So, the sea ice minimum occurred on September 13 with an extent of 4.636 million square kilometers. My extremely way too early forecast on March 22 was for 3 million square kilometers, which I then doubled down on with my way too early forecast on July 16. The actual extent was about 55% more than that, which illustrates why you don't put much weight on way too early forecasts. The extent was at record lows through June, but cold weather late in the summer saved us from a new record minimum extent.
While this was one time I was glad to be wrong, we shouldn't get too comfortable with these results. They are still bad. This minimum was low enough to rank as the eighth-worse, but that doesn't tell the story. The worst minimum is still the one recorded in 2012 at 3.387 million square kilometers. But, the next seven minimum extents were between 4.137 million square kilometers (2016) and 4.636 million square kilometers (2017). (In order, the other years were: 2007, 2011, 2015, 2008, and 2010.) That is a difference of less than 500,000 square kilometers. These minimum sea ice extents are crowding into a narrow range.
The trend is pretty clear. Here is the trend for the September sea ice extent (minimum):
 |
| Source: NSIDC Sea Ice Index |
There is a lot of variability from year to year, but the overall trend is not only unmistakable, but alarming. There is a long-term downward trend, but we see the slope of that trend is more than -13% per decade.
Meanwhile, the current sea ice extent, while greater than at this time last year (the worst winter extent recorded), is still in excess of three standard deviations below the 1981 - 2010 average.
So,while it thankfully was not as bad as I had forecast earlier in the year, we can see the Arctic is very sick and that does not bode well for the rest of the planet.
