. Bonnet, S., de Vernal, A., Hillaire-Marcel, C., Radi, T. & Husum, K. Variability of sea-surface temperature and sea-ice cover in the Fram Strait over the last two millennia. Mar. Micropaleontol. 74, 5974 (2010)
  21. Richerol, T. et al. Evolution of paleo sea-surface conditions over the last 600 years in the Mackenzie Trough, Beaufort Sea (Canada). Mar. Micropaleontol. 68, 620 (2008)
  22. Dickson, R. et al. The Arctic Ocean response to the North Atlantic Oscillation. J. Clim. 13, 26712696 (2000)
  23. Vare, L. L., Massé, G. & Belt, S. T. A biomarker-based reconstruction of sea ice conditions for the Barents Sea in recent centuries. Holocene 20, 637643 (2010)
  24. Kaufman, D. et al. Recent warming reverses long-term Arctic cooling. Science 325, 12361239 (2009)
  25. Trouet, V. et al. Persistent positive North Atlantic Oscillation mode dominated the medieval climate anomaly. Science 324, 7880 (2009)
  26. Cook, E. R., D’Arrigo, R. D. & Mann, M. E. A well-verified, multiproxy reconstruction of the winter North Atlantic Oscillation index since A.D.1400. J. Clim. 15, 17541764 (2002)
  27. Crespin, E., Goosse, H., Fichefet, T. & Mann, M. The 15th century Arctic warming in coupled model simulations with data assimilation. Clim. Past 5, 389401 (2009)
  28. Spielhagen, R. F. et al. Enhanced modern heat transfer to the Arctic by warm Atlantic water. Science 331, 450453 (2011)
  29. Holland, M. M., Bitz, C. M., Eby, M. & Weaver, A. J. The role of ice-ocean interactions in the variability of the North Atlantic thermohaline circulation. J. Clim. 14, 656675 (2001)
  30. Wahl, E. & Ammann, C. Robustness of the Mann, Bradley, Hughes reconstruction of Northern Hemisphere surface temperatures: examination of criticisms based on the nature and processing of proxy climate evidence. Clim. Change 85, 3369 (2007)
  31. Beckers, J. & Rixen, M. EOF calculations and data filling from incomplete oceanographic datasets. J. Atmos. Ocean. Technol. 20, 18391856 (2003)
  32. Christiansen, B., Schmith, T. & Thejll, P. A surrogate ensemble study of climate reconstruction methods: stochasticity and robustness. J. Clim. 22, 951976 (2009)
  33. Lee, T., Zwiers, F. & Tsao, M. Evaluation of proxy-based millennial reconstruction methods. Clim. Dyn. 31, 263281 (2008)
  34. Mann, M. Climate over the past two millennia. Annu. Rev. Earth Planet. Sci. 35, 111136 (2007)
  35. Naes, T., Isaksson, T., Fearn, T. & Davies, T. Multivariate Calibration and Classification (NIR Publications, 2004)
  36. Cook, E., Briffa, K. & Jones, P. Spatial regression methods in dendroclimatology: a review and comparison of two techniques. Int. J. Climatol. 14, 379402 (1994)
  37. Luterbacher, J., Dietrich, D., Xoplaki, E., Grosjean, M. & Wanner, H. European seasonal and annual temperature variability, trends, and extremes since 1500. Science 303, 14991503 (2004)
  38. Michaelsen, J. Cross-validation in statistical climate forecast models. J. Appl. Meteorol. 26, 15891600 (1987)
  39. Martens, H. & Martens, M. Modified Jack-knife estimation of parameter uncertainty in bilinear modelling by partial least squares regression (PLSR). Food Qual. Prefer. 11, 516 (2000)

Download references

Author information


  1. Centro de Estudios Avanzados en Zonas Áridas, Benavente 980, Casilla 554, La Serena, 1720170, Chile

    • Christophe Kinnard
  2. Geological Survey of Canada, NRCAN, Ottawa, Ontario, K1A 0E8, Canada

    • Christian M. Zdanowicz &
    • David A. Fisher
  3. Norwegian Polar Institute, FRAM – High North Research Centre on Climate and the Environment, NO-9296 Tromsø, Norway

    • Elisabeth Isaksson
  4. GEOTOP, Université du Québec à Montréal, PO Box 8888, Montréal, Québec, H3C 3P8, Canada

    • Anne de Vernal
  5. Byrd Polar Research Center, Ohio 43210, Columbus, USA

    • Lonnie G. Thompson


C.K. conducted all data preparation and analyses. C.M.Z. and D.A.F. instigated and directed the research and contributed to the interpretation. E.I., A.deV. and L.G.T. contributed data and participated in the interpretation of results. All authors contributed to manuscript preparation.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Supplementary information

PDF files

  1. Supplementary Information (2.3M)

    The file contains Supplementary Figures 1-11 with legends, Supplementary Methods, Supplementary Table 1, a Supplementary Discussion and additional references.

Excel files

  1. Supplementary Data 1 (1M)

    This file contains the original proxy data used in the study.

  2. Supplementary Data 2 (1.1M)

    This file contains smoothed (5-year lowpass) and infilled proxy records used for calibration against historical sea-ice data.

  3. Supplementary Data 3 (25K)

    This file contains 1-year and 5-year smoothed historical August sea-ice extent used in this study.

  4. Supplementary Data 4 (659K)

    This file contains all data plotted in Figure 3 of the article.

