Arctic Ice Extent Discrepancy: NSIDC versus Cryosphere Today

Foreword: I had originally planned to post a story on this, but Steven Goddard of the UK Register sends word that he has already done a comparison. It mirrors much of what I would have written. There is a clear discrepancy between the two data sources. What is unclear is the cause. Is it differing measurement and tabulation methods? Or, is it some post measurement adjustment being applied. With a 30 percent difference, it would seem that the public would have difficulty determining which dataset is the truly representative one.

UPDATE: The questions have been answered, see correction below – Anthony


Arctic ice refuses to melt as ordered

Published Friday 15th August 2008 10:02 GMT – source story is here

Just a few weeks ago, predictions of Arctic ice collapse were buzzing all over the internet. Some scientists were predicting that the “North Pole may be ice-free for first time this summer”. Others predicted that the entire “polar ice cap would disappear this summer”.

The Arctic melt season is nearly done for this year. The sun is now very low above the horizon and will set for the winter at the North Pole in five weeks. And none of these dire predictions have come to pass. Yet there is, however, something odd going on with the ice data.

The National Snow and Ice Data Center (NSIDC) in Boulder, Colorado released an alarming graph on August 11, showing that Arctic ice was rapidly disappearing, back towards last year’s record minimum. Their data shows Arctic sea ice extent only 10 per cent greater than this date in 2007, and the second lowest on record. Here’s a smaller version of the graph:

Arctic ice not disappearingThe National Snow and Ice Data Center (NSIDC)’s troublesome ice graph

The problem is that this graph does not appear to be correct. Other data sources show Arctic ice having made a nice recovery this summer. NASA Marshall Space Flight Center data shows 2008 ice nearly identical to 2002, 2005 and 2006. Maps of Arctic ice extent are readily available from several sources, including the University of Illinois, which keeps a daily archive for the last 30 years. A comparison of these maps (derived from NSIDC data) below shows that Arctic ice extent was 30 per cent greater on August 11, 2008 than it was on the August 12, 2007. (2008 is a leap year, so the dates are offset by one.)

Ice at the ArcticIce at the Arctic: 2007 and 2008 snapshots

The video below highlights the differences between those two dates. As you can see, ice has grown in nearly every direction since last summer – with a large increase in the area north of Siberia. Also note that the area around the Northwest Passage (west of Greenland) has seen a significant increase in ice. Some of the islands in the Canadian Archipelago are surrounded by more ice than they were during the summer of 1980.

The 30 per cent increase was calculated by counting pixels which contain colors representing ice. This is a conservative calculation, because of the map projection used. As the ice expands away from the pole, each new pixel represents a larger area – so the net effect is that the calculated 30 per cent increase is actually on the low side.

So how did NSIDC calculate a 10 per cent increase over 2007? Their graph appears to disagree with the maps by a factor of three (10 per cent vs. 30 per cent) – hardly a trivial discrepancy.

What melts the Arctic?

The Arctic did not experience the meltdowns forecast by NSIDC and the Norwegian Polar Year Secretariat. It didn’t even come close. Additionally, some current graphs and press releases from NSIDC seem less than conservative. There appears to be a consistent pattern of overstatement related to Arctic ice loss.

We know that Arctic summer ice extent is largely determined by variable oceanic and atmospheric currents such as the Arctic Oscillation. NASA claimed last summer that “not all the large changes seen in Arctic climate in recent years are a result of long-term trends associated with global warming”. The media tendency to knee-jerkingly blame everything on “global warming” makes for an easy story – but it is not based on solid science. ®

Bootnote

And what of the Antarctic? Down south, ice extent is well ahead of the recent average. Why isn’t NSIDC making similarly high-profile press releases about the increase in Antarctic ice over the last 30 years?

The author, Steven Goddard, is not affiliated directly or indirectly with any energy industry, nor does he have any current affiliation with any university.


NOTE OF CORRECTION FROM STEVEN GODDARD:

The senior editor at the Register has added a footnote to the article with
excerpts from Dr. Meier’s letter, and a short explanation of why my analysis
was incorrect.

To expound further – after a lot of examination of UIUC maps, I discovered
that while their 2008 maps appear golden, their 2007 maps do not agree well
with either NSIDC maps or NASA satellite imagery.  NSIDC does not archive
their maps, but I found one map from August 19, 2007.  I overlaid the NSIDC
map on top of the UIUC map from the same date.  As you can see below, the
NSIDC ice map (white) shows considerably greater extent than the UIUC maps
(colors.)  The UIUC ice sits back much further from the Canadian coast than
does the NSIDC ice.  The land lines up perfectly between the maps, so it
appears possible that the UIUC ice is mapped using a different projection
than their land projection.


Click for larger image

Because the 2007 UIUC maps show less area, the increase in 2008 appears
greater.  This is the crux of the problem. I am convinced that the NSIDC
data is correct and that my analysis is flawed.  The technique is
theoretically correct, but the output is never better than the raw data.
Prior to writing the article, I had done quite a bit of comparison of UIUC
vs. NSIDC vs. NASA for this year.  The hole in my methodology was not
performing the same analysis for last year.  (The fact that NSIDC doesn’t
archive their maps of course contributed to the difficulty of that
exercise.)

My apologies to Dr. Meiers and Dr. Serreze, and NSIDC.  Their analysis,
graphs and conclusions were all absolutely correct.  Arctic ice is indeed
melting nearly as fast as last year, and this is indeed troubling.

- Steven Goddard

About these ads
This entry was posted in Science, Sea ice. Bookmark the permalink.

266 Responses to Arctic Ice Extent Discrepancy: NSIDC versus Cryosphere Today

  1. mbabbitt says:

    The question Goddard asks in his Bootnote, “Why isn’t NSIDC making similarly high-profile press releases about the increase in Antarctic ice over the last 30 years?” is a serious, information seeking question if you don’t believe their is an agenda driving the NSIDC data. If you do, then you would probably take the question rhetorically and answer, “Well, duh…”

  2. Fred Nieuwenhuis says:

    There is a good running Sea Ice discussion at CA: http://www.climateaudit.org/?p=3336
    I also like this link: http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

  3. Oldjim says:

    Andrew,
    Can I suggest you modify your story. The link is to The Register – an internet type tabloid – not the Telegraph

    REPLY: Thanks, I had been reading a story on the Telegraph this morning, and had “Telegraph on the brain”. Fixed.

    But to show we all make similar mistakes from time to time, I’ll point out that my name is Anthony, not Andrew. ;-)

  4. Werner Weber says:

    There is another website /www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm from Japan which gives ice coverages. This curve looks similar to NSIDC curve. However, for the last few days, the difference to the 2007 curve appears to be a bit larger.
    Keep in mind that NSIDC gives an area with at least 15 % ‘sea ice’, while the http://arctic.atmos.uiuc.edu/cryosphere/ website probably only shows areas of full ice coverage. Consequently, the UIUC ice area should be considerably smaller than the (presently) 6.07 million km2 of IJIS and a very similar number of NSIDC. Maybe, you could check this from your pixel data.

  5. Richard deSousa says:

    Why doesn’t the NSIDC have a similar graph showing sea ice in Antarctica? Oh, yeah, I forgot, there’s no panic down there about shrinking sea ice.

  6. Ric Werme says:

    There are firsthand reports at http://awberrimilla.blogspot.com/ which is a blog about a sailboat trying to nagigate the NW passage. They’re right in the midst of the diciest part and dealing with new ice forming close to old ice dissapating.

    A lot of the posts are really exchanges between crew and support folk. It takes a lot of reading between the lines to figure out what’s happening, but it’s a lot more interesting than comparing Arctic maps or trying to figure out the weather might be changing them.

  7. youngestjim says:

    The bias at NSIDC appears to be inexplicable. Perhaps they can only get the budget they desire by creating pseudo crises?

  8. David L. Hagen says:

    See detailed comparative graphs by
    Aaron Wells’ Sea Ice Stretch Run #2 Comment 612

    The final revision is in for yesterday’s data. Here are my graphs for yesterday’s ice extent data, based upon IARC/JAXA Ice Extent data.

    Daily Arctic Sea Ice Extent Reduction Day 227

    Arctic Sea Ice Extent Day 227

    2008 Arctic Ice Extent Differences From Previous Years Day 227

    These look much closer to the Cyrosphere comparison above.

  9. Steve W. says:

    I remember that NSIDC ice graphs of a month or so ago showed the blue line almost exactly between the average line and the 2007 line. Now the blue line is touching the 2007 line in June! It looks like they “adjusted” it down. Does anyone have copies of older graphs to check this?

  10. Ric Werme says:

    Oh, I just stumbled on another sailboat taking on the NW passage from the other direction. They’re in the same area as the Berri. Lotsa Polar Bears. Must be waiting for the ice to freeze the boats in place. :-)

    http://www.69nord.com/english/expe/logbook.html

  11. Mikael H says:

    The QuikScat satellite confirm the extent from the maps above:

    http://www.ifremer.fr/cersat/ARCHIVE/images/psi/quikscat/25km/north/2007/QMO/20070812.GIF

    http://www.ifremer.fr/cersat/ARCHIVE/images/psi/quikscat/25km/north/2008/QMO/20080811.GIF

    Also note that the ice thickness seems to be better than last year; so if recovery speeds up compared to the slow recovery in oct-nov-dec last year, the ice will maybe be back to “normal” in 2009.

    Btw, here is NSIDC curve for Antarctic:
    http://nsidc.org/data/seaice_index/images/daily_images/S_timeseries.png
    … and as mentioned; keep in mind the 15% limit.

  12. Joseph Conrad says:

    Ah, El Reg’s Mr. Goddard! Who is he, one may ask, as a “Steven Goddard” doesn’t even exist on the interweb before April 2008.

    Yet the Register chose him to continously enlighten us with his views on climate change without criticism. Is he perhaps having some connection to IT professional John Atkinson, who was allowed to post a similarly incoherent article on the Register? Is it just a coincidence that someone using the moniker “John A.” pops up regularly on the blog of Stephen McIntyre, who works in IT as well and is known for making powerful oil interests heard?

  13. Alan Millar says:

    I am pretty certain that NSDIC retrospectively altered their graph for the period late July to early August. The original graph showed it tracking away from the 2007 position and then it was altered to showing it tracking towards it.

    Has anybody got a copy of the original graph to show the alteration?

  14. Gary Plyler says:

    I have had problems in the past with NSIDC. I was emailing back and forth last February with an NSIDC expert on a technical item (the age of an in-situe formed ice shelf that ‘suddenly’ broke free, like they ever slowly break off, sheesh). I contended that it was a relevant problem of interest since if there was no ice there during the MWP and it formed during the LIA, what’s the problem.

    Eventually the NSIDC expert sent me to RealClimate.org. Go figure. He and his lunchroom buddies have already imbibed in green fool-aide.

  15. terry46 says:

    I’m glad to see that you use white to show the true color of ice at the artic .If you look at the accuweather site on global warming they have a post about the ice melt of course .I’m not sure what year their picture was supposed to be from but the thing that got me was the color of their ice .It is purple in color .I’ve never seen purple ice before .And the red on their map .What is that icey hot.

  16. Oldjim says:

    Sorry Anthony – blame that one on a senior moment

  17. Mike C says:

    [snip ad-hom]

    The rest… there seems to have been a sharp upward spike in the past few days in Antarctic ice, it looks like there may be another record in the south.

  18. R John says:

    Mr. Conrad –

    Regardless of who Goddard is or is not – can you refute the substance of his report? To say that he never existed before 2007 on the internet is absurd. Did you personally look through every Google hit?

  19. tarpon says:

    Anthony, Is the raw data direct from the satellite available? It should be fairly straight forward to do the data processing, especially a pixel by pixel processing, and draw conclusions. It looks like the pictures that are displayed are just that.

    I sure hope we don’t have another science group going green.

  20. Bruce says:

    Mr Conrad,

    Implying that anyone who writes articles that debunk global warming is funded by “powerful oil interests” is kind of silly. I believe the ratio of nutbar pro-warming funding to oil interest funding is 1000 to 1.

    Al Gore himself has a 300 million budget for just this year to propagandize. (Admittedly most of that goes to fund the energy budget of his home and private jets and yacht).

    Tens of thousands of scientists and journalists rely on AGW funding just to pay their salary. They are bought and paid for.

  21. jonk says:

    There is a link on Anthony’s site to the earlier graph you all are talking about.

    http://wattsupwiththat.wordpress.com/2008/07/14/when-graphs-attack/

  22. Jeff C. says:

    Joseph Conrad –

    Interesting suggestion that Steven Goddard and John A. may be the same person. If true, the Goddard penname is likely a mischevious tweak of GISS. So what? Please address the facts of the article. What information that is presented as fact is untrue?

