Pursuant to the previous post on The great 2007 ice crunch – it wasn’t just melt this is a follow up post for questions raised there.
By Steve Goddard
One of the favourite web sites quoted by the AGW faithful, is the University of Washington PIOMASS site – which claims to model ice volume. They show ice volume dropping off precipitously after 2006 and continuing downwards.
On their web site they include the verification graph below, which curiously stops verifying in 2007.
Obviously they have run their models post 2007, so why did they stop updating their verification? The image below gives a clue. I mapped NSIDC November extent (blue) on top of their verification graph, and something stands out like a sore thumb. Ice extent jumped back up after 2007, but their volume model didn’t.
‘
This in itself doesn’t prove anything, because it is possible that extent increased while volume decreased. (Not likely, but possible.)
Below is another one of their verification links, which actually verifies nothing because it only compares their model vs. extent. Since they undoubtedly fed the extent data as an input in to their model, it is rather bizarre that they would present this as verification evidence. It is like predicting the score of a game after it has already been played.
http://psc.apl.washington.edu/zhang/IDAO/summer2007_2008_arctic_seaice.gif
So let’s try doing some real verification. Below are the Navy thickness images from November 2007 and 2009.
Note that across the majority of the Arctic, 2009 thickness is greater than 2007. Notable exceptions being the 2007 pressure bump in the Beaufort Sea, and along the north coast of Greenland.
Conclusion: The PIOMASS models are probably not an accurate representation of current ice conditions. Most of the Arctic has increased in thickness since 2007.
No matter how you slice it, the Arctic is warming. At issue is not whether it is warming but what is causing it. As I pointed out, the abrupt beginning of warming rules out the greenhouse effect. It is completely impossible for atmospheric carbon dioxide to start a sudden, massive warming trend unless its partial pressure simultaneously takes a jump, and this did not happen. Which leaves ocean currents as the only way to start a sudden warming over a geographically large area. The break in warming between 1940 and 1960 fits in too because the simplest way to produce it is to temporarily change the configuration of currents. Which brings up the possibility that if it happened once it can happen again. Hence, a refreeze of the Arctic at some future time is not out of the question.
u.k.(us) said
“Looks like most of the thicker ice, is in no danger of being driven out through the Fram Strait?
My main concern being the polar bears, of course.”
I always thought the polars bears needed “rotten ice” to hunt seals as they came up through the breathing holes; good solid thick ice means they starve!
The only way the PIOMAS charts/model can be right is if ice thickness has declined by about 50% since 1980.
Furthermore, the seasonality of the average ice thickness has to have switched from the early 1980s when the average thickness at the sea ice maximum (in March) actually INCREASED during the season until it the sea ice minimum (in September) whereas today the average ice thickness declines from the maximum to the minimum.
Does the Cryosphere Today area chart look like the thickness has declined by 50% and the average thickness of that ice switched seasonality trends since the 1980s – Nope.
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.area.arctic.png
Ice thickness is actually measured regularly in some places up there. See for example:
http://pafc.arh.noaa.gov/ice.php?img=ice
Notice the keys on the map. First year ice can vary quite a bit in volume. Which leads me to point out that volume cannot be straightforwardly deduced from ice age. The models have that variable wrong. They need to model ice thickness based on concentration (potential for compaction), and wind direction and speed (potential for pressure ridging), to model potential thickness/volume. Of course you need to be able to predict the wind. Good luck with that one.
Ice melt potential is a weather pattern prediction. If the winds are strong enough and the ice has sails on it, the Arctic can flush thick compacted ice just as easily as it can ice cream. Good luck with that one.
Cryosphere Today provide comparisons of sea ice “concentration”.
Comparing May 23 2007 with 2010, concentration is clearly higher this year than in 2007 (solid purple compared to purple with white blotches). Thanks to Juraj V for this link from an earlier thread.
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=05&fd=23&fy=2007&sm=05&sd=23&sy=2010
Now how bizzare would it be if “sea ice concentration” correlated with thickness or even volume.
