We examine the recent (1979–2010) and future (2011–2100) characteristics of the summer Arctic sea ice cover as simulated by 29 Earth system and general circulation models from the Coupled Model Intercomparison Project, phase 5 (CMIP5). As was the case with CMIP3, a large inter-model spread persists in the simulated summer sea ice losses over the 21st century for a given forcing scenario. The initial 1979–2010 sea ice properties (including the sea ice extent, thickness distribution and volume characteristics) of each CMIP5 model are discussed as potential constraints on the September sea ice extent (SSIE) projections. Our results suggest first that the SSIE anomalies (compared to the 1979–2010 model SSIE) are related in a complicated manner to the initial 1979–2010 sea ice model characteristics, due to the large diversity of the CMIP5 population (at a given time, some models are in an ice-free state while others are still on the track of ice loss). In a new diagram (that does not consider the time as an independent variable) we show that the transition towards ice-free conditions is actually occuring in a very similar manner for all models. For these reasons, some quantities that do not explicitly depend on time, such as the year at which SSIE drops below a certain threshold, are likely to be constrained. In a second step, using several adequate 1979–2010 sea ice metrics, we effectively reduce the uncertainty as to when the Arctic could become nearly ice-free in summertime (between 2041 and 2060 for a high climate forcing scenario)

[1] We present the first analysis of snow depth on Arctic sea ice in the Coupled Model Intercomparison Project 5 (CMIP5) because of its importance for sea ice thermodynamics and ringed seal (Phoca hispida) habitat. Snow depths in April on Arctic sea ice decrease over the 21st century in RCP2.6, RCP4.5, and RCP8.5 scenarios. The chief cause is loss of sea ice area in autumn and, to a lesser extent, winter. By the end of the 21st century in the RCP8.5 scenario, snowfall accumulation is delayed by about three months compared to the late 20th century in the multi-model mean. Mean April snow depth north of 70 N declines from about 28 cm to 16 cm. Precipitation increases as expected in a warmer climate, but much of this increase in the Arctic occurs as rainfall. The seasonality of snowfall rate grows, with increasing rates in winter and decreasing rates in summer and autumn, but the cumulative snowfall from September to April does not change. Ringed seals depend on spring snow cover on Arctic sea ice to create subnivean birth lairs. The area with snow depths above 20 cm -a threshold needed for ringed seals to build snow caves -declines by 70%. Citation: Hezel, P. J., X. Zhang, C. M. Bitz, B. P. Kelly, and F. Massonnet (2012), Projected decline in spring snow depth on Arctic sea ice caused by progressively later autumn open ocean freeze-up this century, Geophys.

Under greenhouse warming, climate models simulate a weakening of the Atlantic Meridional Overturning Circulation and the associated ocean heat transport at midlatitudes but an increase in the ocean heat transport to the Arctic Ocean. These opposing trends lead to what could appear to be a discrepancy in the reported ocean contribution to Arctic amplification. This study clarifies how ocean heat transport affects Arctic climate under strong greenhouse warming using a set of the 21st century simulations performed within the Coupled Model Intercomparison Project. The results suggest that a future reduction in subpolar ocean heat loss enhances ocean heat transport to the Arctic Ocean, driving an increase in Arctic Ocean heat content and contributing to the intermodel spread in Arctic amplification. The results caution against extrapolating the forced oceanic signal from the midlatitudes to the Arctic.

[1] Methanesulfonic acid (MSA) has previously been measured in ice cores in Antarctica as a proxy for sea ice extent and Southern Hemisphere circulation. In a series of chemical transport model (GEOS-Chem) sensitivity experiments, we identify mechanisms that control the MSA concentrations recorded in ice cores. Sea ice is linked to MSA via dimethylsulfide (DMS), which is produced biologically in the surface ocean and known to be particularly concentrated in the sea ice zone. Given existing ocean surface DMS concentration data sets, the model does not demonstrate a strong relationship between sea ice and MSA deposition in Antarctica. The variability of DMS emissions associated with sea ice extent is small (11-30%) due to the small interannual variability of sea ice extent. Wind plays a role in the variability in DMS emissions, but its contribution relative to that of sea ice is strongly dependent on the assumed DMS concentrations in the sea ice zone. Atmospheric sulfur emitted as DMS from the sea ice undergoes net transport northward. Our model runs suggest that DMS emissions from the sea ice zone may account for 26-62% of MSA deposition at the Antarctic coast and 36-95% in inland Antarctica. Though our results are sensitive to model assumptions, it is clear that an improved understanding of both DMS concentrations and emissions from the sea ice zone are required to better assess the impact of sea ice variability on MSA deposition to Antarctica.