NOTE: there are some animated GIF’s in this post that may take time to fully load. Patience please.
By Steve Goddard
Monday’s NSIDC Arctic ice extent graph took a turn downwards, and is now showing 2010 a little more than 500,000 km2 higher than 2007. The animation below shows the change from May 1 to May 2.
By contrast, NORSEX shows something very different for May 2. They have no downwards turn, their ice extent measurement is right at the 1979-2006 average, and they show 2010 extent more than 1,000,000 km2 above 2007.
In order to look at this closer up, I superimposed the NORSEX 2010 data (red) on the NSIDC 2010 data (blue) at the same scale, and normalised to 2010, and saw some interesting things. The first problem is that they started to diverge right around the first of April, and as of May 2 they disagree by nearly 500,000 km2.
The next image shows that the X-Y scaling is identical (but normalised) in the two graphs. The grid is from NORSEX. Other colors (besides red) have been removed through chroma keying.
The second discrepancy is that the two sources show a large difference in growth since 2007. The image below normalizes the 2007 data – with identical horizontal and vertical scales. Using this view, NORSEX shows twice as much ice growth as NSIDC since 2007.
The animation below begins normalised to 2010 and finishes normalized to 2007. This technique does not show that either source is in error or has changed their data, rather the animation is done by me to enhance visualization.What it does show is the significant differences between the two records.
I believe that both groups use SSMI so it is difficult to understand what the problem is. Last year we saw something similar. NORSEX has a history of making adjustments in mid-season, so my sense is that NSIDC is probably more accurate. Any ideas from readers?
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Excellent post, thanks as always. I’m sure it’s a total coincidence that the two records diverge right around the time the ice looked to exceed the ’79-00 average.
If you have two watches, they will show two different times. And neither one is likely right.
I think only Al Baby really cares about it.
..and the polar bears, of course.
I think we all need to accept the normal variation in these early months and be patient. The key will be to see what happens at two seminal points; late June and late August. By September we will know whether we continue to rebuild multi-year ice or are running low again.
I’d be interested in seeing how things change over time with regard to the wind direction as correlated to the area of ice loss. It would provide a better predictor than just comparing today with yesterday’s coverage, and would give the less patient of us something more useful to think about.
d.
While they both use SSMI it’s not clear that they use the same satellite. Last year when the SSMI on one satellite (F15?) went down it affected CT, NSIDC and NORSEX and they all took steps to fix it, CT switched to JAXA, NSIDC switched to SSMI on another satellite (F18?) and completed cross calibration, I don’t recall that NORSEX actually said what they did. Since that time the ArcticROOS data seems to do some strange things and be the ‘odd man out’ when compared with the other sources, so I don’t regard it as reliable any more. JAXA, Uni-Bremen, and NSIDC (smoothed) are consistent despite using different platforms.
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
http://iup.physik.uni-bremen.de:8084/amsr/ice_ext_n.png
http://nsidc.org/data/seaice_index/images/daily_images/N_stddev_timeseries.png
Les Johnson says: “If you have two watches, they will show two different times. And neither one is likely right.”
Yes, and these curves are nowhere near as accurate as a watch. They’re more like wrist sundials–lots of guesswork. Maybe a lot of wishful thinking, too, in the case of ice extent. .
Steven: Thanks for this post. I have noticed the relatively sudden drop in the AMSR-E Ice Extent graph that Anthony displays, similar to the NSIDC plot. I cannot imagine why there should be such a discrepancy, however, it amounts to only 3.8%. Is this so significant? I would be more concerned to see such a difference after the Summer melt. I have also noticed that the AMSR-E plot exhibits a sizable blib on 1st June each year from 2002 to 2009, inclusive. What happens in the Arctic ever 1st June? Independance Day?
“Last year we saw something similar. NORSEX has a history of making adjustments in mid-season, so my sense is that NSIDC is probably more accurate.”
Based on last year’s observations and abjustments by NORSEX which of the two was the most accurate?
I’ve noticed that NSDIC sometimes shows open ocean (concentration = 40%, but sometimes >=80%). I’m guessing that Cryosphere Today creates composite images that contain data from several days earlier (as must NSIDC), but I’ve seen the discrepancies linger consistently for protracted periods of time. Discrepancies can be both on the Pacific and Atlantic sides. The biggest difference I’ve seen recently has been in the Sea of Okhotsk (sp?) where NSIDC has consistently shown major open ocean along the coast and in the north that hasn’t existed in the Cryosphere pictures. The difference was even more glaring a week ago, but still exists. It looks as if there is some inconsistency in algorithms that I don’t believe can be explained by differences in binning/smoothing/interpolation.
