Science, Vol 310, Issue 5748, 657-660, 28 October 2005

Reports

Role of Land-Surface Changes in Arctic Summer Warming

F. S. Chapin, III,^1* M. Sturm,⁵ M. C. Serreze,⁶ J. P. McFadden,⁷ J. R. Key,⁸ A. H. Lloyd,⁹ A. D. McGuire,² T. S. Rupp,³ A. H. Lynch,¹⁰ J. P. Schimel,¹¹ J. Beringer,¹⁰ W. L. Chapman,¹² H. E. Epstein,¹³ E. S. Euskirchen,¹ L. D. Hinzman,⁴ G. Jia,¹⁴ C.-L. Ping,¹⁵ K. D. Tape,¹ C. D. C. Thompson,¹ D. A. Walker,¹ J. M. Welker¹⁶

A major challenge in predicting Earth's future climate state is to understand feedbacks that alter greenhouse-gas forcing. Here we synthesize field data from arctic Alaska, showing that terrestrial changes in summer albedo contribute substantially to recent high-latitude warming trends. Pronounced terrestrial summer warming in arctic Alaska correlates with a lengthening of the snow-free season that has increased atmospheric heating locally by about 3 watts per square meter per decade (similar in magnitude to the regional heating expected over multiple decades from a doubling of atmospheric CO₂). The continuation of current trends in shrub and tree expansion could further amplify this atmospheric heating by two to seven times.

¹ Institute of Arctic Biology; University of Alaska Fairbanks, Fairbanks, AK 99775, USA.
² U.S. Geological Survey, Alaska Cooperative Fish and Wildlife Research Unit; University of Alaska Fairbanks, Fairbanks, AK 99775, USA.
³ Department of Forest Sciences; University of Alaska Fairbanks, Fairbanks, AK 99775, USA.
⁴ Institute of Northern Engineering; University of Alaska Fairbanks, Fairbanks, AK 99775, USA.
⁵ U.S. Army Cold Regions Research and Engineering Laboratory Alaska, Ft. Wainwright, AK 997030170, USA.
⁶ Cooperative Institute for Research in the Environmental Sciences, University of Colorado, Boulder, CO 803090216, USA.
⁷ Department of Ecology, Evolution, and Behavior, University of Minnesota, Saint Paul, MN 55108, USA.
⁸ National Oceanic and Atmospheric Administration/National Environmental, Satellite, Data, and Information Service, Madison, WI 53706, USA.
⁹ Department of Biology, Middlebury College, Middlebury, VT 05443, USA.
¹⁰ School of Geography and Environmental Science, Monash University, Clayton, Vic 3800, Australia.
¹¹ Department of Ecology, Evolution and Marine Biology, University of California, Santa Barbara, CA 931069610, USA.
¹² Department of Atmospheric Sciences, University of Illinois, Urbana, IL 61801, USA.
¹³ Department of Environmental Sciences, University of Virginia, Charlottesville, VA 229044123, USA.
¹⁴ Department of Forest, Rangeland, and Watershed Stewardship, Colorado State University, Fort Collins, CO 80523, USA.
¹⁵ Palmer Research Station, University of Alaska Fairbanks, Palmer, AK 99645, USA.
¹⁶ Environment and Natural Resources Institute, University of Alaska Anchorage, Anchorage, AK 99501, USA.

^* To whom correspondence should be addressed. E-mail: terry.chapin@uaf.edu

The Arctic provides a test bed to understand and evaluate the consequences of threshold changes in regional system dynamics. Over the past several decades, the Arctic has warmed strongly in winter (1). However, many Arctic thresholds relate to abrupt physical and ecological changes that occur near the freezing point of water. Paleoclimate evidence, which is mostly indicative of summer conditions, shows that the Arctic in summer is now warmer than at any time in at least the past 400 years (2). This warming should have a large impact on the rates of water-dependent processes. We assembled a wide range of independent data sets (surface temperature records, satellite-based estimates of cloud cover and energy exchange, ground-based measurements of albedo and energy exchange, and field observations of changes in snow cover and vegetation) to estimate recent and potential future changes in atmospheric heating in arctic Alaska. We argue that recent changes in the length of the snow-free season have triggered a set of interlinked feedbacks that will amplify future rates of summer warming.