    Calling people big oil shills just doesn’t work anymore, if it ever did. It resonates in the AGW echo chamber, but the public isn’t buying it. The public view Exxon as a company that provides a product that makes their life easier, not as one hell-bent on destroying Mother Earth. Public opinion polls showing greater than 65% approval for more domestic drilling make this point crystal clear.

    Making your case to the masses, persuading them to elect like-minded leaders, and implementing coices involving sacrifice is hard work. Ad homs are so much easier, but not very effective.

  23. Sam the Skeptic says:

    Mr Conrad,
    What an awfully supercilious little posting. Can you answer this poor layman a couple of simple questions.
    If you don’t agree with Goddard why don’t you post an alternative hypothesis?
    Is it more fun to sneer than to debate?
    Why are the AGW alarmists incapable of reasoned discussion?
    I onkly ask these questions because I have an enquiring mind, you understand. Something that you seem to lack.

    Thanks for the posting, Mr Goddard. Your contributions to sanity are much appreciated out in the world of us non-scientists.

  24. Leon Brozyna says:

    Alan Millar (11:43:57)

    I’ve copied and saved every daily graph from NSIDC since July 31, except for Aug 1. (Also, the Aug 9 graph wasn’t updated; it actually went from Aug 8 to Aug 10}. The axes on the graph change on Aug 5, but it still shows regular ‘adjustments’ to the line’s slope every 2-3 days, sometimes up, sometimes down. The biggest change happened Aug 1, when several days’ worth of an uptick were changed downward. So far, this year’s line is generally running parallel to last year’s line.

    So far I’m up to 14 daily .bmp files totalling 34MB.

    Anthony – You’ve got my e-mail if you want me to send you these images for use here.

  25. counters says:

    Perhaps you all harassing Joseph Conrad (why the heck do posters in the climate change debate always either adopt founding fathers or 18th century authors for their pseudonyms?) could point out Mr. Goddard’s analysis to me. What’s that? He counted the number of pixels of ice on the pretty picture?

    This is ridiculous. There isn’t anything scientific about this piece at all. Let me know when you’ve got an analysis that goes a tiny bit further than counting pixels on pictures… perhaps a mathematical analysis of recent years’ trends and their deviations from the mean ones, or an analysis of the methodology used by NSIDC to calculate their data? I’m not interestied in how many pixels are a different color on two pictures.

    REPLY: COUNTERS, Pixels ARE DATA IN A DIFFERENT FORM. Pixels are formed from numeric data, into a corresponding RGB value. There are graphical histogram programs that count the number of matching pixels of specific RGB values. Using the color scale provided, and its corresponding RGB value, it is an easy and certain mathematical exercise to turn those pixels back into real data, with no loss, no rounding, no floating point error. I’m really surprised that you would snark with “He counted the number of pixels of ice on the pretty picture?” Very unprofessional.

    Take a breather to consider dialing back the snark a bit. – Anthony

  26. Evan Jones says:

    Ah, El Reg’s Mr. Goddard! Who is he, one may ask, as a “Steven Goddard” doesn’t even exist on the interweb before April 2008.

    Geta loada that one, Steve! (I hope you got as big a yuk out of it as I did.)

  27. Alan Millar says:

    Leon

    Thanks for that. I had been following the graphs daily and was amazed to see such an ‘adjustment’! Really seemed to be an about turn.

    Alan

  28. Leon Brozyna says:

    Anthony

    Well, I learn something new everyday. I turned those 14 .bmp files with 34MB total size into 14 .gif files of a bit over 320KB total size. Then my email automatically combined them into a self-extracting zip file. Hope you can make something out of them. Watching the battles of upticks/downticks is interesting.

  29. Steven Goddard says:

    Anthony,

    Thanks for putting this up here.

    The CT maps correlate pretty closely with the NSIDC extent maps. Both show the area where concentration is above about 15 or 20%. The reason I used CT maps is because they archive them every day, whereas the NSIDC maps are only archived once a month, and without a specific day of the month attached.

    Numerical integration by counting pixels is a standard technique in image processing. Any error that is introduced is much less than the 45% difference seen in the latest imagery from today. No one can seriously argue that minor differences in pixel areas will account for this –

    (BTW – I resent the insinuations that I don’t exist. Last time I checked, I was sitting in this chair typing.)

    REPLY: I’ll add that I looked up the geolocation if the IP address from Mr. Goddards posting (WordPress provides them with each comment automatically) and it shows his location to be near Edinburgh, England. JohnA lives in the southern hemisphere now, though he lived in England last year.

    So no more baseless accusations about Mr. Goddard’s identity will be accepted.

  30. McGrats says:

    counters (15:19:01) wrote : “This is ridiculous. There isn’t anything scientific about this piece at all. Let me know when you’ve got an analysis that goes a tiny bit further than counting pixels on pictures… perhaps a mathematical analysis of recent years’ trends and their deviations from the mean ones, or an analysis of the methodology used by NSIDC to calculate their data? I’m not interestied in how many pixels are a different color on two pictures.”

    counters, your knowledge of pixels seems to match you knowledge of climate change. You have a lot of bluster, but few facts. I use pixels day in and day out and believe me, you can obtain an accurate representation of a given area through a “head” count.

    Jack Koenig, Editor
    The Mysterious Climate Project
    http://www.climateclinic.com

  31. Jeff C. says:

    Wow, Goddard must have really hit a nerve with all the gnashing of teeth this one has provoked. Of course the pixel count is scientific, it is one of many plausible methods to determine ice area. As the article also states, the estimate is conservative due to the projection. Pixels at lower lattitudes represent more area than pixels at the pole. With a little spherical geometry applied this anaysis could be refined from the bounding case.

  32. Patrick Henry says:

    One of my professional activities is doing DNA sequencing using electron microscope images. Pixels of various brightnesses are counted and analyzed, to determine the location and type of chemicals present.

    When dealing with micro or macro imagery, pixel colors and brightnesses are normally the only tool you have to work with. How else would you take satellite imagery and turn it into sea ice information?

    Some of the vocal critics might want to slow down their mouths, and do a bit of thinking (for a change.)

  33. Steve Hempell says:

    A little off topic but all you pixel experts out there…..

    I want to integrate an area under a complex curve I have graphed. It always seemed to me that you could do this by counting pixels. I have searched but cannot find one. Is there a program that you can input data for a graph, count pixels under a complex curve and thus find the area under a the complex curve (or a portion of a complex curve)?

    REPLY: Yes, there is a program called ImageJ that will do that, so long as you fill in under the curve with pixels of a specific color that is different from the background. It was developed to do automated cell counts of microscope imagery by the National Institutes of Health. – Anthony

    See features here:

    http://rsbweb.nih.gov/ij/features.html

  34. Mike C says:

    Jeff C,

    I noticed that too. You would think there is a full moon the way they all came out on this one.

  35. Jeff says:

    The whole article is completely worthless, as one would expect from an article written by a fictitious person. It doesn’t make a bit of difference whether this year’s sea ice extent is slightly more or slightly less than last year’s. The University of Illinois webpage cited in the article look has graphs of the sea ice area and sea ice area anomalies and see that the long-term trend is towards smaller sea ice extent year-round. And that’s what’s really important.

    I’m not sure how “Steven Goddard” came up with his 30 percent difference claim. I counted the pixels, using NSIDC’s “Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I Passive Microwave Data” data set and with a 15 percent criteria, and came up with about a 14 percent difference in the number of pixels with sea ice between the 2 dates. NSIDC is reporting sea ice EXTENT, not sea ice AREA, so in fact it’s quite possible that their 10 percent figure is correct.

    REPLY: “The whole article is completely worthless, as one would expect from an article written by a fictitious person.” “Jeff” I take exception to this. Prove Mr. Goddard is “fictitious” or kindly stop posting your opinion about it.

    “Jeff”….hmmm, who is that anyway? Since I don’t ahve a last name you must be “fictitious” right? ;-)

  36. D. Quist says:

    Counters

    I am a skeptic but, I do enjoy someone spicing up the discussion here. Sorry to see you knocked silly on the pixel comment. It is Friday and you might be a bit tired.

    My question is I don’t see a comparison with the NIC IMS product. http://www.natice.noaa.gov/ims/
    It shows the ice extend being much greater than the cryosphere pictures. Especially in the area around Eastern Siberia across to Canada. It shows much more ice. I have been looking at those pictures and comparing them. For example, compare the ice around Wrangel Island north of Eastern Siberia. In the latest NIC picture it is completely surrounded by ice. In Cryosphere it is ice free.

    I assume they use different thresholds.

  37. Flashman says:

    Apparently, the doubleplusungood deniers now tether on the brink of being declared unpersons.

    I would, with all due respect, like to ask Mr Joseph Conrad if he has any factual comment on the relevant post, unless the inquiry itself is apt to render my humble self this perilous fate.

    Alas! No oil money here! But questions remain, unless the asking itself leads to bitter oblivion!

    In a Hansen Camp(tm) one could at least await one’s tribunal and the noose with head held high, but to be blotted out like the MWP at the whim of an acolyte? Nay! That is to much!

    So if these are the stakes, I submit and humbly recant my heresy!

    Otherwise, if it’s not objectionable, I’d really appreciate some enlightenment on the factual issues :)

  38. Jeff says:

    How does the location of an IP address guarantee that the poster actually is at that location? What happens when someone is logged in remotely? The Register happens to have an office in Edinburgh.

  39. counters says:

    Sorry for the snark earlier. I can’t excuse my tone on a couple posts this afternoon, so please accept my apology and I’ll work extra hard to dial it back a bit. I really could’ve used that cup of coffee this morning! :)

    Look, all I’m trying to say is this: This is a serious allegation about the accuracy of a major repository of data. A thorough analysis which fully addresses the methodology of the NSIDC would only be a good thing. The problem is that publishing allegations in an online newspaper is not the proper way to get things done. I know most of you are skeptics here and I’m a “warmist” (although I prefer the less-pretty term “person who is sufficiently satisified with the data forming the basis of AGW theory and the mainstream conclusions in the field”), but maybe you can understand my frustration when every little allegation or comment going against the grain of AGW is jumped on. Skepticism is good and healthy; it’s important that people ask the hard questions and further the science.

    But when an article such as this is posted, the situation changes. While many people will still keep focused on the important thing – verifying the methodology of the NSIDC – others will use this as another convenient one-liner as to why AGW is false. I can see the posts on DotEarth or WU now – “the NSIDC overestimates ice cap melt, therefore AGW is wrong”.

    The bottom line is that when skeptics pursue issues such as this by engaging it in the popular media, a train wreck ensues. Allegations will fly about political motives; ad homs and poor debate will ensue on the blogs. If someone has serious concern with the methodology used to produce a major, then the appropriate channels of communication to be used are the scientific ones. Publish work demonstrating some sort of deficiency or discrepancy in the final products to a Journal, or contact the adminstrators of the product and voice your concerns. I simply can’t see how writing a newspaper column will do anything but fire up the noise machine on each side, and if there’s any doubt as to whether or not this is the ensuing reaction, I submit my own previous, inappropriate post as evidence.

  40. Ric Werme says:

    Jeff (20:01:11) :

    How does the location of an IP address guarantee that the poster actually is at that location? What happens when someone is logged in remotely? The Register happens to have an office in Edinburgh.

    It’s possible, but what’s the gain? Hackers often use multiple levels of logins, even ethical users did that as long ago as the early 1970s on the Arpanet because one computer had a better net interface than the terminal servers.

    Some unavoidable occur at large companies – I worked at a firm that allowed ssh access to the Internet. I live in New Hampshire, but the corporate firewall was in California. Something like that could well be the case, but why do you care so
    much?

  41. Steven Goddard says:

    In order to compare CT maps vs. NSIDC maps, I made a video which superimposes the NSIDC map for August 14 on top of the CT map. As expected, it appears that they are nearly identical, though there are some odd discrepancies – like a small area NE of Svalbard where CT shows >50% concentration and NSIDC shows open water.

    http://nsidc.org/data/seaice_index/images/daily_images/N_daily_extent_hires.png

  42. Jeff C. says:

    Jeff –

    “The University of Illinois webpage cited in the article look has graphs of the sea ice area and sea ice area anomalies and see that the long-term trend is towards smaller sea ice extent year-round. And that’s what’s really important.”

    And I would add The South pole sea ice trend is toward a larger sea ice extent/area. Why isn’t that what’s really important?

    Either way, it’s beside the point. Goddard’s article states that he sees a discrepancy between what is reported and what he observes. I commend you for checking into it yourself and more of us should be doing the same. Perhaps the difference lies in the definition of extent vs. area or different data sources. However the info from the Cryrosphere today does seem at odds with the NSIDC chart.

