I'm focusing on the arctic ice because that is going to happen soon. The difference in temperature if the ice melts will fall drastically changing climate patterns, rainfall, wind, currents.
Antarctica is a different story completely, I don't know much about it, but recently (a few weeks ago) the amundsen Glacier in Antarctica reached it's point of no return surprising a lot of people as that wasn't supposed to happen for decades.
Specifically, the ice extent has been dropping a lot during the last summers. It is noteworthy that it takes a lot of energy loss, and a lot of energy to turn ice to water and vice versa. About as much as three quarters of the way to boiling water. When something like this happens it takes a lot for the water to go from one state to the other.
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm In the summers the arctic receives more sunlight as it is facing the sun, therefore if there is no ice in the summer (or even less ice in the summer) much more heat is absorbed rather than reflected back to space. That is one reason (among others) why the arctic has heated up at an incredible rate, as compared to the rest of the planet.
I'm not a professional in this area however I'm fascinated that such a big change is going to happen so soon. An ice free arctic sea in the summer. We're talking about millions of square kilometers turned from totally reflective to just about the opposite. I'm also a little bit weary of the change as it is a mystery what it will bring.
The range of possibilities here is huge, and possibly the arctic won't be ice free (in the summer) for decades to come, however a simple look at the trends points to a different picture.
Whether the arctic is mostly ice free in 2016, 2020 or 2050, the simple fact that it is lower, by more than 3 million square kilometers, in the summer, when it matters most is a really big deal.
The arctic receives almost all of it's sunlight in the summer, and ice reflects back the heat into space, as opposed to open water.
Go to this site: http://www.ijis.iarc.uaf.edu/cgi-bin/seaice-monitor.cgi?lang=e and put in Septermber 10, 2012, then overlay the historical average data for 1980s, 1990s, 2000sYears like that are likely to be the norm in the future.