16.6 Cloud forcing and feedback during recent Arctic sea ice loss

Thursday, 21 May 2009: 11:45 AM
Capitol Ballroom AB (Madison Concourse Hotel)
Jennifer Kay, National Center for Atmospheric Research, Boulder, CO; and A. Gettelman and K. Raeder

Recent impressive declines in Arctic sea ice extent provide new opportunities to assess the influence of cloud forcing and feedbacks on sea ice loss in observations and models. Our analysis begins by combining A-train satellite observations with complementary datasets to document Arctic cloud and atmospheric structure during summer and early fall. We focus on 2006-2008, a period over which sea ice extent plummeted to record levels, substantial variability in atmospheric circulation patterns occurred, and spaceborne radar and lidar observations became available. Summer cloud forcing is important because it regulates the ice-albedo feedbacks that accelerate seasonal ice loss. Our results indicate large-scale atmospheric circulation patterns and sea surface temperatures primarily controlled Arctic cloud forcing. Cloud feedback on sea ice extent loss is important because it can amplify or dampen ice loss processes. Our results indicate that the presence of cloud feedbacks varied throughout the melt season. There was no summer cloud feedback on sea ice loss because surface inversions inhibited ocean-to-atmosphere turbulent moisture fluxes. In contrast, an early fall cloud feedback occurred because relatively low static stability permitted upward moisture fluxes and low cloud formation over newly open water. In sum, cloud feedbacks play a minor role in regulating ice-albedo feedbacks, but do help sustain ice extent reductions during fall by enhancing downwelling longwave radiation. The 2007 melt season provides a stunning example of cloud forcing and feedback conspiring with winds to promote record-breaking sea ice loss. During 2007, summer cloud reductions enhanced ice-albedo feedbacks, while early fall 2007 cloud increases over newly open water trapped near-surface heat and may have helped delay ice formation. Using the knowledge gained from our observational analysis, we next examine cloud forcing and feedback in the NCAR/NCEP reanalysis model and in the Community Atmosphere Model (CAM) using a data assimilation framework (DART). We find that the observed cloud feedbacks are not well represented in either assimilation system. We explore reasons for incorrect model cloud feedbacks and the implications for the radiative forcing on projected sea ice loss.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner