Geoscience Reference
In-Depth Information
Figure 5. Time series of September ice extent for the (a) run 1 ensemble member and (b-f) sensitivity runs listed in Table
2. The 5-year running mean smoothed ice extent from run 1 is shown as the dotted line in Figures 5b-5f. The thin black
line indicates the September extent, and the bold black line is the 5-year running mean time series. Abrupt events are
indicated by the grey shading.
Our results suggest that the abrupt ice loss events simu-
lated by CCSM3 do not result from threshold or “tipping
point” behavior. However, as mentioned above our inability
to identify a critical state resulting in rapid ice loss does not
necessarily preclude one. In particular, the definition of a
critical threshold is somewhat subjective given the complex-
ity of the system. Additionally, only the sea ice state has
been considered in this analysis. A threshold could instead
exist in other climate variables such as ocean conditions or
snow cover. Regardless, the changing aspects of the Arc-
tic sea ice cover indicated by this study, including increas-
ing summer extent variability and the potential for rapid
change, could have profound consequences beyond the climate
system.
The intrinsic (natural) variability in the September ice
extent increases as the ice thins because the natural vari-
ations in both dynamic and thermodynamic forcing can
more easily produce open water variations. With only a thin
ice pack, large regions of the ice cover can be completely
melted through or easily converged resulting in a large ice
area change. Considerable increases in ice cover still read-
ily occur though since only a thin layer of sea ice needs to
Search WWH ::




Custom Search