Attribution of projected changes in atmospheric energy transport in the Arctic

We analyze meridonal energy transport into the Arctic derived from one simulation of the National Center for Atmospheric Research Community Climate System Model spanning the periods of 1960-1999, 2010-2030 and 2070-2089. The 21st century simulation incorporates the Intergovernmental Panel on Climate Change (IPCC) Special Report on Emission Scenarios (SRES) A2 scenario for CO2 and sulfate emissions. Modeled and observed (European Centre for Medium-Range Weather Forecasting Reanalysis, ERA-40) sea level pressure fields are classified using a neural-network technique called self-organizing maps to create a set of characteristic atmospheric circulation patterns over the region north of 60°N. Model performance is validated for the 20th century by comparing the frequencies of occurrences of particular circulation regimes in the model to those from the ERA-40. The model successfully captures dominant SLP patterns, but differs from observations in the frequency with which certain patterns occur. The model's 20th century vertical mean energy transport profile across 70ºN compares well in terms of structure but exceeds observations by about 12% overall. By relating energy transport to a particular circulation regime, future changes in energy transport across 70°N are assessed and attributed to varying atmospheric dynamics and/or thermodynamics. By the late 21st century, the transport is projected to increase by about 21% in this model realization, with the largest contribution (32%) to the total change occurring in summer. Over 75% of the annual increase is due to changes in atmospheric thermodynamics, while dynamics play a secondary role. A larger poleward energy transport likely constitutes a positive feedback on the system through related increases in latent heat release and longwave emission to the surface.