http://en.wikipedia.org/wiki/Milankovitch_cycles). The southern hemisphere, having much more water, is more stable.
Some what simplistically, let's say we have something that is good at conducting heat through the whole system., but let's say for some reason part of it absorbs heat more easily than the rest (let's say half is painted black and half is painted white and like all things w/ such color differences the black parts absorbs heat better and so is more easily affected).
What happens if you block the black part from absorbing as much heat as it was? The affect is local, but since the heat can be quickly distributed through the system, the effect is also system wide. The thing blocking the black part is local. If you have robust gravity, and you throw it up, it is likely to come down on the black part. The temperature results though are system wide.
Things are more complicated than that because, for example, people do expect Antarctica to be warming faster than it appears to be (we do have issues with measuring how fast Antarctica is warming because of lack of data so how Antarctica is warming (if at all) as a total continent is an area of debate). Partly, though, this is likely because other things are also changing in the southern hemisphere (more so than the northern hemisphere), such as wind and ocean currents.
So we have multiple variables that are changing. I've said in the past, one of the issues with modeling climate change is things keep changing that aren't really related to climate. It is like trying to predict the gas milage of a car where every night somebody comes in and tinkers with the car. Knowing the history of the car, things about the gas put in the car, and the expected driving conditions give you some information, but its very difficult to deal with the other things that are changing in historically unprecedented manners that aren't directly related to the things you are trying to look at, but end up affecting it.