This page has been archived on the Web

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

2011 Literature Review Archives - Climate Trends and Variability

Brown, R. and D.A. Robinson. 2011. Northern hemisphere spring snow cover variability and change over 1922-2010. The Cryosphere, Vol 5, D16111, doi:10.1029/2010JD013975.

An updated record of Northern Hemisphere spring snow cover extent (SCE) reveals significant declines in spring SCE over the period 1922-2010.  Long-term warming of spring temperatures over the mid-latitudes is the dominant cause of the observed trend.

Scientists from Environment Canada and Rutgers University have compiled a record of Northern Hemisphere spring (March, April) snow cover extent (SCE) for the period 1922-2010.  This new record is based on multiple datasets (snow cover data from visible and microwave satellite observations, surface snow depth observations and reconstructed snow cover) following the technique used by Brown and colleagues to develop a similar but shorter record for the Arctic last year.  This record updates the SCE record used in the Fourth IPCC Assessment Report and provides improved uncertainty estimates.  The uncertainty estimates are larger for Eurasia than North America and in both cases they are greatest prior to 1967 (i.e. before the satellite era) and decrease thereafter as more and improved datasets became available.  As a whole, there has been a significant reduction in Northern Hemisphere spring snow cover extent over the past ~90 years and the rate of reduction has been most rapid over the past 40 years.  The rates of reduction in NH SCE calculated over the 1922-2010 period are estimated to be -3.24 x 106 km2 per 100 years for March and -4.72 x 106 km2 per 100 years for April.  The NH SCE trends are driven primarily by increasing spring air temperatures over the mid-latitudes.  For Eurasia alone the decline is evident in both March and April whereas for North America a significant reduction in SCE is only observed for April.  The significant decrease in March snow cover extent over Eurasia appears to also be related to changes in regional circulation patterns.


Csank, A.Z., A.K. Tripati, W.P. Patterson, R.A. Eagle, N. Rybczynski, A.P. Ballantyne, J.M. Eiler, Jand M. John. 2011. Estimates of Arctic land surface temperatures during the early Pliocene from two novel proxies. Earth and Planetary Science Letters, Vol 304, pp. 291-299, doi.org/10.1016/j.epsl.2011.02.030.

 The early Pliocene (3.5 - 4 million years ago) was the last sustained period of time when the Earth was warmer than present and is therefore considered a good analog for 21st century climate change.  New paleo-temperature records indicate that over this period, prolonged atmospheric carbon dioxide concentrations of between 365-415 ppm resulted in Arctic temperatures 11-16oC warmer than present.

The climate of the Pliocene (~2.6 to 5 Ma ago) is often considered to be an example of climate conditions that could be associated with prolonged, elevated levels of atmospheric CO2.  Over this period atmospheric concentrations of carbon dioxide (CO2) are estimated to have been between ~365-415 ppm (present level is ~390ppm) and the configuration of continents was similar to today.  Model-based studies suggest that global temperatures during the Pliocene were 2-3oC warmer than pre-industrial but that Arctic temperatures were 11-17oC warmer than today.  To better constrain the magnitude of the Arctic amplification over the early Pliocene, Csank and colleagues develop proxy temperature records for the Arctic.  Terrestrial growing-season temperature proxies for the early Pliocene are inferred from oxygen isotope data from peat and mollusc shells collected at a site on Ellesmere Island in the Canadian Arctic.  The results (obtained using two different techniques) indicate that temperatures were between 11-16oC warmer than present (in agreement with model-based and other paleo-estimates) and that the Arctic Ocean was likely seasonally ice free.


Gagen, M., E. Zorita, D. McCarroll, G.H.F. Young, H. Grudd, R. Jalkanen, N.J. Loader, I. Robertson and A. Kirchhefer. 2011. Cloud response to summer temperatures in Fennoscandia over the last thousand years. Geophysical Research Letters, Vol 38, L05701, doi: 10.1029/2010GL046216.

