Hi folks.
The situation, as I understand it anyway, is that the positive warming from CO2 and other long-lived radiative trace gases (e.g., methane) cannot account for the observed warming in the arctic. In other words, the polar amplification seen is much larger than current models can account for using only CO2. The explanation for this follows a couple of different tracks, depending on whether you are a climate skeptic or not. There is some overlap between these since the atmosphere-ocean dynamics of the arctic are poorly understood, especially in the case of coupling between the arctic and known decadal processes like the North Atlantic Oscillation and the Pacific Decadal Oscillation.
Track 1: It's all natural variability. Skeptics will argue that a shift in the NAO or PDO from one phase to the other shifts circulation in the arctic and that has caused the sea ice to melt. There are NASA websites around with articles by Jamie Morison detailing how this might work. There are however, arctic scientists who do agree with the conclusions of the IPCC that also believe dynamics might be at least partly responsible for the warming (and loss of sea ice) in the arctic (Cecelia Bitz is, I think, an example of this and her article on RealClimate.org is excellent and highly recommended, at least by me).
Track 2: There is some other feedback going on that is warming the arctic that isn't accounted for by greenhouse forcing from CO2. If you search the website from Atmos. Chem. and Physics for "P. K. Quinn" you can find a paper detailing what this entails. In a nutshell, current climate models don't account for warming from black carbon (i.e., soot) or from short-lived greenhouse gases like ozone. Atmos. chemists like Quinn et al. know that Europe, Russia, and N. America to a lesser extent, deliver these to the arctic. The warming from these might be enough to provide the missing forcing since black carbon has a huge effect on albedo.
Anyway, the thing that is interesting from my perspective is this is a huge hole in climate models. The one thing that all models predict and that has been observed is a polar amplification, especially in the arctic. Yet the models crater severely reproducing the extent of the observed warming. This is a very active area of research, and going here:
http://www.agu.org/meetings/fm08/waisfm08.html
and searching for "arctic warming" returns 40 sessions (not papers, but sessions containing anywhere from 10 to 35 individual papers) discussing arctic warming. Skeptics haven't picked up on this model failure and beat the everloving snot out of it simply because the professional skeptics don't want to do point it out because they would have to admit the arctic is warming and the ditto-heads on the internet who parrot what the Balls and Lindzens say don't understand the problem.
My personal opinion is that Quinn et al. (and I might be a little biased here) are on the right track and that dynamics has less to do with this than radiative forcing. If it were all circulation, the interior of Greenland wouldn't be melting like it is. But I can't back that with data.
-Bill
gcnp = Got Creek, No Paddle
References:
Bitz: http://www.realclimate.org/index.php/archives/2007/01/arctic-sea-ice-decline-in-the-21st-century/
Morrison news release: http://www.engr.utexas.edu/news/articles/200711131355/index.cfm (not sure if Morison put this in print yet)
Quinn et al. http://www.atmos-chem-phys.net/8/1723/2008/acp-8-1723-2008.pdf