So I guess Cryosphere Today has to be added to the list of pro-skeptic sites and dismissed as inaccurate. The data are directly opposite to the PIOMASS model calculations. Whence this discordant cacophany? Someone is singing from the wrong hymn-sheet.
Mike G
Wait a little longer and you’ll be able to up that debt exponent again.
The Arctic is warming? Check out the unofficial (as in records from stations BEFORE they get sent out for Officialness) records for Alaska:
http://pafc.arh.noaa.gov/climate/
Where are the warming trends? Under the table? No, not there.
George Turner says:
May 28, 2010 at 4:06 pm
“Bleh. I downloaded Gimp and it can provide an exact pixel count on the images.”
George, do you have a pixel count on the colors? It is not obvious to me whether there is more or less ice in 2009. Or someone else?
Assuming a current minimum of 5.5M km2 and a steady rate of decline of 15,000km2 / 30 years …
It will be 11,000 years before the Arctic is ice free in September.
Maybe it will be called the Older Anti-Dryas.
Well, presumably PIOMAS stopped calibration because ICESAT died. Also presumably they are trying to suggest how conservative they (think) themselves to be by showing that the PIOMAS model actually shows a *higher* volume than the the last ICESAT thickness reading suggested. So the implied messaging for post-2007 is “Hey, if anything we’re probably too high”.
I don’t hate on PIOMAS. But it is what it is, and no one should expect it to be more than it is with that limited amount of calibration. Another 10 years of data from the Euro satellite and they might have something worth putting at least a little bit of reliance upon.
I’m a huge fan of iteration. I really am. But there are no shortcuts. You can’t get from there to there without going thru all the points in between.
Pamela Gray wrote, with respect to unofficial Alaska weather station data,
“Where are the warming trends? Under the table? No, not there.”
Have you found a way to download their raw data? I’d be interested to see those. From just eyeballing the trends in their unofficial graphs, it mostly looks like things are warming.
http://pafc.arh.noaa.gov/papers/updated_climo_graphs_09.pdf
rbateman writes,
“Assuming a current minimum of 5.5M km2 and a steady rate of decline of 15,000km2 / 30 years …
It will be 11,000 years before the Arctic is ice free in September.”
There’s another prediction, to put alongside Watts’ prediction that ice extent will recover this year, or Goddard’s linear projection that September ice will be gone in 2065, or Serreze thinking by 2030. So we’ll watch nature and see whose hypotheses she supports.
“Note that across the majority of the Arctic, 2009 thickness is greater than 2007.”
Just eyeballing it, Steve, I can’t say I find that obvious. And even if it’s true, there’s the question of how thick and what the total volume is. There are some red patches (5m) for 2007 that look to be around the size of mainland Britain, c. 200 km^2, and maybe more green and yellow and it’s hard to compare extents precisely. Maybe you have some figures to back up what you say? I’m not doubting you, just saying that eyeballing doesn’t do it for me.
Better keep on eye on nsidc arctic ice extent – it’s dropping like a stone. Exactly what we’d expect if the majority of recent freezes are only 1-2 years old and very thin.
The University of Washington sea ice volume chart is based on this model;
http://psc.apl.washington.edu/zhang/Global_seaice/model.html
and the page states that:
“Satellite sea ice concentration data are assimilated in GIOMAS using the Lindsay and Zhang (2005) assimilation procedure. The procedure is based on “nudging” the model estimate of ice concentration toward the observed concentration in a manner that emphasizes the ice extent and minimizes the effect of observational errors in the interior of the ice pack.”
According to this paper:
http://journals.ametsoc.org/doi/full/10.1175/JTECH1871.1
“Because of the errors in the summer Gice dataset ice concentration in the interior of the pack (as well as errors in summer ice concentration based on passive microwave observations), assimilation of ice concentration is accomplished in a method that emphasizes the extent over the concentration. The observations are weighted heavily only when there is a large discrepancy between the model and the observed concentration. Each day the model estimate Cmod is nudged to a revised estimate Ĉmod with the relationship.”