Reconstructed changes in cloudiness during warm and cool periods of the last millennium are found to support a negative cloud feedback (i.e. a dampening effect) on climate change on multi-decadal timescales in a northern boreal region.  At long timescales, increased (reduced) cloudiness decreases (increases) the amount of sunshine received by the Earth’s surface, contributing a cooling (warming) effect. This result is in line with climate model projections for high-latitudes in response to continued warming over the 21st century.

The magnitude, sign and spatial variability of the cloud response to increasing global temperatures remains one of the greatest sources of uncertainty in future climate projections.  Attempts to better constrain the cloud response are impeded by the limited length of the observational records of both sunshine duration and cloud cover.  Gagen et al. developed a proxy-record of mean summer near-ground solar radiation (inversely related to cloud cover) for Fennoscandia for the past millennium to evaluate the cloud response to pre-instrumental temperature anomalies.  The record of summer near-ground solar radiation was derived from a stable carbon isotope record extracted from a precisely dated Scots pine tree-ring chronology (AD 866-2001) sampled in the northern boreal forest zone of Finnish Lapland.  In this region, the concentration of CO2 in conifer needles relates most strongly to the rate of photosynthetic assimilation of carbon as controlled primarily by growing season sunlight.  The stable carbon isotope series shows a strong relationship with a local record of sunshine duration available for the latter half of the twentieth century.  Comparison of the mean summer insolation reconstruction with an independent tree-ring based reconstruction of summer temperatures developed previously for the same region reveals that, in this region, the warmest centuries of the medieval period were cloudy and the coolest centuries of the ‘Little Ice Age’ were sunny.  The authors note that this response is indicative of a negative shortwave cloud feedback at multi-decadal timescales at this latitude.  These findings are in line with studies that suggest that future warming will result in a negative cloud feedback (i.e. higher temperatures are expected to increase evaporation and therefore cloudiness ultimately resulting in less incoming shortwave radiation at the surface than at present) at high-latitudes.  The authors conclude that this technique for reconstructing past sunshine duration could be attempted in other regions (e.g. the tropics where a positive cloud feedback is projected for the future).


Gardner, A.S., G. Moholdt, B. Wouters, G.J. Wolken, D.O. Burgess, M. J. Sharp, J. G. Cogley, C. Braun and C. Labine. 2011. Sharply increased mass loss from glaciers and ice caps in the Canadian Arctic Archipelago. Nature. Published online April 20, 2011. doi:10.1038/nature10089.

The rate at which ice is being lost from glaciers and ice caps in the Canadian Arctic Archipelago is shown to have increased sharply in recent years, largely in response to warmer summer air temperatures.

The Canadian Arctic Archipelago (CAA) region has experienced some of the warmest summers on record recently, with four of the five warmest years since 1960 occurring since 2004. A scientific team, that included a number of Canadians, has investigated changes in the amount (mass) of ice in glaciers and ice caps of the CAA over the period 2004-2009. They used three independent methods for estimating changes in glacier mass balance in the northern CAA (surface mass budget modeling + ice discharge; repeat satellite laser altimetry (ICESat), and repeat satellite gravimetry (GRACE); the latter two methods were also applied to the southern CAA.  All three approaches show consistent and large ice mass losses which, when averaged, give an estimate of total mass loss for the CAA of 368 ± 41Gt for the years 2004-2009, or 61 ± 7 Gt/yr.  Between the periods 2004-2006 and 2007-2009, the rate of mass loss sharply increased from 31 ± 8 Gt/yr to 92 ± 12 Gt/yr  and they estimate that three-quarters of all mass loss over the 6 year observation period occurred in these last three years. The majority (92%) of the ice loss was from meltwater runoff, with a much smaller contribution from icebergs (calving off from glaciers terminating in the sea). These changes are found to be primarily a response to warmer summer temperatures. Although the observation period is short, when comparing results to other major regions with significant small ice caps and glaciers (i.e. excluding Greenland and Antarctica), the authors conclude that the Canadian Arctic Archipelago region has been the single largest contributor to sea level rise of over the period 2007-2009.


