Ice thickness models 2

As promised I'm returning to the sea ice thickness models I wrote about two weeks ago, highlighting some of the current differences between PIPS, PIOMAS and TOPAZ. The satellite that will give us a definitive answer on ice thickness and volume, CryoSat-2, is currently being calibrated with the help of surveys by ESA and NASA. It will take some more time before definite data will be made publicly available, so in the meantime we'll continue looking at what sea ice thickness models have told us so far.

PIPS 2.0 2004-2011

PIPS 2.0 (Polar Ice Prediction System) has an excellent archive, so I have retrieved all the ice thickness images from the period 2004-2011 and put them side by side (click for a larger image):

 

This year the overall sea ice thickness seems to have almost returned to levels last seen in 2004 and 2005. I'd be surprised if this were true, considering the fact that last winter was pretty mild, and of course the fact that we've seen some pretty dramatic shrinking of Arctic sea ice extent in the past few years. Although multi-year ice has recovered somewhat in the past 2 years, levels are still 10% below those of 2005:

Some other thing I mentioned in passing in yesterday's blog post on the situation around Franz Josef Land, is that PIPS seems to project 5 meter thick ice where there is in fact open water and/or a slush of small ice floes. The same goes to a lesser extent for those red dots off the coast of Alaska and northwest Canada that do not seem to be confirmed by satellite images.

But at least most of the really thick ice has disappeared from the channels in the Canadian Archipelago according to the PIPS ice thickness model, which seems to be consistent with the transport of sea ice we witnessed last year in places like Ballantyne Strait and the Prince Gustaf Adolf Sea.

TOPAZ 2008-2011

Unfortunately the TOPAZ model (delivered as part of the MyOcean project) doesn't go back such a long way as PIPS, as it is a much more modern model. But I still managed to retrieve images for 2008, 2009 and 2010 from their old site and made a composite image from their brand new dynamic viewer (click for a larger version):

The scales differ somewhat, but this year resembles 2009 most in that there is hardly any thick ice on the Siberian side of the Arctic. Most of it is pressed against Greenland and the Canadian Archipelago (but again, not in it, in contrast with previous years).

I have discussed some of the differences between TOPAZ and PIPS in the previous post on sea ice thickness models, so there's no need to do it here (but do not hesitate to comment if you think I've missed something, because I always do). Unfortunately there is no archive of PIOMAS images. The only thing I have is the animated GIF I included in a blog post from last year.

Circumstantial evidence

I don't have much circumstantial evidence, and would love to hear from others who managed to pick up some tidbits here and there. One thing I found on the ESA CryoSat blog was a mention of them directly measuring a thickness of 1.80 meters at 85.6° N 69.8° W, which would be approximately where the white dot is flashing:

After asking for some more info on their direct measurements of sea ice thickness, CryoSat validation manager Malcom Davidson replied:

Still working on the data. However as a rough indication first and second year ice were 1 – 2m thick at the two sites (85.6N and 69.8 W and 83.3 N and 62.9 W) and multi-year ice highly variable from 2m to 6m or more.

So there you have it. Checking out the Arcus data resource page I came across this 'Graph of Ice Thickness Initialization Field Used for Ensemble Simulations in 2010' from the Alfred Wegener Institute:

But that's from 2010, and thus old news.

What happened in the Arctic last week is old news.

UPDATE May 15th

Scientists from the Alfred Wegener Institute have done another survey of parts of the Arctic, as was reported by commenter Chris K. a few days back. This survey was also done to support ESA and NASA operations for validating CryoSat-2 sea ice thickness data. One other commenter, Enno Zinngrebe, was so kind as to translate the part that pertains to this blog post, another piece of circumstantial evidence on actual ice thickness:

A preliminary look at the measurements shows that the one-year ice in the Beaufort Sea () this year is 20-30 cm thinner than in the previous two years. The ice thickness was in 2009 on average 1.7 m, in 2010 on average 1.6 m, and 2011 on average 1.4 m.

ScienceDaily now has an article on this.

 

Comments

[Artful Dodger]

Nice work Nevin. It's not surprising that PIPS 2.0 shows little difference between years, since it's model uses 1979 climate forcings, rather than dynamically collected weather data.

It might also be worth reminding Readers about the Navy's plan to replace PIPS, as you've written about previously.

Warm thoughts, Lodger.

[Andrew Xnn]

Looks like May 10,2011 is the first year that there is no significant sea ice south of the Hudson Strait for this date. During previous years, the sea ice ran for hundreds of miles south of the strait.

It was also very slow to freeze up last fall.

[Kevin McKinney]

Yes, the paucity of ice on Canada's east coast last winter was really remarkable (and, in fact, remarked upon.) It caused some difficulties for, and displacement of, the seals who normally use that ice for a nursery.

I think it's unclear what impact that will have on the seal population so far, but the harp seal, at least, is anything but threatened at the moment, with population estimates approaching 10 million (IIRC.)

Of course, if this winter turns out not to be an isolated anomaly (which would not surprise me in the least) that could change.

There was once a whole bunch of passenger pigeons, too.

[Enno Zinngrebe]

From the AWI german article that was put in the other thread, by Chris K, being German I try to translate:

++++++++++++

one of the centerpieces of the mission were large-area ice thickness measurements in the inner arctic, done in close cooperation between AWI and Univ. Alberta scientists. They used a four meter long electromagnetic ice thickness sensor, called EM-Bird. For the investigations, the Polar 5 [a plane] dragged the sensor suspended from an 80m cable in about 15m height above the ice. A preliminary look at the measurements shows that the one-year ice in the Beaufort Sea () this year is 20-30 cm thinner than in the previous two years. The ice thickness was in 2009 on average 1.7 m, in 2010 on average 1.6 m, and 2011 on average 1.4 m. "I would expect that this thin first year ice will not survive the summer melt", estimates Dr Stefan Hendricks. In several weeks, his colleagues from the Marine Ice Group at the AWI will present their model estimates for the summer ice minimum of 2011, which will make use of the now-measured new data.

+++++++++

I hope this is useful for those who don´t speak German. All errors of translation mine of course etc.

[Neven]

Enno, thanks a lot for the translation. I have used part of it for an update of this blog post. Thanks also goes to Chris K. for linking to the AWI article.

Great stuff!

[michael sweet]

Enno,
Thanks for the translation. If the ice is that much thinner than in the past in the Beaufort sea what does that portend for the rest of the Arctic? Hopefully when the June Search projections come out we will get to learn about more of this data. Presumably other scientists have data from other parts of the Arctic. It was warm through most of the Arctic last winter, if the ice is that thin everywhere it will be an interesting melt season.