    This is an article in an online journal that raises a question, not some “peer-reviewed” paper claining to settle the science. There are many smart people that are regular commenters here. Perhaps we can all do some auditing ourselves and figure it out.

  43. steve says:

    IP address is not *all* that meaningful in determining someone’s location. It can in some cases (probably most) let you know where they’re posting something from, but it’s possible to post through a proxy or intermediary server and the IP you will get in that case, is the IP of the in-between server and not that of the original posting location. People use this to get around those web sites that only allow streaming content in the US by tunneling their requests through a US based server.

    I don’t know who he is but it doesn’t really matter does it? I agree with the sentiments suggesting that the substance of the article is what matters.

    I guess my only question would be… what does any of it mean? What do we mean when we talk about artic sea ice? Overall area? Thickness? area + thickness? Do satellite images in any way infer thickness of the ice? or is it just surface area? For instance, could the ice be wide and very thin and is that the same as a smaller thick area of ice? Certainly counting pixels will give you a count of some kind of real data, but I’m not sure what it actually tells you. Maybe someone can answer that.

    All we really need is a valid measuring tool, and a consistent measurement (ie using the same reasonably valid methods) and a comparison of those same method’s measurements over time. Maybe that’s possible with some kind of satellite pixel approach.

  44. Phil. says:

    Jeff C. (16:45:03) :
    Wow, Goddard must have really hit a nerve with all the gnashing of teeth this one has provoked. Of course the pixel count is scientific, it is one of many plausible methods to determine ice area. As the article also states, the estimate is conservative due to the projection. Pixels at lower lattitudes represent more area than pixels at the pole. With a little spherical geometry applied this anaysis could be refined from the bounding case.

    Goddard has made a mistake and his results disagree with all the published data. His mistake is probably because he is counting from the transformed data in the image rather than use the original data which can be downloaded? He and Jeff here appear to have a misapprehension concerning the projection used which is based on the secant at 70ºN not the tangent at the pole, which might be the source of the error.

  45. dreamin says:

    “I can see the posts on DotEarth or WU now – ‘the NSIDC overestimates ice cap melt, therefore AGW is wrong’.”

    I would analogize the evidence for CAGW to the evidence for alien visitation. There is lots and lots of evidence, but most of the time when you take a careful look at any one piece of evidence, it just doesn’t stand up to scrutiny. At which point, defenders of the faith are free to say “well maybe Crop Circle #1 is a hoax, but what about Crop Circle #2?”

    It seems to me there’s a pattern here which seriously undermines the credibility of the warmists as a group. Of course, if warmists were better at thinking critically and admitting error when appropriate, the movement would have probably fizzled out by now.

  46. Steven Talbot says:

    NSIDC is reporting sea ice EXTENT, not sea ice AREA, so in fact it’s quite possible that their 10 percent figure is correct.

    Jeff’s explanation seems highly plausible. Note this graph, from Cryosphere Today, which shows current area at c.4 million km2 –

    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.jpg

    By contrast, NSIDC’s graph of extent shows in excess of 6 million km2:

    http://nsidc.org/images/arcticseaicenews/20080811_Figure2.png

    So, Mr Goddard’s comparison is based on a basic error. Can we look forward to you clarifying this on your site, Mr Goddard?

  47. Steven Talbot says:

    A further point, Mr Goddard.You state that the NSIDC data “shows Arctic sea ice extent only 10 per cent greater than this date in 2007″. This is incorrect.

    The August 11th announcement read as follows:

    Arctic sea ice extent on August 10 was 6.54 million square kilometers (2.52 million square miles), a decline of 1 million square kilometers (390,000 square miles) since the beginning of the month. Extent is now within 780,000 square kilometers (300,000 square miles) of last year’s value on the same date and is 1.50 million square kilometers (580,000 square miles) below the 1979 to 2000 average.

    http://nsidc.org/arcticseaicenews/index.html

    Using those figures, I calculate the then current extent as being 13.5% greater than the equivalent 2007 date, bit 10% as you have stated.

  48. tarpon says:

    The answer is the same reason all the TV weathermaps now show yellow for 60 degree temperatures — The map now looks all ” reddish and hot” to the naive viewer. If they had a disclaimer about the scale change, who would care — But they don’t. It’s clearly designed with a single intent in mind, try and convince people it’s getting hotter. Why they would want do this, I have no idea.

    The ‘icy truth’ lies in the data processing software, I see no reason why the science group would not openly provide source code, and access to the raw pixel data. They could at least tell us what the change log looks like and explain any changes to the code. That would be good “open” science, instead of wasting people’s time trying to figure it out.

    Something which is designed to fix just these types of problems … The U.S. has a law, Data Quality Act of 2001, this law requires federal agencies to ensure the integrity of the information they use and distribute. It also allows outside parties to petition to force the correction of information they believe is wrong.

    The Data Quality Act (DQA) passed through the United States Congress in Section 515 of the Consolidated Appropriations Act, 2001 (Pub.L. 106-554). The Government Accountability Office calls it the Information Quality Act, while others call it the Data Quality Act.

    This public law also applies to all Federal grantees, and all Federally funded groups.

  49. Steven Talbot says:

    Further note: the 13.5% greater figure I have calculated, as above, is of course remarkably close to Jeff’s recalculation of the pixel count, viz.:

    I counted the pixels, using NSIDC’s “Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I Passive Microwave Data” data set and with a 15 percent criteria, and came up with about a 14 percent difference in the number of pixels with sea ice between the 2 dates.

    We seem to be progressing towards an explanation, which is simply that the discrepancy proposed by Mr Goddard doesn’t exist.

    It seems wise to check over figures before leaping to a conclusion. That seems to me to be the proper scientific and sceptical approach.

  50. old construction worker says:

    counters
    ‘If someone has serious concern with the methodology used to produce a major, then the appropriate channels of communication to be used are the scientific ones.’
    And I’m sure you were the first one to complain to IPCC about HOW new& improved “hockey stick” made it into their report.

  51. kim says:

    Interesting thought, Phil. Anyway to pinpoint the error? Steve?
    =======================================

  52. Steve Keohane says:

    I overlaid the iceextent line from 08/11 onto the earlier chart from 07/14 and found the line has been changed from even the stating point at the beginning of May. I have been following the extent all summer and have seen discussion re: the slope change last month, but more than that has been changed. I may be off a pixel or two, but believe this is pretty accurate:
    http://i33.tinypic.com/2rgyzvm.jpg

  53. rjb says:

    Why is there so much attention on the Arctic ice? I’m reading this fascinating book “1421: The Year China Discovered America” by Gavin Menzies. In the section on the voyage of Zhou Wen, he reviews the claims that the Chinese circumnavigated Greenland around 1421-2. He discusses evidence of a much warmer climate in Greenland between 1422 and 1428 before the onset of the mini ice age in the 1430s, including a change in the types of flies found during excavations dated to this period in Greenland. He mentions Captain Nares’s voyage in 1875 when all but 15 miles of the Greenland coastline were free of ice. Browsing the image archives at http://arctic.atmos.uiuc.edu/cryosphere/, I don’t see any satellite captures showing an ice free route around the coast of Greenland in recent history (including the most extreme times of the 2007 melt). So why all the attention on the Arctic ice now, when it appears there is ample evidence to suggest the ice has melted more in the recent past than it has melted in the past decade? From the layman’s perspective, this appears to be a Red Herring fallacy.

  54. Jeff C. says:

    Phil – thanks for clarifying the projection used in the data. I assumed that the projection used standard k-space, i.e. the sphere tangent to the projection plane at the pole. Now that you bring it up, I do recall the 70 deg comment from the Sea Ice thread at CA.

    Your comments may be inconvenient, but they are always thoughtful. It is appreciated.

  55. Jack Simmons says:

    Mike C (18:14:33) :

    Well, uh, in fact the full moon is out.

  56. John McDonald says:

    Can some define what Extent is?

    Area is an easy concept for me, Extent seems like it can have many defintions.

  57. Steven Talbot says:

    [I said Jeff twice above when I meant Phil - sorry!]

    John McD, extent is the area of ocean containing 15% ice, ‘area’ is the calculated area of the ice itself.

  58. dreamin says:

    Why is there so much attention on the Arctic ice?

    For the same reason there was so much attention paid to hurricanes a few years ago and global surface temps before that. The warmist spin machine is desperate to find weather events which can be trumpteted as evidence of catastrophic anthropogenic global warming.

    If artic ice rebounds, the COGW will seize on something else.

  59. Steven Goddard says:

    There is no mystery to my calculations, and I explained them clearly in the article.

    I took the August 12, 2007 and August 11, 2008 maps from Cryosphere Today – located here :
    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/ARCHIVE/

    I counted the number of pixels in the two images marked as being >15% ice concentration (NSIDC’s definition of “extent”.) 2008 was 30% greater than 2007.
    You can see the difference visually here:

    It is obviously much greater than a 10% increase.

    The NSIDC 10% figure was taken from their graph issued with the press release, which showed about 6.3 vs 5.7 last year.
    http://nsidc.org/images/arcticseaicenews/20080811_Figure2.png

    Additionally, you can see that Arctic melt has nearly stopped this year.

    The point of the article is that some of the NSIDC data seems inconsistent with other sources. Some posters here are trying to use NSIDC data to prove NSIDC data – which is obviously a flawed approach.

  60. Glenn says:

    John McDonald,

    Well the NSIDC chart of “Sea Ice Extent” shows “Area of ocean with at least 15% sea ice”.
    http://nsidc.org/data/seaice_index/images/daily_images/N_timeseries.png

    So extent is a measure of two dimension area.
    I don’t know how that would differ from the “atmos” pics of “sea ice concentrations” except they can go lower than 15%. However, I’ve not seen that.
    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/ARCHIVE/20070812.jpg

  61. Steven Goddard says:

    One more point. The UIUC maps are centered directly over the pole. I did not make any corrections to the data in the article, but had I – the discrepancy would have been increased due to latitudinal distortion as you move away from the poles.

    This is very simple stuff. Some people here are making bold declarations that I am in error, without actually addressing the methodology that I used.

  62. Steven Talbot says:

    Steven Goddard,

    You say:

    The NSIDC 10% figure was taken from their graph issued with the press release, which showed about 6.3 vs 5.7 last year.

    So you have eyeballed a graph and think that your guestimate of what it shows will do, when you could use the actual figures, as I have quoted to you? These figures were given in the very same press release from which that graph comes! To repeat:

    Arctic sea ice extent on August 10 was 6.54 million square kilometers (2.52 million square miles), a decline of 1 million square kilometers (390,000 square miles) since the beginning of the month. Extent is now within 780,000 square kilometers (300,000 square miles) of last year’s value on the same date and is 1.50 million square kilometers (580,000 square miles) below the 1979 to 2000 average.

    http://nsidc.org/arcticseaicenews/index.html

    That is 13.5% greater than last year, not 10% as you claimed. This is basic arithmetic! You are significantly in error, and should have the integrity to say so (and given this blog’s proper concern for accuracy, that is an expectation that I suggest should be widespread here).

    As for your comment here –

    The point of the article is that some of the NSIDC data seems inconsistent with other sources. Some posters here are trying to use NSIDC data to prove NSIDC data – which is obviously a flawed approach.

    I presume by ‘other sources’ that you mean the Cryosphere Today site? What, pray, is the source of their data, other than the NSIDC? It is actually you who is trying to use one representation of NSIDC data to question another! The data is the same in both cases!.

    Anyone can make a mistake, Mr Goddard, but not everyone has the capacity to admit it, it seems.

  63. ed says:

    Goddard “error” (i’ll use this euphamism for what he really did) is obvious even to the layman. He totally ignores the scale on the image that shows the various colors represent different levels of sea ice coverage. Red is 60% while purple is 100%. Goddard treats them equally but it is clear from inspection the current year has more area with less % coverage – so goddard’s primitive calculation is bound to be way off.

    Its pretty funny how you guys can question the NSDIC’s numbers based on this obviously bogus analysis.

  64. Steven Talbot says:

    rjb,

    How can I best put this? Gavin Menzies’ book has been the object of considerable contempt from professional historians. Here’s a basic Wiki article on the subject –

    http://en.wikipedia.org/wiki/1421_hypothesis

    Of course, the ‘consensus’ of historians might be wrong and Menzies might be right, so you’re welcome to make your own judgment. Lord Monckton has also referenced Menzies in support of his arguments (http://www.telegraph.co.uk/news/uknews/1533290/Climate-chaos-Don%27t-believe-it.html), so I guess you have company if you judge it to be plausible.

  65. Phil. says:

    Steven Goddard (10:39:20) :
    One more point. The UIUC maps are centered directly over the pole. I did not make any corrections to the data in the article, but had I – the discrepancy would have been increased due to latitudinal distortion as you move away from the poles.