Li, J, S-P. Xie, E.R. Cook, G. Huang, R. D’Arrigo, F. Liu, J. Ma and X-T. Zheng. 2011. Interdecadal modulation of El Niño amplitude during the past Millennium. Nature Climate Change, Vol 1, p. 114-118, doi: 10.1038/nclimate1086.

 A long-term record of El-Niño/Southern Oscillation (ENSO) variability identifies cycles of 50-90 years in ENSO intensity that appear to be linked with long-term changes in sea surface temperatures in the eastern-central tropical Pacific.  As ENSO events of greater intensity are observed during warm periods, this suggests that increased sea surface temperatures in this region in the future may lead to more intense ENSO events.

 The El-Niño/Southern Oscillation (ENSO) phenomenon (an alternation between anomalously warm (El Niño) and cool (La Niña) conditions in the eastern tropical Pacific at 2-8 year intervals) is the greatest source of year-to-year climate variability on Earth.  Variations in the intensity of ENSO events have a large impact on the occurrence and predictability of climate extremes around the world.  The ability to detect and predict changes in ENSO intensity is limited because the instrumental record is too short to characterize its natural variability. Li and colleagues describe the development of the longest, continuous, annually-resolved reconstruction of past ENSO activity developed to date.  The record is based upon a network of drought reconstructions derived from a dense collection of North American tree-ring chronologies. The relationship between the instrumental record of ENSO events and the paleo-drought data is used to develop a 1,100 year reconstruction of ENSO history.  Over the past millennium, the intensity of ENSO events shows significant quasi-regular cycles of varying length (50-90 years).  The analysis further showed that during past warm (cool) periods in the Eastern tropical Pacific, ENSO events (El Niño and La Niña) were typically more (less) intense.  The authors speculate that if SSTs in the eastern-central Pacific continue to increase with anthropogenic global warming, they may lead to more intense ENSO events with more extreme impacts.  The authors conclude that this new reconstruction provides observational constraints for improving model-based projections of future ENSO behaviour.


Pederson, G.T. S. T. Gray, C.A. Woodhouse, J. L. Betancourt, D. B. Fagre, J. S. Littell, E. Watson, B. H. Luckman, and L. J. Graumlich. 2011. The unusual nature of recent snowpack declines in the North American Cordillera. Science Express. Published online 9 June 2011. doi: 10.1126/science.1201570.

 Recent declines in snowpack, in a region of the mountainous west extending from southern B.C to the northern U.S., are shown to be almost unprecedented in over 1000 years. The spatial pattern of change identified over a broader longitudinal range also indicates that in response to regional warming, a fundamental shift has likely occurred from precipitation to temperature as the dominant control of snowpack.

 The Columbia, Colorado and Missouri Rivers are the primary water source for >70 million people and 60-80% of the water in these basins originates as snowpack.  An understanding of the range of past snowpack variability, particularly at decadal to multidecadal timescales, and how this is linked to variations in precipitation and temperature is important for considering how water resources may change in response to future regional climate change. Pederson et al develop a set of annually resolved, tree-ring based snowpack reconstructions for 27 sub-watersheds in the Colorado, Columbia and Missouri River headwaters.  The reconstructions, which range in length from 500-1000+ years, were developed from a network of 66 tree-ring chronologies sampled at sites along the Cordillera (from B.C. to New Mexico) where growth is conditioned by the amount of water available from the previous winter’s snowpack.  The reconstructions reveal that recent snowpack decreases in the northern part of the study region are of almost unprecedented magnitude.  Furthermore, a fundamental shift in the pattern of snowpack declines across the region was evident. Over the past millennia (with only a few exceptions) there has been a north-south contrast in snowpack anomalies across the region: when the north was wet (more snow) the south was dry (less snow) and vice versa.  The authors relate this pattern to changes in the location of the winter storm track (and hence precipitation delivery) which in turn is linked to shifts in patterns of sea surface temperatures in the Pacific Basin.  Since the 1980s, the snowpack decline has been synchronous across the study region indicating a breakdown in the prevailing pattern.  This likely indicates a shift in the dominant driver of the snowpack from precipitation to temperature related to the recent warming which may have fundamental implications for future water supplies in the mountainous west.