Zip files

  1. Supplementary Data 5 (20K)

    This file contains Matlab codes/ calibaration comprising: calibrateproxies.m - steps for PLS calibration of historical sea ice extent with the proxy network; call_calibrateproxies.m - script to load matlab data and call calibrateproxies.m / associated codes; Various Matlab functions called by calibrateproxies.m

  2. Supplementary Data 6 (324K)

    This file contains Matlab data comprising: AugustIceExtentWCPOL_infilled.mat: August historical sea ice extent (SeaIceExtent.xls) in Matlab format; PROXIESpre1995_infilled_withnoise.mat: smoothed (5-year lowpass) and infilled proxy records (ProxyData_5YR_infilled.xls); PROXYMETADATA1.mat: Matlab structure containing information on proxy records, called by calibrateproxies.m.


  1. Report this comment #58731

    Arno Arrak said:

    The authors conclude their article with the statement that '...anthropologically forced ("greenhouse gases") warming stands out as a very plausible cause of the record atmospheric and oceanic warmth of the recent decades...' This is just a statement of opinion because there is nothing whatsoever in this paper to back it up. I have no idea why reviewers and editors let it go through. Their theory, such as it is, is entirely inadequate to explain the Arctic warming that controls the ice cover. In my paper which appeared two years ago 1 and which they ignore, I provided a full explanation of Arctic warming during the past century. I ruled out greenhouse warming as a cause because the laws of physics do not permit it. Having carbon dioxide in the atmosphere in and of itself is not sufficient to start a greenhouse warming. If we are dealing with Arrhenius type warming then starting one requires a simultaneous increase of atmospheric carbon dioxide. That is the only way to increase absorption because the IR absorptivity of carbon dioxide is a property of the gas and cannot be changed. The present Arctic warming began suddenly at the turn of the twentieth century. There was no parallel increase of atmospheric CO2 at the time which immediately rules out carbon dioxide as its cause. The credit for discovering the warming goes to Kaufman et al. 2 who published it in 2009. Using lake sediments they worked out a two thousand year record of Arctic temperatures. For most of these two thousand years there was nothing but slow, linear cooling, most likely due to orbitally driven reduction of summer insolation. This came to a sudden end at the turn of the twentieth century when temperatures turned sharply upward. Since carbon dioxide was ruled out as a cause it became obvious to me that the only logical cause could be a rearrangement of the North Atlantic current system that started carrying warm water north. The source of that warm water is of course the Gulf Stream. That is why the Arctic is still warming when global warming as a whole is at a standstill. According to Kaufman the warming paused in the middle of the century, then resumed, and is still going strong. There are numerous observations of this warming going back to the twenties. Some are documented in my paper. The most recent observation is by Spielhagen et al. 3 who measured Arctic temperature directly and took a foraminiferal core near Svalbard. They found that temperatures reaching the Arctic today exceeded anything seen for the last two thousand years. Both Kaufman and Spielhagen have made seminal discoveries about Arctic warming but both are still firm believers that the warming is anthropogenic. Kaufman puts it this way: "...This shift correlates with the rise in global average temperature which coincided with the onset of global anthropogenic changes in global atmospheric composition..." Here he has just made an important discovery but he throws the effect away with this gobbledygook that has no science content. Great job, good technician, but not a scientist is my judgement about that. Judging by the literature he is not alone. His data for the twentieth century were hard to see so I used NOAA Arctic Report Card for 2010 to get a more detailed picture of twentieth century temperatures. It showed the warming in more detail than Kaufman did and also showed that the pause in mid-century lasted thirty years. Furthermore, it was not just a warming pause but an actual cooling at the rate of 0.3 degrees Celsius per decade. It is very likely that what happened was a temporary return of the former flow pattern of ocean currents. If so, we should be ready for a repeat performance. Nature is fickle and what has happened before can happen again. Needless to say, this is of huge importance to the exploitation of Arctic resources. Now lets look at some of the temperature curves in this paper. What strikes me most is their choice of scale. The last part of the curve shows an abrupt drop at the edge of the graph. Knowing that the warming started at the turn of the twentieth century it would be important to see what happened before and after the start of warming. Which means the last two centuries but that is impossible with their compressed scale. And this is probably not from ignorance but a deliberate choice because he does list Kaufman's paper among his references. Another peculiarity is that more ice loss takes place during the Little Ice Age than during the Medieval Warm Period. This is simply contrary to expectations. The small scale sea ice graph in Figure 2c shows a drop in sea ice after 1970 which I can believe because 1970 was the year when the warming resumed after a pause. But I do not see an equal drop between 1900 and 1940 when the warming was equally intense. Their hypothesis is that their proxy network contains an embedded climate signal specifically related to sea ice variability. To me the reversal of expectations for LIA and MWP means that this hypothesis is not working too well. I personally would not want to use their data for anything important to do with the Arctic.

    1 Arno Arrak, "Arctic Warming is not greenhouse warming," E&E 22(8):1069-1083 (2011)
    [2} Darrell S. Kaufman, Davud P. Schneider, Nicholas P. McKay, Caspar M. Ammann, Raymond S. Bradley, Keith Briffa, Jonathan T. Overpeck, Bo M. Vinther, Arctic Lakes 2K Project Members, "Recent Warming Reverses Long Term Arctic Cooling" Science 325:1236-1239 (4 September 2009)
    3 Robert F. Spielhagen, Kirstin Werner, Steffen Aagaard Sorensen, Katarzina Zamelczyk, Evguenia Kandiano, Gereon Budeus, Katrine Husum, Thomas M. Marchitto & Morten Hald, "Enhanced Modern Heat Transfer to the Arctic by Warm Atlantic Water, " Science, 331:450-453 (28 January 2011)

Subscribe to comments

Additional data