By Steven Goddard
The topic this week is verification of data sources. NSIDC reports :
At the end of the month, extent fell near the level recorded in 2006, the lowest in the satellite record for the end of May
That sounds bad! Let’s see how 2010 compares with ten, twenty and thirty years ago :
Then NSIDC discussed my favorite graph from PIOMAS, showing a record low anomaly of Arctic ice volume.
PIOMAS shows an anomaly nearly 50% greater than 2008. If that is correct, then surely we will have a record minimum this summer. However, the PIOMAS data seems inconsistent with PIPS2 data, which shows thicker ice covering a much larger area in 2010. My guess is that PIOMAS is looking at only a small region of the Arctic interior.
Next data source to look at are the Arctic Ice Concentration maps. UIUC has archived lower resolution maps since the start of the satellite record, such as the one below for June 11, 2010.
Their newer maps appear to show a lot more detail.
Note that the newer map shows a lot of low concentration ice in the Beaufort Sea (green, yellow, light pink.) Let’s see how that compares with a current satellite photo.
I must be missing something, because I don’t see much of that low concentration (green) ice in the satellite photo. The two videos below compare satellite vs new and old concentration maps. New comparison first :
The old concentration maps correlate much more closely with the satellite photo.
You can also see in the video below how erratic the bands of yellow and green are in the newer maps. They appear one day and disappear the next.
In the low colour version of the video, it appears that they may be having trouble differentiating between cloud cover and low concentration ice. We see large areas of open water switching to saturated ice overnight, and vice-versa.
It appears that the precision of the newer maps is much greater than their accuracy. Better to use a low precision map which is matched correctly with its accuracy.
Moving on, Barrow Sea Ice is starting to show first signs of thinning. By June 16, 2007 the ice had already broken up.
Temperatures in the Arctic have continued below normal and below freezing for the last few weeks.
NCEP is forecasting below freezing temperatures for much of the Arctic Basin during the next week.
This week, the sun reaches its highest point in the sky. Next week it starts its decline towards winter.
Now, the really interesting stuff. Arctic Basin sea ice volume (calculated from PIPS maps) remains close to the 2006 track, and well above the last three years
Area is slightly below 2006 and 2009.
Thickness is nearly identical to 2006 and 2007.
Conclusion : 2010 minimum extent is on track to come in just below 2006. With the cold temperatures the Arctic is experiencing, the likelihood of a big melt is diminishing.
NSIDC shows extent lower than 2007. The modified NSIDC map below shows where they believe the gaps are (in red.)
On the surface, this appears to disagree with my measurement above that Arctic Basin ice area is greater than 2007. The difference is that I am only considering regions that have perennial ice – i.e my PIPS calculations assume that there is no summer ice in Hudson Bay or the Barents Sea, and that they are a “don’t care” for estimating minimum extent. You can see in my PIPS ROI (Region of Interest) maps for June 14 below, that there is excellent agreement between PIPS and NSIDC. 2007 had less ice in the Chukchi and Laptev Sea, and more in the Beaufort Sea.
PIPS June 14, 2010
PIPS June 14, 2007
The modified NSIDC image below shows ice loss since June 3.
The modified NSIDC image below shows ice loss since early April.
And finally, the modified NSIDC image below shows regions which have below normal ice extent.
My analysis indicates the highest late summer extent since 2006. In discussing, please remember that the regions in red above normally have no ice in September. They don’t figure in to the summer minimum.