Is so much easier if you can take advantage of expert opinion. So I did some back-of-the-enveloping, then decided to ask someone who would know much more than me.

Who best to ask? WTF: I emailed Mark Serreze the following enquiry:

I have noticed that Bill Chapman’s site at CT/UIUC is currently showing a negative areal anomaly of more than -2.7 Million Km2, which as far as I can ascertain is the largest recorded anomaly for this metric in satellite records.

Can you tell me if the same is true for your method of calculating extent, i.e., if the present anomaly (for the past 5 days, for example) is now larger than it was during the week of minimum extent, Sept. 16th?…

…I am also curious to know how far you think the anomaly might go before next April; is there any chance we might see a total areal anomaly of -3.5Mkm2, or even more?

Never let it be said that scientists aren’t good, decent considerate folk. Heck, these guys have their jobs to do, as well as handle the press, and could do without dumb emails from the likes of the Old man. So I wasn’t surprised when Mark replied, though I was extra pleased by the bonus email from Walt Meier. Here are a couple of extracts from their replies:


… in terms of ANOMALIES, yes, I would not be surprised if we still had a record, both in terms of ice area and extent. Check our last posting, you will see that extent is still way below normal.

Walt: do you have the numbers handy in terms of current ice extent anomalies?

Ice extent and area are increasing only slowly this autumn, the reason being that there is still so much heat in the Arctic Ocean. Will anomalies grow through this winter? I just don’t know. However, the ice we have by April will likely be pretty thin, setting us up for another big loss next summer.

Mark C. Serreze

And Walt replied, too:

Hi Fergus,

Yes, we have been tracking the extent anomalies relative to the climatology (1979-2000) and indeed it did continue to grow well after the minimum extent was reached on Sept. 16.

In fact, it reached as much as -3.25 million sq km about a week ago. Since then it has moderated some, but it’s still nearly -3 million sq km.*

This isn’t too surprising. There was such a huge area that was ice-free, allowing the ocean to warm up. This means it’s going to take longer to cool to the freezing point allowing ice to form. I would expect that the anomaly magnitiude will decrease through the winter**, but may still stay larger than the previously anomalies of ~1 million sq km we’ve seen the past couple of years. Also, as Mark mentions below, if where there is ice, it will be thinner.


* a quick technical point; the numbers I use are for the CT ‘area’ metric, Walt’s are the NSIDC ‘extent’ anomaly numbers.

**walt emailed to mention the typo, where he put ‘summer’ rather than ‘winter’; it has now been adjusted.

So, the anomaly did increase after the minimum was reached, by both measures. CT is still going downwards, though this could change within a week; to answer a potential disagreement on the earlier thread, my ‘eye’ estimate reads the ‘tale of the tape’ showing an areal anomaly of -2.75 Mkm2 +/- 0.05.

From this, plus my unique method of pseudo-straight-line analysis, I will now estimate the forthcoming seasonal mean extents for Autumn (October-December) and Winter (January-March). This can be compared against the fourth of the top-line graphs at CT – the one with the pretty colours.

The OND mean extent will probably not exceed 10Mkm2, and is more likely to be as low as 9.3Mkm2, +/- 0.2.

JFM should be below 13.5; my guess is 12.8 Mkm2 +/- 0.4. The mean annual extent, calculated at the end of Autumn, will be below 11 Mkm2, perhaps as low as 10.2.

I couldn’t work out how to estimate the Spring mean.

This sets the scene nicely for the bets between William, Eli and the others over at Stoat, for next Summer’s minimum. There is a very good chance that the maximum in March will be down by more than 1 Mkm2, and it could be as much as 1.5. The ice in April will be thin, and will melt or dissipate more rapidly as a result. Next year’s minimum might be very close to this year’s all-time record decline. I don’t know which side of the line it will fall, but I’ll hazard that it will be considerably lower than the 2005 (former) record minimum.

If anyone thinks my estimates are likely to be out by more than 1 Mkm2 either way, I’m willing to consider a small wager…

My sincere thanks go to Mark and Walt, and the excellent people who work at both the NSIDC and the UIUC to produce the data; you’re all heroes.