    No as pointed out above the minimum distortion is at 70ºN and increases as you move away!

    The NSIDC calculations and the JAXA calculations both agree that the extent is today ~9.9 Mm^2 even though they use data from different satellites (SSMI vs AMSR-E). Based on their data the current extent is 8% above last year. You are the odd one out!

    Also the Arctic melt has not ‘nearly stopped’, in fact it’s continuing even faster than last year as shown by Aaron Wells graph from CA: http://img352.imageshack.us/img352/4760/seaiceextentlossratebi9.jpg

    And that despite the fact that if you’re interested in ‘melt’ you should be looking at area which is even closer

  66. Steven Goddard says:

    Ed,
    You are confusing the calculation of ice area with extent. “Extent” treats all regions greater than 15% concentration equally – just as I did.

    Phil and Steven Talbot,
    If the UIUC maps are correct, there is nothing in my extremely straightforward calculations which could be broken. You seem to be (rather obtusely) attempting to make an argument that the UIUC maps do not correctly map the satellite data on to the polar projection they use. Polar map projections have minimum distortion at the pole, where the land is perpendicular to the viewer.

    If you wish to argue that the UIUC maps are incorrect, that might be an interesting topic of discussion. However, you have yet to make any attempt to analyze the methodology I used – which (obviously) assumes the correctness of the maps.

    According to the UIUC maps, there has been almost no change in the amount of Arctic ice since August 8. Once again, if you believe those maps are incorrect – please state your case – and I suggest that you take it up with Phil Chapman over at CT.

  67. rjb says:

    Steven Talbot:
    I appreciate the links (although like Middlebury’s history department, I do not consider wikipedia a reliable source; rather a convenient research tool, littered with inaccuracies). I only just picked up the book, so it is a starting point for me. I was unaware of the fervent opinions surrounding the publication.
    Thanks,
    rjb

  68. Frank Lansner says:

    My Pixel for pixel, calculations for 2005, 2007 and 2008 16´th august:
    (ice area 2007 16 aug = 100% )

    2007: 100%
    2008: 130,3%
    2005: 132,7%

    I thus get 30,3 % more ice in 2008 than 2007, very close to Mr Goddards 30 % in the article. I can only agree.
    Notice also how close 2008 and 2005 are. We are as good as back to 2005 of ice extend.

    I then measured the colour nuances, and the area*concentration with 2007 as 100% is :

    2007: 100%
    2008: 125,0%
    2005: 123,6%

    So yes, if we incalculate concentrations, 2008 hass not 30 % larger ice extent but ”only” 25% larger ice extend than 2007.

    BUT, notice that doing so reveals that there is now more ice in 2008 than there where in 2005…

    Another thing, i see here and there in this debate the argument that “NSIDC measures from 15% ice extend”. The thing is, is changes nothing realy. If you compare 2007 with 2008 you use SAME method on both pictures, so if you on BOTH pictures set the limit to 15% this cannot realy change outcome when comparing. Unless of course that one year had extremely much 15% ice. But thats certainly not likely. Check the cryosphere photos and you will see that the big areas are never with low concentration. Such low cincentrations areas of ice mekt away very quickly.

    Summa whether you use 15% or 25% changes nothing as long as you use same method on the 2 figures you compare.

    Thanks Anthony and Goddard for pointing out this example of distorted data.
    Its realy appreaciated by many many many peoble, thankyou.

    Cryosphere opereates with concentrations in their graphs, but NSIDC does not, and therefore has no obvious excuse for their underestimation of this years ice extend. It just does not look good as so many many other things in this strange debate.

  69. Frank Lansner says:

    See for yourself what i measured above, there is now slightly more ice in 2008 than 2005, 16´th august:
    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=16&fy=2005&sm=08&sd=16&sy=2008

  70. McGrats says:

    Phil wrote: “Also the Arctic melt has not ‘nearly stopped’, in fact it’s continuing even faster than last year as shown by Aaron Wells graph from CA: http://img352.imageshack.us/img352/4760/seaiceextentlossratebi9.jpg

    Although you claim the the above, I could not find the link anywhere on climateaudit.org that led to the above. Could you provide a definitive URL on CA that leads to the above?

    Thanks!

    Jack Koenig, Editor
    The Mysterious Climate Project
    http://www.climateclinic.com

  71. Steven Goddard says:

    Phil,

    A bit more on your post. The graph of melt you sent over shows an average of about 50,000 km2 / day. If it kept that up through the end of the month – which it won’t – the total would only be a loss of an additional 8% before the season ended.

    You made the comment that I should be looking at area rather than extent. That is a more difficult calculation and beyond the scope of a newspaper article. According to CT, the minimum area was hit on August 16 or 17 last year.
    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.jpg

    Temperatures around the North Pole have dropped well below freezing the last couple of days. NOAA Buoy 30065 is reporting -4.3C and their North Pole Buoy is reporting -1.5C.

    I did a comparison of the CT map projection vs. Google Earth. It appears that CT uses a polar projection taken from an altitude of about 10,000 miles over the North Pole. Unlike the NSIDC extent maps, they do not attempt to elongate lower latitudes – instead preserving the view that a satellite would see from 10,000 miles. That may be what is causing some readers here to be confused.

    Apologies – I meant to say William Chapman in my last post, rather than “Phil” Chapman.

  72. Steven Talbot says:

    Steven Goddard,

    However, you have yet to make any attempt to analyze the methodology I used – which (obviously) assumes the correctness of the maps.

    Sigh. I have demonstrated that your premise of 10% above 2007 (on the basis of the graph) was incorrect, and that the correct figure was 13.5%. Sorry to repeat myself, but you seem to be ignoring this incontrovertible fact.

    If your name were Hansen or Mann you would be eviscerated here for making such an error, and for avoiding recognition of it. As it is, you’ve received plenty of messages of ‘support’, since your article seems to fuel the notion of data being cooked. Such is the nature of ‘scepticism’ when one gets a proper look at it, I guess.

    But I give up. I didn’t really imagine that you would want to respond to ‘review’ in the way that we (quite rightly) expect scientists to do.

  73. Claire Solt PhD says:

    It is interesting that a self-described warmer objects to internet publication when it was just such quick fast communication that birthed the internet and moves science along much faster than the peer review publication system can.

  74. Glenn says:

    Ed:
    “Goddard “error” (i’ll use this euphamism for what he really did) is obvious even to the layman. He totally ignores the scale on the image that shows the various colors represent different levels of sea ice coverage. Red is 60% while purple is 100%. Goddard treats them equally but it is clear from inspection the current year has more area with less % coverage – so goddard’s primitive calculation is bound to be way off.

    Its pretty funny how you guys can question the NSDIC’s numbers based on this obviously bogus analysis.”

    Perhaps you should take a look at the NSIDC graph header: “Area of ocean with at least 15% sea ice”. Why should Goddard not treat the colors equally?
    And why is his a bogus analysis? The Cryosphere maps are or should be based on real data, and that data should match data that other analyses are based. My understanding is that Steven has attempted to pull data out of the Cryosphere map and compare that data to the NSIDC’s, and I see nothing in principle wrong about that concept.

  75. Steven Talbot says:

    Glenn,

    “My understanding is that Steven has attempted to pull data out of the Cryosphere map and compare that data to the NSIDC’s”

    It’s the same data. It all comes from the NSIDC. That is why Goddard’s comments about comparing different sources make no real sense.

    Look, the NSIDC data is expressed in a graph (which Goddard has mis-described in terms of difference to 2007). Then it’s represented in a map. Then Goddard counts up pixels and thinks there’s a discrepancy. And all the ‘sceptics’ leap to the conclusion that someone’s cooking the data somewhere. But it’s the same data!

    If NSIDC were into the game of cooking their data, then the map would be cooked as well as the graph, no?

    Darn, I thought I’d given up. I really, really do now, promise ;-)

  76. Steven Talbot says:

    But I have to break my promise already, since I realise I haven’t yet commented on the ‘leap year trick’.

    1)Goddard’s premise is that the graph shows 10% greater for the same date. This is false, as I have demonstrated: it is actually 13.5% greater.

    2) He then compares this with the ‘map’ renditions which he offsets on the grounds that it’s a leap year. Er, hang on a minute, wasn’t it a leap year when you were looking at the graph???????

    3) So, he already i)has a false premise and ii)is not comparing the same days, the difference being in favour of the discrepancy he wants to display!

    Frankly, how gullible does one have to be to be classed as a true ‘sceptic’?

  77. Smokey says:

    All the impotent handwaving by the climate deceivers disregards these facts:

    click1

    click2

    click3

    click4

    click5

    click6

    click7

    Mr. Goddard has fully answered his critics. In globaloney parlance, his analysis is “robust.” The links above support Goddard’s article.

    Want more? I got ‘em. Just ask.

  78. Glenn says:

    Steven,

    Are you sure both Cryosphere maps and NSIDC charts use the same data?

    Here’s the closest I could come to data sources
    http://nsidc.org/data/collections.html
    http://arctic.atmos.uiuc.edu/SEAICE/arctic.historical.seaice.doc.txt

    Looks like several satellites and other methods are available.

  79. Jeff says:

    Jeff C.

    “And I would add The South pole sea ice trend is toward a larger sea ice extent/area. Why isn’t that what’s really important?”

    If there is a real trend, it is very small in terms of a percentage of the sea ice amount. And it certainly is not due to cooling over the oceans surrounding Antarctica; in fact, the atmosphere appears to have warmed over the oceans surrounding Antarctica in the past few decades. Perhaps the warmer air, which can hold more water vapor, is leading to an increase in the snowfall which in turn is leading to an increase in sea ice.

  80. Jeff says:

    I’m enjoying the irony that GW skeptics have been claiming that the small sea ice extent in 2007 had more to due with an unusual wind regime in the Arctic than with atmospheric warming and now they’re claiming that the greater amount of sea ice in 2008 vs. 2007 proves that the atmosphere isn’t warming.

  81. Patrick Henry says:

    Jeff,

    An increase in snowfall can increase the thickness of sea ice, but it can’t increase the area. of the ice. If snow falls on open water, it melts immediately.

    The only thing which can increase sea ice area is cold temperatures.

  82. randomengineer says:

    Jeff — “I’m enjoying the irony that GW skeptics have been claiming that the small sea ice extent in 2007 had more to due with an unusual wind regime in the Arctic than with atmospheric warming and now they’re claiming that the greater amount of sea ice in 2008 vs. 2007 proves that the atmosphere isn’t warming.”

    Say, have a nice steaming cup of strawman with that enjoyment — on me!

    What people here are responding to is the contention that 2008 was going to OBVIOUSLY be worse than 2007 because all of that 2008 ice was just “first year ice” and not the reliable crusty thick multi-year stuff.

    The fact that 2008’s extent isn’t significantly BELOW the 2007 number because of the aforementioned claimed problem of lacking multi-year ice is simply remarkable. So much for the predictions of a worse year. The AGW alarmist’s 2008 predictions weren’t merely just wrong, they were in a class of abysmal prognostication skill that ranks up there with psychics. Witch doctors with chicken guts or orangutans with darts could have fared better.

    Oh, and it wasn’t the skeptics claiming things about wind and current. They got that from this outfit called NASA. You may have heard of them.

  83. omnologos says:

    I disagree with Patrick Henry.

    Another factor that can increase sea ice area is the relative calmness of the sea and of the air above. Ice can be broken up mechanically. Keeping temperatures unchanged, stronger weather systems can result in larger areas of open water.

    Actually, one wonders if the decrease in sea-ice area is anthropogenic indeed, but due to the combined action of hundreds of cruise vessels freed up by the collapse of the Soviet Union. In Antarctica, in fact, the decrease in sea-ice is strictly limited to the Pensinsula, where too a number of cruises is organized to travel to 8-)

  84. Steven Goddard says:

    Here is an update from the Canadian Weather Service showing new ice rapidly forming in the Northwest Passage.
    http://awberrimilla.blogspot.com/2008/08/yesterdays-new-ice.html

    The Australian sailors attempting to navigate it are having a very difficult time with the ice , snow and cold, and are stuck in a snowstorm with strong northerly winds. Hopefully they won’t become the next victims of Arctic warming media hype.

    Jeff,
    Your logic is again flawed. NSIDC predicted that Arctic Ice extent would >12% lower than last year. Even using their highly questionable measurements, they are off by at least 25%. The point of the article is that the experts were wrong. The North Pole did not melt and the Arctic is not ice free.

    Steve Talbot,
    As far as the 10% vs. 13% goes, it sounds like you are making the argument that the NSIDC graph does not accurately represent their published data, showing less ice than there actually is. Regardless, it is irrelevant as the maps now (August 16) show 40% greater ice extent than 2007. The point is that there is a large disconnect between the maps and the graph. I don’t see any significant difference between the NSIDC maps and the UIUC maps, so the mystery remains unsolved.

