Will Climate Change Affect Snow Cover Over North America?

Gavin Gong (Assistant Professor, EEE)

Allan Frei (Associate Professor, Hunter College)


Continental-scale snow cover extent (SCE) is a potentially sensitive indicator of climate change, since it is an integrated measure of multiple hydroclimatological processes, and it is the most prominent seasonal landsurface feature in the extratropical Northern Hemisphere. Conversely, feedback processes may cause SCE changes to in turn affect the direction and magnitude of other climate changes across the globe. In this study, current and future decadal trends in winter North American SCE (NA-SCE) are investigated, using nine general circulation models (GCMs) of the global atmosphere-ocean system participating in the upcoming Intergovernmental Panel on Climate Change Fourth Assessment Report (IPCC-AR4).

Figure 1. Annual time series (thin line), overlaid with nine-year running means (thick line), of January NA-SCE, including both 20th century (20C3M) and 21st century (SRESA1B) scenarios, for nine GCMs. Results are shown in two panels to aid in visualization.

Simulated annual timeseries of January NA-SCE spanning the 20th and 21st centuries are shown for all nine models in figure 1, based on a set of current and hypothesized future socioeconomic developments with associated greenhouse gas emissions, designated the 20C3M and SRESA1B scenarios, respectively. Most models do not exhibit a 20th-century trend, and significant between-model variability is apparent, with most models underestimating the observed NA-SCE over the 20th century. This exemplifies the considerable uncertainties that still plague GCM simulations. Nevertheless, all nine models exhibit a clear and statistically significant decreasing trend in 21st century NA-SCE, although the magnitude of the trend varies between models.

For five of the models, two additional 21st-century scenarios were simulated. The SRESA2 scenario represents accelerated greenhouse gas emissions compared to SRESA1B, while the COMMIT scenario unrealistically assumes that emissions are maintained at year 2000 levels throughout the 21st century. Figure 2 shows that though trend magnitudes again vary between models, in all cases the decreasing trends for SRESA1B and SRESA2 are statistically significant and comparable in magnitude to each other. In addition, they both decrease at greater rates than under the COMMIT scenario. For all models, the SRESA1B and SRESA2 decreasing 21st-century trends are statistically distinguishable from 20th-century conditions. In contrast, 20th- and 21st-century values are statistically indistinguishable under the COMMIT scenario.



Figure 2. Annual time series (thin line), overlaid with nine-year running means (thick line), of January NA-SCE for 20th- and 21st-century scenarios from five GCMs. 20C3M, COMMIT, SRESA1B and SRESA2 scenarios denoted by black, red, blue and green lines, respectively.

These results show that climate change associated with increased greenhouse gas emissions may indeed affect future snow cover extent over North America. Additionally, the response to greenhouse gas forcing may be nonlinear: perhaps there is a threshold in emission rate or concentrations above which increased emissions result in no additional snow trend. The predicted decrease in this integrated hydroclimatological metric suggests a strong and persistent climate change response at the land surface, which could have considerable impacts of water supply, ecological cycles, recreation, and subsequent climate change feedbacks.

Frei, A. and G. Gong, 2005. Decadal to Century Scale Trends in North American Snow Extent in Coupled Atmosphere-Ocean General Circulation Models. Geophysical Research Letters, 32:L18502, doi: 10.1029/2005GL023394.

500 W. 120th St., 918 Mudd, New York, NY 10027    212-854-2905                
©2012 Columbia University