Rignot, E., I. Velicogna, M.R. van den Broeke, A. Monaghan and J. Lenaerts. 2011. Acceleration of the contribution of the Greenland and Antarctic ice sheets to sea level rise. GRL 38, L05503, doi:10.1029/2011GL046583.

 Study shows that the Greenland and Antarctic ice sheets have been losing mass at an accelerating rate over the past two decades and may soon become the dominant contributor to global sea level rise.

Projections of future sea level rise account for contributions from the thermal expansion of warming ocean waters and from land-based ice masses (glaciers, small ice caps and the large polar ice sheets). Although small ice gaps and glaciers have contributed more to observed sea level over past decades than the ice sheets have, it is expected that in the future, loss of ice from the ice sheets will become dominant. Rignot and colleagues provide a basis for improving future estimates of the latter by developing refined estimates of the contributions from both the Greenland and Antarctic ice sheets to recent observed sea level rise. Two independent methods for determining mass balance (mass budget method and gravity method) were evaluated over a common observation period (2002-2009). Excellent agreement between the methods is shown, for both ice sheets, for changes in total ice sheet mass balance, with both methods showing accelerating rates of mass loss. On the basis of this agreement, the longer 18-year record, developed using the mass budget method, is used to provide a trend in acceleration of ice mass loss of 21.9 ± 1Gt/yr2 for Greenland and 14.5 ± 2Gt/yr2 for Antarctica with a combined loss of 36.3 ± 2 Gt/yr2. This means that each year over the 18 year period, the two ice sheets lost a combined average of ~36.3 Gt more than they did the year before. The total contribution from both ice sheets to sea level rise amounted to 1.3± 0.4mm/yr over the period 1992-2009. Other research has demonstrated an accelerated loss of ice mass from glaciers and small ice caps over the last few decades, but this work by Rignot and colleagues demonstrates that the acceleration in loss from the ice sheets is about three times larger than for these smaller ice masses. If this trend continues, ice sheets will become the dominant contributor to sea level rise in the next few decades, well in advance of model forecasts. The authors conclude by discussing some reasons why this trend can be expected to continue and then, using the observed acceleration of ice mass loss of 36.3 ± 2 Gt/yr2, project a contribution from the ice sheets to total sea level rise of 15±2 cm by year 2050 compared to 2009/10, about twice the contributions from each of glaciers and small ice caps, and thermal expansion of the ocean.


Spielhagen, R.F., K. Werner, S. Aagaard, K. Zamelczyk, E. Kandiano, G. Budeus, K. Husum, T.M. Marchitto, M. Hald. 2011. Enhanced modern heat transfer to the Arctic by warm Atlantic water. Science, Vol 331, p. 450-453, doi:10.1126/science.1197397.

 Reconstructed temperatures of Atlantic Water flowing into the Arctic Ocean derived from a marine sediment record indicate that recent temperatures are unprecedented over the past two millennia.

 The relatively warm Atlantic Water that flows northward into the Arctic Ocean via the Fram Strait Branch of the North Atlantic current is the major source of oceanic heat advection to the Arctic and is therefore important to Arctic sea ice distribution and associated climate feedback mechanisms.  Continuous historical temperature records of these waters span only the last ~150 years.  Spielhagen et al. present a ~2000 year marine sediment record from a core drilled on the continental margin of western Svalbard, Norway.  This site lies in the path of the Atlantic Water flow into the Arctic Ocean via Fram Strait.  Changes in species abundance (different species are associated with unique temperature requirements) and Mg/Ca ratios from planktic foraminifers in the marine sediment record were used to generate two independent reconstructions of mid-summer temperatures, with multi-decadal resolution, for the uppermost part of this Atlantic Water layer.  The sediment record reveals a dramatic increase in the abundance of subpolar species since ~1900 reflecting an increased influence of relatively warm Atlantic Water in this region.  The two independent temperature reconstructions show that temperatures of Fram Strait Atlantic Water since ~1850 were ~2oC higher than the average for the preceding two millennia. For the two reconstructions, mean temperatures averaged 3.4OC and 3.6OC before ~1850 and 5.2OC and 5.8OC afterwards, consistent with other observational and paleoclimate series.  The rapid warming of this Atlantic Water over the past ~120 years is unprecedented over the past 2000 years and is amongst a growing body of evidence that Arctic summer temperatures are rising rapidly despite decreases in summer insolation in the Arctic related to orbital forcing.  The authors conclude that the increased temperatures of the Atlantic Water layer flowing into the Arctic Ocean may be a key element in the transition to sea-ice free conditions projected for the Arctic Ocean in the future.