    All of the expert predictions of Arctic ice collapse in 2008 were dead wrong. The NSIDC maps and the NASA satellite images clearly show that there is a lot more ice than last year. If people can’t see that, then it must be because they don’t want to.

  85. Steven Talbot says:

    Glenn,

    Are you sure both Cryosphere maps and NSIDC charts use the same data?”

    Well, according to Steven Goddard Cryosphere uses NSIDC data. He wrote in the article:

    “A comparison of these maps (derived from NSIDC data) below shows…

    Is this just another mistake that needs correcting?

    Eyeballing the 8/15/08 daily maps from Cryosphere and NSDIC suggests very similar representation of coverage, although the resolution’s lower on the NSDIC map –

    http://arctic.atmos.uiuc.edu/cryosphere/NEWIMAGES/arctic.seaice.some.000.png

    http://nsidc.org/data/seaice_index/images/daily_images/N_daily_concentration_hires.png

    Some more work there for the pixel counters!;-)

    It’s interesting to compare the Cryosphere hi-res image above with the Cryosphere side by side comparison screen for August 15th

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=15&fy=2007&sm=08&sd=15&sy=2008

    That looks kinda different to me. One test of the pixel-counting methodology would be to compare these two images for the same day from the same site. Not something I can be bothered to do, since I am not in pursuit of a conspiracy theory, but I have a hunch that some pixels would get counted as ice cover from the smaller image but not from the larger. But this is all silly stuff really, when the article’s premise has a 35% error anyway ( I note that when it comes to a piece like this another poster considers such an error margin to be “robust”!).

  86. Phil. says:

    Steven Goddard
    “Phil and Steven Talbot,
    If the UIUC maps are correct, there is nothing in my extremely straightforward calculations which could be broken. You seem to be (rather obtusely) attempting to make an argument that the UIUC maps do not correctly map the satellite data on to the polar projection they use. Polar map projections have minimum distortion at the pole, where the land is perpendicular to the viewer.”

    No you ‘rather obtusely’ continue to deny the facts regarding the maps being used, I shouldn’t have to tell you three times, your assumption regarding the projection used is incorrect! See below for detail on the projection used.

    If you wish to argue that the UIUC maps are incorrect, that might be an interesting topic of discussion. However, you have yet to make any attempt to analyze the methodology I used – which (obviously) assumes the correctness of the maps.

    The maps are correct, they just aren’t what you assume them to be.

    According to the UIUC maps, there has been almost no change in the amount of Arctic ice since August 8. Once again, if you believe those maps are incorrect – please state your case – and I suggest that you take it up with Phil Chapman over at CT.

    Just about 2%/day!

    As you can see below NSIDC are performing much more accurate calculations than your pixel counting on the low resolution graphic at CT (as mentioned elsewhere in the thread why didn’t you use the high res graphic?).

    “The polar stereographic projection specifies a projection plane or grid secant to the Earth at 70 degrees. The planar grid is designed so that the grid cells at 70 degrees latitude are 25 km by 25 km. For more information on this topic please refer to Pearson (1990) and Snyder (1987). This projection often assumes that the plane is tangent to the Earth at the pole. Thus, since there is a one-to-one mapping between the Earth’s surface and the grid at the pole, there is no distortion at the pole. Distortion in the grid increases as the latitude decreases, because more of the Earth’s surface falls into any given grid cell, which can be quite significant at the edge of the northern SSM/I grid where distortion reaches 31 percent. For the southern grid, the SSM/I grid has a maximum distortion of 22 percent. To minimize the distortion, the projection is true at 70 degrees rather than at the poles. This increases the distortion at the poles by three percent and decreases the distortion at the grid boundaries by the same amount. The latitude of 70 degrees was selected so that little or no distortion would occur in the marginal ice zone. Another result of this assumption is that fewer grid cells will be required, as the Earth’s surface is more accurately represented. This saves about 100 megabytes per year in data storage.
    The polar stereographic formulae for converting between latitude/longitude and x-y grid coordinates have been taken from map projections used by the U.S. Geological Survey (Snyder 1982). Several different ellipsoids were compared to the Hughes ellipsoid and in each case, differences were less than 1 km over the SSM/I grids. However, differences of up to 9 km were found if a sphere rather than an ellipsoid was used. Thus, it is an explicit requirement that an ellipsoid be used in processing the data.
    An ellipsoid is defined by equatorial radius and eccentricity. The ellipsoid used in the Hughes software assumes a radius of 3443.992 nautical miles or 6378.273 kilometers (km) and an eccentricity (e) of 0.081816153. To properly convert these coordinates to a polar stereographic grid (the projection of choice), the conversion should assume the Hughes ellipsoid.”

  87. Phil. says:

    Steve Goddard:
    “Your logic is again flawed. NSIDC predicted that Arctic Ice extent would >12% lower than last year”

    Care to say where they said this? I’ve been unable to find it, however I did find the following.

    NSIDC comment from April:
    “Despite strong growth of new ice over the winter, sea ice is still in a general state of decline. The ice that grew over the past winter is relatively thin, first-year ice that is susceptible to melting away during the summer. Although natural variability in the atmospheric circulation could prevent the ice pack from breaking last year’s summer record, a closer look at sea ice conditions indicates that the September 2008 minimum extent will almost certainly be well below average.”

  88. Phil. says:

    Although you claim the the above, I could not find the link anywhere on climateaudit.org that led to the above. Could you provide a definitive URL on CA that leads to the above?

    Thanks!

    Jack Koenig, Editor
    The Mysterious Climate Project
    http://www.climateclinic.com

    Here you go: http://www.climateaudit.org/?p=3336#comment-289313

  89. Jeff says:

    Steve Goddard wrote
    “Your logic is again flawed. NSIDC predicted that Arctic Ice extent would >12% lower than last year. Even using their highly questionable measurements, they are off by at least 25%. The point of the article is that the experts were wrong. The North Pole did not melt and the Arctic is not ice free.”

    A pathetic attempt to divert attention from the issue. Steve Goddard claimed that NSIDC’s was claiming that the sea ice EXTENT was 10 percent greater when in fact it actually was 30 percent. I showed that the actual difference in AREA was 13 percent (I incorrectly stated 14 percent yesterday), not the 30 percent that Mr. Goddard mysteriously came up with, meaning that NSIDC’s claim of an approximately 10 percent difference in EXTENT is highly plausible. Now he trying to obscure the issue by bringing up an alleged forecast of a 12 percent decrease that he claims that someone made and stating that there was a 25 percent error in the “forecast.”

    Mr. Goddard now is claiming that the measurements are “questionable”. The algorithm has been published, so I’m interested in Mr. Goddard’s take on what is wrong with it.

    And of course he repeats the lie that the “experts” forecast that the North Pole would be ice free this summer. He then invents a whole new lie that there was a forecast that the arctic would be ice free. What will he claim next?

  90. Jeff says:

    Patrick Henry
    “The only thing which can increase sea ice area is cold temperatures.”

    Wrong. Winds and currents play a role. If you can find a data set showing that temperatures around (as opposed to over the interior) of Antarctica are cooling, please post it. Every one that I’ve seen has shown warming.

  91. Steven Goddard says:

    Phil,

    You keep describing the NSIDC map projection (ad nauseum,) which is not the one that CT uses in their maps and not the map set that I used. Rather, I used the CT maps which are higher resolution than the NSIDC maps, have more detail, and are archived every day. No matter how many times you describe the NSIDC projection, it doesn’t have any impact on this discussion.

    As I have attempted to explain to you before, the CT projection is the view of an astronaut 10,000 miles over the North Pole. The only point which is undistorted is right at the pole. As you move away from the pole, the distortion increases approximately as the sine of the latitude. (At 30 degrees latitude, a pixel would represent 2X as much area as one at the pole.) Thus my calculations are conservative, because the actual amount of ice increase away from the pole is greater than the pixel count. This is not a question of pixel precision, which has very little impact on the 30% number.

    NSIDC forecast in May that extent would break last year’s minimum record by about 13% at 3.59 Mkm2. See figure 4.
    http://nsidc.org/arcticseaicenews/2008/050508.html
    If you accept their current +14% number, that is a misprediction of 27%.

  92. Werner Weber says:

    From the UIUC data you can add up all polar sea areas covered by ice: I get 3.3 million km2 for August 16, 2008. The IJIS value (15 % ice coverage and more) of 08/16 is 5.91 million km2, a change of 0.07 million km2 from 08/15. The NSIDC value is very similar. The corresponding numbers for 2007 are 0.7 to 0.8 million km2 lower, both for IJIJ, NSIDC and for UIUC (the latter is estimated ‘by visual inspection’).
    In Germany, the low 2007 ice coverage was seen by the AGW protagonists as the tipping point, beyond which any skepticism was no longer allowed or should even be forbidden.

  93. Jeff says:

    randomengineer, you only need to go back to the Oct. 2007 “issue” to see where our host was playing the 2007 record minimum was due to winds not warming card so the decrease in 2007 was not evidence of global warming. If the decrease from 2006 to 2007 was not proof of warming then why is the increase from 2007 to 2008 proof of global cooling?

  94. Steven Goddard says:

    Steven Talbot,

    I’m keeping a running count of 2008 vs. 2007 from the full resolution (850×850) CT maps. August 15, 2008 shows 41% greater extent than August 16, 2007.

    An interesting statistic is that according to CT, Arctic ice area bottomed out by August 17 last year – which means that ongoing losses in extent after that date were probably due to compaction, rather than melting.

  95. Glenn says:

    Steven Talbot said
    “Glenn,
    “Are you sure both Cryosphere maps and NSIDC charts use the same data?”
    Well, according to Steven Goddard Cryosphere uses NSIDC data. He wrote in the article:
    “A comparison of these maps (derived from NSIDC data) below shows…
    Is this just another mistake that needs correcting?”

    I’d say so. You didn’t answer the question, and it was you who insisted on the data originating from the same source. Not sure whether the above quote from Goddard is correctly interpreted as a claim that the Cryosphere
    map is derived from NSIDC. It may be that he used assigned the map area found to the NSIDC numbers – his comparison used the NSIDC data, from the NSIDC chart.
    Another mistake of yours is the attempt made (02:37:07) to confuse sea ice area with arctic ocean area with a minimum sea ice concentration. Both
    the NSIDC graph and the Cryosphere maps depict ocean area, and is what Goddard has compared.
    Another is your criticism of Goddard’s 10% figure, which is much closer than your 13.5%, upon analysis of the NSIDC graph itself that Goddard used.
    I get a 2007 minimum for the dated graph of 5.7 mil, and the 2008 area as 6.3 mil. 10% of 5.7 added to 5.7 is 6.27, real close to being 6.3 mil. You used some quote from an NSIDC webpage to make your calcs. So you again used a different source to refute a claim based on a different source.
    Yet another is the big deal you made of a leap day, and it isn’t a big deal at all. IMO it is quite fair and reasonable to compare “same day as last year” as 365 days ago, not 366 or 3 or 498.

    It appears that Goddard has a point, whether the data source is the same or not. And I don’t need to count pixels, just use a pencil as a straightedge on the monitor looking at the Cryosphere maps. It is most obvious just eyeballing that quite a bit more area is covered than a half mil from the same time last year. And his 30% more probably isn’t far off if at all.

  96. Jeff says:

    Steven Goddard
    “You keep describing the NSIDC map projection (ad nauseum,) which is not the one that CT uses in their maps and not the map set that I used. Rather, I used the CT maps which are higher resolution than the NSIDC maps, have more detail, and are archived every day. No matter how many times you describe the NSIDC projection, it doesn’t have any impact on this discussion.”

    The CT maps might contain more pixels than the NSIDC maps but that doesn’t change the fundamental resolution of the data. It just creates an image that is larger on a computer monitor.

  97. Glenn says:

    “The CT maps might contain more pixels than the NSIDC maps but that doesn’t change the fundamental resolution of the data. It just creates an image that is larger on a computer monitor.”

    You lost me there, Jeff. The number of pixels IS resolution. And i fail to understand how “larger on a monitor” is relevant. Your monitor bigger than mine, or what.

  98. Steven Talbot says:

    Glenn,

    You didn’t answer the question, and it was you who insisted on the data originating from the same source.

    As I have indicated, I was naive enough to accept that Goddard’s statement was accurate. This confused me until you made clear that it could not be. Yes, I was insisting the data was the same source, since that was what Goddard had stated. My argument was proceeding from his stated premises.

    Not sure whether the above quote from Goddard is correctly interpreted as a claim that the Cryosphere map is derived from NSIDC.

    Huh? The statement was “these maps (derived from NSIDC data)”. That seems a pretty clear process of interpretation to me, Glenn!