NSIDC does some average-ing to generate their graphics. If ASME data is any indication, the NSIDC curve will trend towards being less steep in the days to come.
All I really know is my wife says the ice in the freezer is on the decline and I think it is well within the norm…course she drinks Margaritas and I bourbon…neat.
perspective is all lets try balancing today against an average from 1980 to 2010?? neat
I think unless the two systems have been calibrated against each other, one should expect them to be different. The important thing will be the trends when comparing data from each source. It would be like time keeping – any two watches will keep different time. But if I am measuring hours, the differences wouldn’t matter.
I expect the trends are more important than the exact daily number, and we need to look at NSIDC versus NSIDC and NORSEX versus NORSEX. DMI for example is quite different since they use 30% ice, but I still like it for trending but it is DMI compared to DMI.
Wayne
Interesting find Steve. With sea ice really falling way down in the Barents Sea, I tend to think that NSIDC has it correct, but the reason for the large difference is curious.
Nice update today on the NSIDC site for current conditions in the arctic. They focus on the warmth the artic has been seeing this year:
http://nsidc.org/arcticseaicenews/index.html
But they also focus on something I think is even more interesting, and that’s the pick up in ice flow out through the Fram Strait since the winter. A nice map is included showing the ice motion:
http://nsidc.org/images/arcticseaicenews/20100504_Figure6.png
I think the most accurate data today would show that we are now slightly below 2009’s level for sea ice extent for the same date, and getting very close to 2008’s level. I think with the heat, the rate of melt, the increased flushing through the Fram Strait, and the generally thin ice conditions because the winter was mild in the arctic, we are headed for a summer low near 2007’s record low (or slightly above, as I predict).
Most importantly, the sea ice volume anomaly continue downward:
http://psc.apl.washington.edu/ArcticSeaiceVolume/images/BPIOMASIceVolumeAnomalyCurrent.png
Lower volume, higher temps, increase Fram Strait flushing…probably the only thing to prevent a near record summer minimum would be Katla in Iceland behaving very very badly…and there is a reasonable chance of that as well.
Are you sure that NORSEX uses SSMI? The NORSEX curve looks very similar to the ASMR-E curve in the time frame in question.
Got caught in formatting. Should have said that NSIDC shows open ocean where Cryosphere Today shows ice. And not just near the concentration boundary of 15 percent.
Ice coverage algorithms can produce different although consistant results. Consistancy is the key, compare apples to apples each year, but comparing apples to oranges is a waste of time.
For a more accurate ice coverage you should use the AMSR-E data (6.75Km resolution v’s 25Km for SSMI). Incidently this also shows the drop on the 3rd May caused by the melt of thin ice in the Sea of Okhotsk. Link is in the Live Weather Roll on this page or for a different algorithm here http://www.iup.uni-bremen.de:8084/amsr/ice_ext_n.png
I think someone forgot to turn off an Aprils fool Joke.
Aren’t there any sattelite pictures to compare? 500.000 km of ice isn’t likely not to “dissapear” on those.
The missing ice went off to hide with the missing heat. This will result in a missing sea level rise.
Obviously James Hansen has gotten his hands on things.
At Bob
“I have also noticed that the AMSR-E plot exhibits a sizable blib on 1st June each year from 2002 to 2009, inclusive. What happens in the Arctic ever 1st June? Independance Day?”
They change algorithms to account for melt pools forming on the surface of the ice. This prevents underestimating sea ice area.
Not a good analogy in this day and age; consider Radio Controlled Wristwatches; “A look at radio controlled wristwatches, including those that synchronize to WWVB.”
.
.
“The map is not the territory”
“If the map shows a different structure from the territory represented — for instance, shows the cities in a wrong order. . . . then the map is worse than useless, as it misinforms and leads astray.”
Alfred Korzybski
Perhaps someone saved a big chunk of ice for all the Daiquiris (Ice+Rum+Lemon juice) which will be prepared in the next Cancun Climate Change Summit, just to relax all climate scientists, who have worked so hard, during the year, inventing all kind of convenient theories about a wide variety of armageddonian scenarios, providing all his excellencies-politicians and most holy prophets of the New Age climate change creed with the needed input in order for them to wiseacre about the necessity of a world “governance” which will empower them so as to govern upon our destiny without any hindrance whatsoever.
I watch the ice daily and the downturn doesn’t jive with visual changes at http://nsidc.org/data/seaice_index/