When I was a wee mustelid, I wrote about this; and there is an RC piece, and as far as I know, its all still valid.
But this post is aimed at those who put an “s” into poles in “What exactly is the mechanism that causes the poles to warm faster than the tropics as a result of climate change?” when they mean the real world, as opposed to a simplified or equilibrium one. Mind you, since the ar4 sez Models of the 21st century project that future warming is amplified at high latitudes resulting from positive feedbacks involving snow and sea ice, and other processes (10.3.3.1) you can hardly blame anyone else for being confused. Traditionally, the IPCC doesn’t pay much attention to Antarctica, and we help them by not publishing :-)
Because, you see, there is only amplified warming at the North Pole; and this is what is predicted. I’m speaking now about the bulk of Antarctica, rather than just the excitingly overheating Peninsula. And by “predicted” I mean, comes out of climate models.
The first plot is the zonal mean of a difference of a 2090-2095 mean and a 2005-2010 mean from the IPCC ar4 sresa1b set, annual mean sfc T. The data have been somewhat crudely processed so don’t publish based on this. Black lines are full zonal means, and the thick blue is the average (including the cr*ppy models). Red lines are the land-only means (which is why they have a gap at 60S and at the NP).
Ve zee zat zee North Pole warms nearly 6 oC (all now taken from the avg) and the South 3 oC. Looking at the land-only (red) we see that, apart from the northern high latitudes, most latitudes warm by about 3 oC and there is no SH polar amplification.
Looking at a map of the difference of the average…
It becomes clear that, away from the NP, land warms 2-3 oC, ocean 1-2 oC. The TAR has a similar pic. The commonplace explanation is that the oceans are a nice heat sink and hence delay warming (see-also Rowan Sutton) and there is a nice big ocean all the way around Antarctica.