    Another mistake of yours is the attempt made (02:37:07) to confuse sea ice area with arctic ocean area with a minimum sea ice concentration. Both
    the NSIDC graph and the Cryosphere maps depict ocean area, and is what Goddard has compared.

    Yes, I accept that, though it wasn’t an “attempt to confuse”, but a mistake. Please note that I accept when I have made a mistake, unlike some others.

    Another is your criticism of Goddard’s 10% figure, which is much closer than your 13.5%, upon analysis of the NSIDC graph itself that Goddard used…..You used some quote from an NSIDC webpage to make your calcs. So you again used a different source to refute a claim based on a different source.

    No, that is wrong – it is the same source, the 11th August press release, to which Goddard refers. The graph was not released independently but as part of that press release, which includes the actual figures (in preference to either your or Goddard’s eyeballing). It seems pretty obvious to me that the graph does not have the resolution to compute percentage differences accurately and that one should refer to the figures which were given.

    Yet another is the big deal you made of a leap day, and it isn’t a big deal at all. IMO it is quite fair and reasonable to compare “same day as last year” as 365 days ago, not 366 or 3 or 498.

    You seem to have missed my point entirely. That may be reasonable in itself, but Goddard only applied this principle in comparing maps. With the graph he compares the same calendar date. Therefore, the offset was in favour of showing a greater differential. (Please note that I have accepted a mistake above. This is a mistake, or a failure to understand the point, on your part).

    It appears that Goddard has a point, whether the data source is the same or not. And I don’t need to count pixels, just use a pencil as a straightedge on the monitor looking at the Cryosphere maps. It is most obvious just eyeballing that quite a bit more area is covered than a half mil from the same time last year. And his 30% more probably isn’t far off if at all.

    It’s certainly possible that there is a point to be made about how effective these maps are as a visual representation of sea ice extent. Not very, I think. But the map is not a data source but rather a representation of a data source (just as a graph is not a data source). Goddard stated “The problem is that this graph does not appear to be correct. Other data sources show Arctic ice having made a nice recovery this summer…”. Three points here: 1) there are no grounds to presume that the graph is “not correct” rather than that the maps are “not correct” and 2) the NISDC maps exhibit the same visual coverage as the Cryosphere ones, so the insinuation that “other sources” show the graph to be “not correct” is unsound and 3) Goddard ignores the actual data sources, which were available.

    The article allows the inference that someone is “cooking the data” in a biased fashion. IMV, that was its intent. IMV also, that is unjustified.

  99. Brian D says:

    I wonder how this will affect the Artic region in the near term. The sulphur cloud from Kasatochi is circling the Artic right now. And I have been hearing of cold air building a little early in the Artic, also.

    2008-08-16 04:49:20
    Volcanic unrest continues at Kasatochi volcano. Seismic activity is still being detected on the AVO seismic network located on Great Sitkin Island, 25 miles to the west.

    AVO has received no new reports of activity or ash emissions. A sulfur dioxide cloud produced by the Aug. 7 eruption remains at high altitudes in the circum arctic region.

    http://www.avo.alaska.edu/image.php?id=15049

    Apparently this cloud is the one of the largest observed since 1991 Hudson eruption in Chile.
    http://earthobservatory.nasa.gov/Newsroom/NewImages/images.php3?img_id=18118

    I understand this will have no long term effect because it didn’t get high enough, but it can create an aerosol haze that, in the short term, could reflect the little sunlight that is now entering the Artic region. And some mets are talking about an early build up of cold air. Models do seem to suggest this. Could we possibly see an early ice extent minimum this year? By about 2 weeks, maybe?

    I guess it’s a wait and see.

  100. Brian D says:

    I just found this site. Daily SO2 images from around the globe.
    http://satepsanone.nesdis.noaa.gov/pub/OMI/OMISO2/index.html

  101. Steven Goddard says:

    Resolution of the image has nothing to do with the monitor. Most Image files contain a sequence of bytes representing the location, color, transparency and brightness of the pixel. The CT image files have a resolution of 850×850 which is 722,500 pixels. The calculations are done by software analyzing each of the pixels. You don’t need any monitor to do this analysis.

    I don’t believe that anyone is doing anything intentionally wrong with the data. There is an inconsistency between the CT images and the NSIDC published extent numbers. There also is an inconsistency between the CT images and the CT area measurements – which are likely derived from someone else’s extent numbers. I’m keen to find out what the source of the discrepancies are. From comparing with satellite images, my sense is that the CT images are golden.

    My beef with NSIDC is that they have a habit of taking spectacular stories to the press, which get misquoted and propagated in high places. They should take a more conservative approach to dealing with the press. They should also be more conservative in their predictions. Scientists should be careful to avoid the appearance of pursuing an agenda, as it potentially affects their objectivity.

    Suppose that 2008 had of set another record melt. Why couldn’t they wait until September to discuss it, rather than making high profile predictions to the press in May? It is not normal behaviour for scientists to be so impatient.

  102. Steven Talbot says:

    Steven Goddard,

    I don’t believe that anyone is doing anything intentionally wrong with the data.

    I’m very pleased to hear that, and thank you for making that statement. You will be aware that many others on this thread have leaped to a different conclusion.

    Suppose that 2008 had of set another record melt. Why couldn’t they wait until September to discuss it, rather than making high profile predictions to the press in May? It is not normal behaviour for scientists to be so impatient.

    Their statement in May was:

    Taken together, an assessment of the available evidence, detailed below, points to another extreme September sea ice minimum. Could the North Pole be ice free this melt season? Given that this region is currently covered with first-year ice, that seems quite possible.

    I think that remains possible, FWIW, though I don’t think it would prove anything one way or another, and I wouldn’t bet on it. The concern is with a falling trend rather than what happens in a particular year.

    It is possible that some scientists are “impatient” not because they are agenda-driven (not because they are part of a massive leftist conspiracy to raise taxes on the back of fake science, yada yada…) but because they are genuinely worried, on the back of their scientific understanding. Of course, they might be wrong, but I won’t personally object to scientists seeming impatient if their best judgment is that there are pressing concerns to address.

  103. Glenn says:

    Steven Talbot,

    “As I have indicated, I was naive enough to accept that Goddard’s statement was accurate. This confused me until you made clear that it could not be. Yes, I was insisting the data was the same source, since that was what Goddard had stated. My argument was proceeding from his stated premises.”

    If you are referring to the data sources being the same, I did not make it clear that it could not be. I asked you for your evidence, since you had employed the assumption in a reasoned argument against Goddard’s claim.
    This issue is still unresolved, but:

    “Huh? The statement was “these maps (derived from NSIDC data)”. That seems a pretty clear process of interpretation to me, Glenn!”

    is a quote mine. To use your partial quote from a previous post:
    “A comparison of these maps (derived from NSIDC data) below shows…”

    Again, it just isn’t clear whether Goddard meant that the comparison was derived from, or employed, NSIDC numbers, or whether he is claiming that the maps themselves are derived from NSIDC data. Nor should it be an issue of whether the same source of data was used for both.

    “That may be reasonable in itself, but Goddard only applied this principle in comparing maps.”

    I didn’t miss the point you claim is made here. The problem is that the maps are not the graph, and he was comparing the maps when he included a leap day. He could not have with the graph. You may be right that this favored a greater difference, but not a substantial one. Nit picking.

    “Just as a graph is not a data source”

    Now that is a hoot. Yes, Steven, graphs are sources of data. They aren’t satellite instruments, but representations of data allow data to be drawn from them. Were it not so, we should be thinking that we might as well be looking at cartoon reruns of Popeye. Data can reasonably be taken from the NSIDC graph, which is what I did, and resolution has little to do with quality here with respect to hundreds of thousands of square kilometers of ocean.

    As to your claim about NSDIC data, your source reports for Aug 10, not Aug 11: “Arctic sea ice extent on August 10″, and if you look at Goddard’s archived graph of Aug 11, this year’s line drops a bit at the last day. That in itself makes your data from the NSDIC site and Goddard’s different by some amount. So aren’t you as guilty here as you claim Goddard is with respect to the leap day inclusion?

  104. Glenn says:

    Steven Goddard,

    Wouldn’t these U of Illinois maps be generated by an algorithm? Programs, or models, are only as good as the programming.
    I share your concern for consistency in data, and from whatever source. Here’s just one other source that doesn’t match:
    http://www.iup.uni-bremen.de:8084/amsr/amsre.html

    So many different measurements leaves me with little confidence in any. I don’t want to accept this as good science. Global temps are shown down to
    the thousanth of a degree, yet different sources vary tremendously more, almost as much as the claimed increase for the past 30 years. Yet that increase is only said to be a fraction of a degree itself. Hard to gain confidence in any, but especially hard with the sources that employ ground stations next to air conditioners. I see absolutely no responsibility shown by the AGW crowd.

  105. This blog has put up two very different pictures of Antarctica temperature anomalies, both NASA origin it seems. Connolley believes the high-cooling one is the work of a “PR droid” (naturally) but I have other thoughts I put here.

    More discrepancies in polar data? Any thoughts?

  106. Steven Talbot says:

    Glenn,

    As for your first point, I think we must agree to differ in our interpretation.

    Your second point, that I am nit-picking regarding the leap year offset. Perhaps so – one would have to consider the actual data for the day’s difference in order to know. But in any case, I thought that nit-picking about data accuracy was something people here thought to be important?

    Now that is a hoot. Yes, Steven, graphs are sources of data. They aren’t satellite instruments, but representations of data allow data to be drawn from them.

    No, Glenn, graphs are not the sources of data. They are representations of data, as you rightly say. The maps are also representations. Any representation may have inherent error, or may need careful interpretation, as is the case with the maps, evident from associated notes, for example:

    Please note that our daily sea ice images, derived from microwave measurements, may show spurious pixels in areas where sea ice may not be present. These artifacts are generally caused by coastline effects, or less commonly by severe weather. Scientists use masks to minimize the number of “noise” pixels, based on long-term extent patterns. Noise is largely eliminated in the process of generating monthly averages, our standard measurement for analyzing interannual trends.

    http://nsidc.org/arcticseaicenews/disclaimer1.html

    You continue:

    As to your claim about NSDIC data, your source reports for Aug 10, not Aug 11: “Arctic sea ice extent on August 10″, and if you look at Goddard’s archived graph of Aug 11, this year’s line drops a bit at the last day. That in itself makes your data from the NSDIC site and Goddard’s different by some amount. So aren’t you as guilty here as you claim Goddard is with respect to the leap day inclusion?

    The comment and the graph were both archived on the 11th. Therefore, your point does not stand (i.e., the graph represents data to the 10th also). You simply underline the point that Goddard has compared data from different days!

  107. Jeff says:

    I now have my new-and-improved sea ice area calculation. I did 2 things differently:
    1) I limited the area to the “Arctic” by only looking for grid cells north of 63 degrees.
    2) I assumed that the “pole hole” in the NSIDC data has more that 15 percent sea ice (a rather safe assumption)

    I found 8467 grid cells with > 15 percent concentration on Aug. 12, 2007 and 9649 on Aug. 11, 2008. Since NSIDC also supplies a file with the area of each grid cell, I converted these to areas and came up with an area of 5523918 km^2 on Aug. 12, 2007 and 6281339 km^2 on Aug. 11, 2008.

    I’m still at about a 14 percent difference.

    REPLY: What was the rationale for 63 degrees N?

  108. Jeff says:

    For Aug. 11-16, I got the following sea ice areas (in km^2) from NSIDC’s data:
    6291339 6207404 6125202 6067534 5949942 5869797

    This appears to show that the sea ice is still decreasing, though my understanding is that data less than 5 days old are subject to revision.

    I’m trying to remember how I got started in this thread. One year changes in any quantity are usually not meaningful.

  109. Steven Talbot says:

    So many different measurements leaves me with little confidence in any. I don’t want to accept this as good science. Global temps are shown down to
    the thousanth of a degree, yet different sources vary tremendously more, almost as much as the claimed increase for the past 30 years.

    You might have a point if they were all purporting to be measuring in equivalent ways, but I am sure you are aware that they are not. The significant consideration is trend, and concern should be for the consistency of measurement within each set. Satellites, for example, exclude measurement of polar regions, the Himalayas and parts of the Andes, as I expect you know.

    I see absolutely no responsibility shown by the AGW crowd.

    Whereas I see that you’re happy to accept eyeballing a graph over calculating from raw data, accepting a pixel count of a mapped representation without assessing projection and disregarding the possible impact of spurious pixels, so long as it is supportive of the anti-AGW crowd?

  110. Charles says:

    The important point is not how this year’s minimum compares with last years record–on the log-term (by records standards, short term by geological or climatological standards.