Tivy, A., S.E.L. Howell, B. Alt, S. McCourt, R. Chagnon, G. Crocker, T. Carrieres and J. J. Yackel. 2011. Trends and variability in summer sea ice cover in the Canadian Arctic based on the Canadian Ice Service Digital Archive 1960-2008 and 1968-2008. JGR Vol 116, C03007, 25 pp. Also, Howell, S.E.L., A. Tivy, T. Agnew, T. Markus and C. Derksen. 2010. Extreme low ice years in the Canadian Arctic Archipelago: 1998 versus 2007. JGR Vol 115, C10053, 16 pp.

Two studies, using new long term data sets of sea ice in Canadian Arctic waters, confirm long term declines in summer sea ice cover in response to regional warming and support the expectation of increasingly navigable waters in the future. They also both confirm that in the Canadian Arctic Archipelago, including the Northwest Passage, multi-year ice import from the Arctic Ocean is an important factor in summer ice conditions. As long as this import continues, multi-year ice in the NWP will continue to create potentially hazardous ice conditions as the transition to a summer time sea ice-free Arctic continues.

Much of the analysis to date on trends and variability in Arctic sea ice has used satellite data that extends back to the late 1970s. A team of Canadian scientists, including some with Environment Canada’s Canadian Ice Service and Climate Research Division, have now extended the period available for analysis of long term trends in the Canadian Arctic back to the 1960s, by using records of sea ice cover in the Canadian Ice Service Digital Archive (CISDA), validated for climate studies. The same data set was also used in a separate study, by some of the same authors, to look specifically at the two record low sea ice years in the Canadian Arctic Archipelago (CAA) region. Tivy et al. report that for the period 1968-2008, summer sea ice cover (‘all ice’, capturing both multi-year and first-year ice) has decreased by 11.3% ± 2.6% per decade in Hudson Bay, 8.9% ± 3.1% per decade in Baffin Bay, 5.2% ± 2.4% per decade in the Beaufort Sea and 2.9% ± 1.2% per decade in the CAA. In comparison with published trends based on satellite data for periods beginning in 1979, these trends are somewhat weaker over most of the Canadian Arctic region. This is a reflection of the large inter-annual variability in sea ice cover and the impact of the start year for trend analysis. More important is the robust, four decade long declining trend in summer sea ice cover revealed by this new analysis. The authors show that between 10 and 58% of the year-to-year variability in summer all ice cover in almost all regions is explained by surface air temperature in the preceding spring season, a result that supports the growing expectation of extended open water seasons in the Arctic in response to future regional warming.  Interestingly, for the majority of Canadian Arctic waters, no significant changes were observed in summer multi-year ice, with only a very few exceptions in specific sub-regions. Other studies have documented strong declining trends in multi-year ice in the Arctic Ocean so this result points to the significance of sea ice dynamics, particularly the import of multi-year ice from the Arctic Ocean, for changes in multi-year ice cover in Canadian Arctic waters.  The importance of sea ice import into the Canadian Arctic Archipelago region is reinforced by Howell et al. in examining the conditions that led to the two extreme sea ice minima recorded in 1998 and 2007 in the CAA. Although in both years, strong warming during summer - and in 1998, early Fall - were major factors in the sea ice minima, sea ice dynamics actually contributed to the 1998 minimum by inhibiting replenishment from the Arctic Ocean whereas in 2007, conditions facilitated replenishment, thereby preventing that year’s minimum from falling below the record from 1998.


Date modified: