Besides, the linear fit is scary enough as it is. But you’re right about “dropping off the trendline”. The rate at which the summer ice is melting is not only increasing, the rate at which it is increasing is increasing even faster.

On the NSIDC archive there are monthly graphs like this for every single month of the year, for every year. The regression drops to the right on every single one, and the slope of the line tends to trend farther down as more data is collected.
It is flatter in winter, but it is the summer months where it is most striking.

These are also the months where the physical significance is magnified. The more ice melts, the more sunlight is absorbed by the sea, the longer will autumn freezing be delayed, the thinner the ice gets and the more severe next year’s thaw will be. I guess you could call it an, ahem, snowball effect.

It’s positive feedback, and it will stabilize only when the thermosphere undergoes a change of state to some new level of equilibrium. How that new equilibrium will translate to overall global climate remains to be seen, but I submit the climactic anomalies we have witnessed this year in N America are a credible predictor of things to come.

Here is last year’s (2011) graph for the minimum ice month, September (the September graph is published early in October). When it is published this year, a 2012 data point will be added, and the new trendline will drop even faster. The slope of the linear regression line calculated for Sep 2011 was -12%/decade, and is calculated relative to the 1979 to 2000 average, when the melting was more moderate. This year’s record melt will pull it down further. Note that the previous record of 2007 was smashed just a few days ago, and the melt continues, perhaps for another month.

Based on the numerical data here alone, coupled with the most conservative assumptions, some climatolgists project the first totally ice-free arctic summer to about 2030. As long as some ice remains in the arctic at summer’s end, a damper in the overall warming trend is still in place, and the sudden shift to a new equilibrium is delayed. When the summer ice vanishes, all bets are off.

This new equilibrium may be stable, but it will be different than current conditions. I do not believe the world will come to an end–after all, it has never done so in the past. But it will change and it will change rapidly, perhaps faster than we can economically and politically adapt to it. It may not happen in my lifetime, but it will happen in the lifetime of people alive today. If you have children, they will see it, and they will face the consequences.