Summer warming in arctic Alaska and western Canada has accelerated from about 0.15° to 0.17°C decade^1 (19611990 and 19661995) (1, 3) to about 0.3° to 0.4°C decade^1 (19612004; Fig. 1). There has also been a shift from summer cooling to warming in Greenland and Scandinavia, more pronounced warming in Siberia, and continued summer warming in the European Russian Arctic.

The pronounced summer warming in Alaska cannot be readily understood from changes in atmospheric circulation, sea ice, or cloud cover. Changes in the North Atlantic Oscillation and Arctic Oscillation are linked to winter warming over Eurasia. Variations in the Pacific North American Teleconnection, the Pacific Decadal Oscillation, and El NiñoSouthern Oscillation have strong impacts on Alaskan winter temperatures, but their influences on summer temperatures are comparatively weak (46). There has been a pronounced decline in the extent of summer sea ice, especially north of Alaska and Siberia (1). This implies that solar energy is increasingly augmenting the sensible heat content of the ocean, some of which can then heat the atmosphere over the ocean and adjacent coast (Fig. 2). However, this mechanism fails to explain strong summer warming over interior Alaska (Fig. 1) (7). Further, regional warming trends associated with declining summer sea ice should be more clearly expressed in autumn and winter (8), when much of the additional ocean heat gained in summer will be released back to the atmosphere. The satellite record shows increased summer cloud cover in Alaska (Figs. 2 and 3), similar to patterns described for the circumpolar Arctic (9). The surface cloud radiative forcing in summer over the low-albedo Alaskan land surface tends to be negative, meaning that the decrease in downwelling shortwave radiation to the surface exceeds the increase in the downwelling longwave flux. The consequent reduction in surface net radiation (Fig. 3) would tend to dampen warming resulting from other causes (9).

The summer warming in Alaska is best explained by a lengthening of the snow-free season, causing sensible heating of the lower atmosphere to begin earlier (Fig. 2). Snowmelt has advanced 1.3 days decade^1 at Barrow (coastal), Alaska (10); 2.3 days decade^1 averaged over several (mainly coastal) stations (10); 3.6 days decade^1 in the northern foothills of the Brooks Range (11); 9.1 days decade^1 for the entire Alaskan North Slope [calculated from the satellite data set of Dye et al. (12)]; and 3 to 5 days decade^1 for the region north of 45°N (12). Similarly, spring soil thaw has advanced 2.0 to 3.3 days decade^1 over North American and Eurasian tundra (microwave satellite) (13) and leaf-out date has advanced by 2.7 days decade^1 in Alaska (model estimate) (14) and by 4.3 days decade^1 in North America above 40°N (satellite record) (15). We calculate that the observed snow-melt advance of about 2.5 (1.5 to 3.5) days decade^1 in the Alaskan Arctic increases the energy absorbed and transferred to the atmosphere per decade by about 26 MJ m^2 year^1 [3.3 W m^2 (Table 1)]. This regional decadal change is comparable (per unit of area) to the global atmospheric heating associated with a doubling of atmospheric CO₂, which is projected to occur over multiple decades.

Table 1. Observed changes per decade in summer atmospheric (atmos.) heating (by latent plus sensible heat flux) in Alaskan tundra and potential future changes if arctic tundra were completely converted to shrub tundra or spruce forest. The observed changes are subdivided into changes due to the longer snow-free season and those due to the increased areal extent of shrublands and forest. Also shown is the change in heating associated with a doubling of atmospheric CO2. Cause of change   Atmos. heating   (MJ m-2 year-1)*   (% of total)   (W m-2)   Observed change in atmos. heating over tundra (per decade)             Due to snowmelt advance   25.53   95   3.28       Due to vegetation change                 Shrub expansion   0.59   2   0.08           Forest expansion   0.88   3   0.11       Total change   27.00   100   3.47   Maximum potential change in atmos. heating over tundra             Due to complete conversion to shrubland                 Effect of snowmelt advance   19.48   28   2.51           Effect of shrub expansion   49.50   72   6.37           Total change   68.98   100   8.88       Due to complete conversion to forest                 Effect of snowmelt advance   10.60   5   1.36           Effect of forest expansion   190.80   95   24.54           Total change   201.40   100   25.90   Atmos. heating change caused by doubling of atmos. CO2         ©1998 to 2010 The Heat is Online Maintained by Monochrome