    The ice which is there now, is not the same as the ice that covered the Arctic 3,4,5 years ago–this ice is largely very young ice. It also has been frozen out of water with a lower salinity than previously, thanks to last years minimum.

    These facts have implications such as: Lower salinity water freezes faster, younger ice is thinner and more fragile than older ice, so the ice will regrow faster as the salinity drops and the freezing point rises.

    While last year’s minimum was exciting because of how little ice there was historically (a very short period in climate studies,) what was most noteworthy last year was that it was also the first time since the 80’s that the minimum dropped for three consecutive years below the long-term average slope.

    The magnitude got most of the press, but it is the rapid increase in the rate of decline shown my the successive minimums which is of more concern.

    If this year’s minimum manages to be above the long-term slope, then perhaps, there is no reason for excitement. If there is a fourth year in a row below that average, then the function describing the curve may have hit an area where the output of the physics has changed.

    Frankly, basing any sort of reaction on such a small data point is an exercise in futility. The ONLY weather/climate predictions we can make that are reliable are for about 4 days in the future, getting excited either way because a prediction was right or wrong for this kind of thing is a valid as getting mad at the weather people because they said it would be a “normal” crop year and it was, except for that 2 week drought that timed just right to destroy the crop.

    By its nature, we will be uncertain of what is happening with the climate long-term until it has happened.

    Unfortunately, we MUST deal with the problems, permanent or temporary regardless of their duration.

    There is no doubt anywhere, that humanity and the Earth will survive this kind of crisis–we habitate too many different areas to be easily wiped out.

    But climate change will not kill us all the same way any more than overpopulation.

    Next summer, baring major changes in current weather trends, there will be many placed in the Equatorial zone who will suffer drought and famine–they do not know or care if it is global warming, when you are thirsty and hungry, you only watn water and food. This is the kind of situation which historically has led to wars and toppled governments.

    People faced with death tend to fight and the rules for fighting tend to go away.

    Of course, at present, we probably can feed and water nearly everyone we do today–If we are willing to do so. This may not remain true for more than a few months.

    It is also possible, but looks less likely, that the patterns could shift, the monsoons return, the droughts recede.

    The US SouthWest has lived for over 100 years knowing that they cannot sustain the lifestyle. Now they are in a major drought, but major droughts are historically common in that area, and they have often run for decades and in at least once case into centuries.

    The US SouthEast is also in a drought, that is not historically a normal pattern since settlement.

    In the Midwest, our rain and snow belt has shifted south by about 100 miles from 30 years ago, we are now edging to the other of our 2 modes in Western Wisconsin: we’re drying enough to edge to being prairie after a couple hundred years at least, of being forest.

    Using different models, difference measuring systems and differing algorithms to determine ice cover does obscure the subject and makes comparisons difficult to impossible, but there is no standard, recognized measure for ice coverage, and there is no 100% reliable algorithm to calculate. Remember that te original data used for calculation is very likely different in quality or dates, and even a small difference in how the program determines if pixel is ice or not could result in major varations.

    The difference between 10 and 30 % additional coverage seems significant, but given the experimental error in making the measurements, it may well be with observational error.

    The most important point of all seems overlooked:

    In the past 60 years we have learned that the “stable” planet upon which we live is not stable at all, and gradually we have learned that the large events of the past which we have been unable to nail down to within a century or two in many places, do not necessarily take 10’s 100’s 1,000’s or more years. Instead, we find increasingly that very large changes can happen in timespans considered short by individual humans–under a decade perhaps.

    Many of these changes are drastic enough to greatly damage our civilization. Some in the past have been enough to threaten our species.

    Such events seem to have happened in the past on a semi-regular cycle, and we have identified some things with cycles of orbital change and axial tilt, others appear to have no causative factor–mostly because we are probably very short on data.

    Can melting the Arctic cause an Ice Age? Many of our models, which we routine trust on other matters, say yes, it can. Our experts agree that it is “unlikely”–but use the model to predict other climate changes.

    What I know is that I know of no system, natural or designed which behaves according to a smooth function at all times. And I am CERTAIN that no system which involves positive feedback can remain a smooth function.

    The fastest glacier in the world is moving at 300 feet per day. We have no idea how fast they are capable of moving (the fastest clocked avalanche was 250mph) If the East Antarctic ice sheet started moving that fast, it would dump many, many cubic kilometers of ice into the ocean, which, coming from land, would displace their mass in seawater, raising the sea level.

    We know that the faster a glacier moves, the more cracks it develops, and the more cracks the more melt water gets to the bottom, the more water, the faster the glacier.

    Similarly, methane release from sea floor methane hydrate has been strongly implicated in at least one major extinction event. A major release could easily trigger enough warming to cause more releases.

    Round and round she grows..the story of positive feedback.

    Is the climate changing? Probably. What route will it take? We can only try to predict and make educated guesses. Will it cost us to attempt to slow the now apparent changes? Yes. But the real question is: How much will it cost us if we fail to act and or predictions come true? That is the question which determines how much we need to invest in insurance and that includes efforts to slow or stop the change.

    The main possible events I see as possible, in order of desirability:
    1) The Arctic ice melts and triggers a mini-ice age, reversing the warming trend enough to bring us back to a climate which is close to what we consider “normal.”
    2) We have an asteroid, meteor shower, or comet hit us, or volcano which is large enough to affect the climate and small enough not to devastate us.
    3) A limited nuclear war. Recent calculations from 2007 show that a limited nuclear exchange would have similar climatic effects to a full MAD scenario.
    4) The melt continues, when Greenland has lost about 1/2 of her ice, the West Antarctic sheet breaks free and slides into the Pacific. This combination raises sea level around 7 meters. Warming continues and the East Antarctic Sheet begins to accelerate and drop hundreds of cubic kilometers of ice each season into the sea–perhaps eventually a single very large event.
    5) The methane hydrate melts and accelerates the melting of other such reserves and we have an extinction level event similar to 55 million BCE.

    I favor 1-3 because they are the scenarios which cost the least number of lives.
    #4 is survivable for the civilization in a drastically altered form, as a large percentage of our population would die. #5 would stand a good chance of destroying our civilization, if not the species.

    At this point, our simulation model, Project 6, says that #3 is most likely, and sometime next year. Depending upon harvests and the US foreign policy among other factors.

    The disadvantage of foretelling the future is not in being wrong, but in being right and unable to stop or change the outcome.

  111. Jeff says:

    Steven Goddard
    “I’m keeping a running count of 2008 vs. 2007 from the full resolution (850×850) CT maps. August 15, 2008 shows 41% greater extent than August 16, 2007.”

    Using the actual data, I get a difference of about 13.6 percent.

  112. Jeff says:

    Glenn
    “You lost me there, Jeff. The number of pixels IS resolution. And i fail to understand how “larger on a monitor” is relevant. Your monitor bigger than mine, or what.”

    If after a snowstorm I were to measure the snow depth at 1 kilometer intervals and then interpolate the data to a 100-meter resolution grid, would that be the same as measuring the snow depth at 100-meter intervals?

  113. Glenn says:

    Steven Talbot,

    Hopefully you are aware that alphanumeric characters are only representations of data as well, and not the true source. Recording, reproducing and representing data can all suffer from error. Sheesh, even the data source can and and often does contain error. That doesn’t mean that a graph isn’t a source of the data used to create it. Again if that were so, graphs would not be used. Some representations are better than others, of course. Your complaint seems to be rooted in the belief that these graphs and maps are error ridden. If so, then it would be immoral to publicize them. Your claim that graphs aren’t data sources is not believable, since you realize that they are representations of data.

    As to the graph and your quote from the NSIDC page, all I know is that archiving is done when it is done, that the graph itself is said to be updated daily online, and when I read “On Aug 10th…” I take it that the event described is represented as having occured on that date. The graph URL is dated Aug 11 and is Goddard’s reference, and the data you used to falsify his conclusions was said to be from Aug 10, and resided on a separate webpage. Clearly you used different data.

  114. Smokey says:

    Compare last year’s sea ice on this date with today’s sea ice: click [this page takes a few seconds to load].

    I prefer looking at the maps rather than engage is this endless true believer nitpicking. So, who are you gonna believe? The maps of sea ice, separated in time by exactly one year? Or your lying eyes? Personally, I believe what my eyes tell me. Others have their religion to guide them.

    And this whole discussion centers around the Northern Hemisphere. Note that the Southern Hemisphere has significantly more sea ice than in previous years. Added together, global sea ice is rapidly increasing. Inconvenient, huh?

  115. Glenn says:

    Steven Talbot,

    “You might have a point if they were all purporting to be measuring in equivalent ways, but I am sure you are aware that they are not. The significant consideration is trend, and concern should be for the consistency of measurement within each set. Satellites, for example, exclude measurement of polar regions, the Himalayas and parts of the Andes, as I expect you know.”

    The significant consideration is accuracy, my friend. You are right that there are different methodologies for measuring, and comparisons should only be made with data measuring the same thing(s). Yet different methodologies should generate the same data, since the reality is that there is only one Earth and one “global temperature”. If you feel comfortable will all this inconsistency, I think you are being just as irresponsible as the experts who generate it. If I hire two people to come into my house and make termperature measurements, I expect they will both arrive at the same result, or both be considered wrong and their data discarded. I’m puculiar about *measurements* that way, and science is or should be especially so.

    Me: “I see absolutely no responsibility shown by the AGW crowd.”

    “Whereas I see that you’re happy to accept eyeballing a graph over calculating from raw data, accepting a pixel count of a mapped representation without assessing projection and disregarding the possible impact of spurious pixels, so long as it is supportive of the anti-AGW crowd?”

    Excuse me, but I’ve offered no conclusions beyond the maps being inconsistent with the graph. This can not be seen as being supportive of anything other than consistency. Regardless, it appears that the Arctic melt is largely if not exclusively a result of events not related to “global warming”, so for me this issue has nothing directly to do with being supportive of the “anti-AGW crowd” or critical of the “AGW crowd”.

    But making claims about what I “happily” haven’t done or have disregarded, in the absence of evidence of that, isn’t quite honest. It’s a tactic I expect from an AGWer. Are you one, or rather do you accept the “projections” made by the various AGWers that spell doom and gloom if we don’t all stop drivin that hot rod Lincoln?

  116. Phil. says:

    Steven Goddard (11:52:39) :
    Phil,

    You keep describing the NSIDC map projection (ad nauseum,)

    Three times!

    which is not the one that CT uses in their maps

    According to you, although CT uses the NSIDC data which is archived in the polar stereographic format that I described.

    and not the map set that I used. Rather, I used the CT maps which are higher resolution than the NSIDC maps, have more detail, and are archived every day. No matter how many times you describe the NSIDC projection, it doesn’t have any impact on this discussion.

    As I have attempted to explain to you before, the CT projection is the view of an astronaut 10,000 miles over the North Pole.

    Elementary trigonometry shows us that that is not the projection that CT uses! (hint the earth’s radius is 3963 miles)

    The only point which is undistorted is right at the pole. As you move away from the pole, the distortion increases approximately as the sine of the latitude. (At 30 degrees latitude, a pixel would represent 2X as much area as one at the pole.)

    I guess trig isn’t your strong point! (30º isn’t visible on the CT map either, although it would be if they used the projection you claim.)

  117. Jeff says:

    Glenn
    “I’ve offered no conclusions beyond the maps being inconsistent with the graph.”.

    Ignoring the data is a tactic that I expect from an [snip- you don't get to use that word here]. NSIDC’s graph is consistent with the data. I’ve posted the location of the data, so you can do your own count. NSIDC’s graph is also consistent with the “recent ice area” on the CT website.

    The only “evidence” to support the “conclusion” that NSIDC’s graph is inconsistent with the maps is someone’s claim to have counted pixels on re-projected images created using an unspecified algorithm (actually, there were likely multiple interpolations. NSIDC sea ice data is stored in polar stereographic grids. I’m guessing that the data first had to be interpolated to a lat-lon grid and then interpolated again into the image space).

    Personally, I tend to believe objective data rather than someone’s guesstimate. But to each his own. Some people prefer to believe that what they believe is true. Then again, when a friend showed me the maps I guesstimated that there was about a 15 percent difference. Maybe my eyes are just better calibrated than most peoples’.

  118. Jeff says:

    Smokey
    “Added together, global sea ice is rapidly increasing.”

    Really? You know this from where? You certainly didn’t find this out from the “Global Sea Ice Area” graph on the CT site.

  119. old construction worker says:

    Charles
    ‘At this point, our simulation model, Project 6, says that #3 is most likely, and sometime next year. Depending upon harvests and the US foreign policy among other factors.’
    And, of course, your model incorporates as many of the forecasting principals as possible as laid out in the Principals of Forecasting hand book. If not your model is nothing more than a reflection of your guestimation.