So the University of Washington used an erroneous data set, weighted heavily when observations didn’t fit the model and then “nudged” its output to the results that they wanted. Seems more akin to sausage manufacturing than science…
OMG!!! I just saw the lastest anomlay for the Antartic on Cryospheres website.
Its very clear from this that Antartic ice is in a reverse death spiral and the next ice age is just round the corner but thankfully will only affect the Southern Hemisphere.
/sarc off
Steve,
As I mentioned in the earlier article, the Navy’s PIPS analysis shows a 20% volume loss from 2007 to 2008. Both PIOMAS and PIPS show the arctic ice has lost more than a third of its volume at maximum melt during the last decade. Why do you think they are both wrong?
Tom P
As I mentioned many times, why are you setting up a straw man? I never mentioned anything about the last decade. What I am questioning is the last three years.
Nice try though.
Bill Illis says:
May 28, 2010 at 6:08 pm
The only way the PIOMAS charts/model can be right is if ice thickness has declined by about 50% since 1980.
That’s about right, from a winter thickness of 3.6m in 1980 to 1.9m in 2009.
R. Gates says:
May 28, 2010 at 4:21 pm
The Navy also uses a model, and does not have direct data over the whole Arctic, and no one will until Cryosat2, and in fact the current Navy model PIPS 2.0 will soon be replaced by a “better” model, PIPS 3.0 which will be updated to use the same Los Alamos CICE modeling algorithms as PIOMAS.
It’s nice to know that Steve Goddard has such trust in Government scientists:
So let’s try doing some real verification. Below are the Navy thickness images from November 2007 and 2009.
Ah yes.
Real verification.
Navy thickness “images”.
Too bad PIPS 2.0 doesn’t actually verify their sea ice thickness “images”, as this Award makes clear:
http://www.nrl.navy.mil/content_images/75awards.pdf
Yup, input some nice “satellite-derived ice data (ice concentration)” data to the ice/ocean model, and presto, out comes ice thickness.
In a nice “image” format.
How’s that PIPS 3.0 coming along ?
Ah, glad you asked, it’s all done, here’s the User’s Manual:
http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA489794&Location=U2&doc=GetTRDoc.pdf
And who started the development of PIPS 3.0 ?
http://www.oc.nps.edu/~pips3/
Why, it was Dr. Wieslaw Maslowski , of the Oceanography Department, Naval Postgraduate School.
Yes, that Dr. Maslowski:
http://news.bbc.co.uk/2/hi/7139797.stm
12 December 2007
Finally, someone from the Navy is speaking out on the Arctic sea ice death spiral.
None other than Mr. Pips 3.0
Also very involved with CICE.
http://oceans11.lanl.gov/trac/CICE/wiki/EarlyRefsPubs
Small world.
Thanks Anu, a nice bit of detective work…
Anu says:
May 28, 2010 at 9:09 pm
excellent detective work anu, but i’m afraid all that demonstrates is that *surprise!*
the naval post graduate school has also been politicized!
who would have figured?
Steve,
Steve,
Am i missing something here?
If you draw a line from November 2007 to November 2009 on the PIOMASS graph you get a slight recovery in ice volume which is perfectly consistent with the Navy images you gave?! This is because November 2007 was a record low ice volume month while November 2009 contained a small recovery peak in ice volume before it moved down again.
Whats all the hoopla about?
Pamela Gray,
It is clear that the Arctic has warmed over the period of satellite measurements, and by three to five times as much as the planet as a whole.
See here:
http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt
And soot was mentioned in a previous thread as a possible explanation for the amplified warming in the arctic. Yes, it is indeed one of the explanations:
http://www.giss.nasa.gov/research/news/20090408/
Anu,
You are aware that Professor Wieslaw Maslowski forecast an “ice free Arctic” by 2013?
http://news.bbc.co.uk/2/hi/7139797.stm
Now that you have shown your admiration for him, I will have to hold you to his prediction.
Steve,
I’m glad It looks like we’re agreed that both the Navy PIPS and the UofW PIOMAS teams have published figures showing that a large proportion of the September ice volume has been lost during the last decade.
You have never before suggested such a loss had occurred before, in fact quite the opposite, so it’s hardly a strawman.