  120. dipole says:

    Anthony says in reply to counters (15:19:01)

    Using the color scale provided, and its corresponding RGB value, it is an easy and certain mathematical exercise to turn those pixels back into real data, with no loss, no rounding, no floating point error.

    Not so sure about that. I tried my own pixel count (using the Python image processing library). There are difficulties with the UIUC daily images as the map colours do not match the scale too well. This might be due to colour distortions from JPEG compression, or some other image processing artifact. It would certainly help if Steven Goddard explained his method in more detail so someone else could reproduce his results.

    To quantify of my own visual impression, I also tried the even less scientific method of using the free select tool in the Gimp image processing app to outline the ice extent, then measuring the selected area. The result surprised me – 20080817 was only 17.5% above 20070818. Just eyeballing the images I would certainly have expected a bigger difference.

    REPLY: Be careful how you save the image from the website. Doing it the wrong way can add an extra level of JPEG compression (from whatever tool used to save it) and may make the image far worse thatn the original. An image saved once with JPEG compression, but saved again as a bitmap image or other lossless format will only have one level of JPEG compression, and often that compression is reversable. – Anthony

  121. Steve says:

    If this story gets 1% the coverage the original alarmist story got I will be shocked.

  122. Evan Jones says:

    It doesn’t make a bit of difference whether this year’s sea ice extent is slightly more or slightly less than last year’s.

    What does make a difference is whether or not NSIDC is accurately reporting the data.

    It also makes a difference if there is a strong reversal in trend, even if it is only over the course of a year. Especially as it coincides with outside factors such as PDO and AO reversal.

    As to what direction the trend takes from this point we simply don’t know now. In the meantime it would be awfully nice to be getting a correct graph of ice extent from government agencies.

  123. Glenn says:

    Dipole,

    Remember Goddard’s comparison is for a specific time, Aug 11, not the 17th.

  124. dipole says:

    Anthony replies to my previous post (22:39:44)

    Be careful how you save the image from the website.

    I am using wget from the command line (on a linux system), so I assume I have an exact copy of the image on the uiuc site. I suspect the colour distortion is introduced by whatever software uiuc use to produce the image.

    REPLY: Ok that should be a good save then. I’ll point out that even with distortions, you can still delineate the colors used for ice from open sea to get the total number of pixels representing areas of sea ice @ 15% or greater… the hues differences, even with distortion introduced by JPEG compressions should be easy enough to capture and turn into pixel counts.

    I didn’t run this experiment, Mr. Goddard did, but that is how I think he approached it. Correct me if I’m wrong.

  125. Glenn says:

    Jeff says “Ignoring the data is a tactic that I expect from an AGW denier”

    Just eyeballing these two maps

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=11&fy=2007&sm=08&sd=11&sy=2008

    should convince most anyone without an agenda that the right side is quite a bit more than 10% more than the left side map. The 10% comes from data taken from this graph

    http://nsidc.org/images/arcticseaicenews/20080811_Figure2.png

    It needn’t be rocket science to see the inconsistency, and no need to ignore any data.

  126. Oldjim says:

    Phil. (20:12:24) :
    I guess trig isn’t your strong point! (30º isn’t visible on the CT map either, although it would be if they used the projection you claim.)
    The fact that a projection as claimed would include a lot more than the area shown is irrelevant. It all depends on the area included in the image and there is no point including parts of the globe which are nowhere near the arctic

  127. Oldjim says:

    Ignore the above post _ I was looking at the wrong picture

  128. dash says:

    The NSIDC graph shows current NH sea ice coverage to be around 700,000 sq km or 15% greater than the same time last year, not 10%.

  129. Smokey says:

    Jeff (21:31:35) :

    Smokey:
    Added together, global sea ice is rapidly increasing.

    “Really? You know this from where?”

    As stated, the N.H. maps come from: “University of Illinois, the Cryosphere Today”. The S.H. chart, showing an increase in sea ice of more than one million square kilometers, comes from NSIDC/NASA.

    If you don’t believe your eyes, go argue with them.

    Honestly, some folks seem to be afflicted with what George Orwell would describe as “up is down, black is white, evil is good, and increasing global sea ice is decreasing global sea ice. Get a grip. Or at least get some facts, instead of sniping from the sidelines.

    Sorry, it’s late. But it gets tiresome when the increasingly desperate AGW Bovine Fecal Purveyance Specialists flail around trying to discredit someone like Goddard or Monckton, who do their own investigation, their own research based on credible sources, and write their own articles — instead of writing an article themselves. I suspect the reason the true believers in the disappearing sea ice meme know that such any such article will quickly be torn to shreds, based on empirical evidence. The fact is, the climate always fluctuates. Naturally. The climate today is well within normal historical parameters. If anything, conditions today are quite benign.

    Mr. Goddard has patiently answered each criticism in turn. More bluster won’t change that.

    Why not try writing your own article instead? Ask Anthony if he would post it, and we can see how it stands up. Now that would be very interesting.

  130. Oldjim says:

    Please ignore the above post – I misread the image

  131. Steven Talbot says:

    Glenn,

    You say:

    “The significant consideration is accuracy, my friend.”

    I agree that is significant, it goes without saying, which is why I do not think eyeballing is a good enough basis upon which to draw your conclusion of “the maps being inconsistent with the graph” (my eyeballing is inconsistent with yours, by the way).

    If you feel comfortable will all this inconsistency, I think you are being just as irresponsible as the experts who generate it.

    That’s pretty close to a straw man, Glenn. The inconsistency I feel most concerned about is that between the satellite records, since they are purportedly measuring the same things, AFAIAA, and they also show greater divergence than between other types of record.

    Regardless, it appears that the Arctic melt is largely if not exclusively a result of events not related to “global warming”

    That’s an interesting, and very confident, claim. Would you be able to explain to me how you quantify “largely if not exclusively”? I am well aware that wind patterns will influence the particular extent in a given season, for example, but do you have evidence to establish that the long-term trend is unrelated to warming?

    But making claims about what I “happily” haven’t done or have disregarded, in the absence of evidence of that, isn’t quite honest. It’s a tactic I expect from an AGWer. Are you one, or rather do you accept the “projections” made by the various AGWers that spell doom and gloom if we don’t all stop drivin that hot rod Lincoln?

    I am not a member of any church, Glenn. At present I am persuaded that the risks of AGW are high. I look forward to whatever new evidence, either way, may be forthcoming. I don’t count your eyeballing of graphs and maps to be useful evidence of anything much.

  132. Jeff says:

    Glenn
    “Just eyeballing these two maps

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=11&fy=2007&sm=08&sd=11&sy=2008

    should convince most anyone without an agenda that the right side is quite a bit more than 10% more than the left side map.”

    [snip - Jeff, knock off the insulting ad homs please]

  133. Jeff says:

    Evan Jones
    “What does make a difference is whether or not NSIDC is accurately reporting the data.”

    And I have shown that the NSIDC esimates are reasonable [snip - thats ad hom]

  134. Steven Talbot says:

    The 10% comes from data taken from this graph

    http://nsidc.org/images/arcticseaicenews/20080811_Figure2.png

    Sigh. No, the 10% comes from your (and Goddard’s) eyeballed guesstimate of the data. The figures were published at the same time as that graph, directly alongside it, and they showed 13.5% above the previous year, a figure that is remarkably consistent with Jeff’s 14% from a count of grid cells!

  135. Phil. says:

    dash (03:39:32) :
    The NSIDC graph shows current NH sea ice coverage to be around 700,000 sq km or 15% greater than the same time last year, not 10%.

    Current JAXA data (higher res than NSIDC) shows 13% ((5.839-5.166)/5.166)

    Oldjim (02:52:28) :

    The fact that a projection as claimed would include a lot more than the area shown is irrelevant. It all depends on the area included in the image and there is no point including parts of the globe which are nowhere near the arctic

    So CT took the trouble to show the globe in perspective out to ~32ºN then couldn’t be bothered to complete it and decided to add the star background instead! I repeat that projection isn’t what Goddard thinks it is.

  136. Jeff says:

    I incorrectly stated that I had posted the location of NSIDC’s data. That actually was in another forum. The data can be found at

    ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/nasateam/

  137. Jeff says:

    “Just eyeballing these two maps

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=11&fy=2007&sm=08&sd=11&sy=2008

    should convince most anyone without an agenda that the right side is quite a bit more than 10% more than the left side map. The 10% comes from data taken from this graph”

    Nobody who doesn’t have an agenda would refuse to look at the actual data.

    You don’t seem to be able to comprehend that I actually COUNTED the number of grid cells with sea ice on each day and therefore I actually KNOW what the difference in areas is. My 14 percent figure is not something that “looks right” to me, it is the actual difference. There’s no other way to explain this. I’m trying to make this as simple as possible.

  138. Jeff says:

    Smokey
    “As stated, the N.H. maps come from: “University of Illinois, the Cryosphere Today”. The S.H. chart, showing an increase in sea ice of more than one million square kilometers, comes from NSIDC/NASA.

    If you don’t believe your eyes, go argue with them.”

    So you’re talking about the change from one year to the next? Have you looked at the long-term trend?

    Where I live, we’ve already have 50 percent more rain this month than in all of August 2007. Therefore, the climate here is getting wetter, right? (unless you take into account that we have only received 60 percent of the average rainfall for the first 230 days of the year)

    “Honestly, some folks seem to be afflicted with what George Orwell would describe as “up is down, black is white, evil is good, and increasing global sea ice is decreasing global sea ice. Get a grip. Or at least get some facts, instead of sniping from the sidelines.”

    I have facts, unlike Mr. Goddard who mysteriously comes up with numbers without explaining how he got them.

    “Sorry, it’s late. But it gets tiresome when the increasingly desperate AGW Bovine Fecal Purveyance Specialists flail around trying to discredit someone like Goddard or Monckton, who do their own investigation, their own research based on credible sources, and write their own articles — instead of writing an article themselves. I suspect the reason the true believers in the disappearing sea ice meme know that such any such article will quickly be torn to shreds, based on empirical evidence. The fact is, the climate always fluctuates. Naturally. The climate today is well within normal historical parameters. If anything, conditions today are quite benign.”

    I obviously did my own research, because I obtained the actual sea ice data files and looked at the actual values rather than just saving an image from a web page. There have been numerous publications on global warming. That you aren’t aware of them is pretty startling.

    “Mr. Goddard has patiently answered each criticism in turn. More bluster won’t change that.”

    He hasn’t explained how he got his numbers. He hasn’t explained where he got the 10 percent difference from. He hasn’t explained why he uses JPEGs. He hasn’t explained why he doesn’t use the actual sea ice data files. Maybe he doesn’t even know that JPEGs use lossy compression, so that the color value in any particular pixel does not necessarily represent the original data value. And maybe he doesn’t understand the artifacts that can be introduced into data when it is interpolated multiple times.

    “Why not try writing your own article instead? Ask Anthony if he would post it, and we can see how it stands up.”

    Why would I write an article showing that NSIDC was right when NSIDC was obviously right? You [snip] have already said that you’d rather rely on your own “impression” of what the data is rather than look at the actual numbers, so printing the facts won’t change anyone’s mind.

  139. Steven Talbot says:

    …the disappearing sea ice meme…

    It looks like we’re heading towards the second lowest summer coverage on record, contending with 2005….

    …no problem, then – that must mean the world is cooling. eh?

  140. Steven Goddard says:

    It appears that the Northwest Passage has closed up again after only three days.
    http://nsidc.org/data/seaice_index/images/daily_images/N_daily_extent_hires.png

    Eyewitness accounts here –
    http://awberrimilla.blogspot.com/2008/08/lancaster_18.html
    They got out just in time. Resolute is freezing in behind them

    I’m guessing that NSIDC will not make a high profile press release about this.

  141. iceFree says:

    Someone better warn Gordon Pugh not to launch his kayak.

    Pugh said, “Over recent decades, we have lost more than half the Arctic summer sea ice cover; decades ahead of predictions, showing climate change has been hugely underestimated. We must insist that our leaders take urgent action to halt climate change.”

  142. Steven Talbot says:

    It appears that the Northwest Passage has closed up again after only three days.
    http://nsidc.org/data/seaice_index/images/daily_images/N_daily_extent_hires.
    png

    Lol, now you pick the NISDC map! Here’s another view from UIUC, whose map you seemed to favour before when you wanted to make a different case –

    http://arctic.atmos.uiuc.edu/cryosphere/NEWIMAGES/arctic.seaice.some.000.png

    We can all pick cherries ;-)

    As for a small boat getting stuck in “shortlived, thin new ice”, well, I guess that must prove the world is cooling and that AGW is a scam, eh?

  143. Steven Talbot says:

    By the way, Steven Goddard, have you not figured out where the Berrimilla is (whose blog you link to as an “eyewitness account” of the NW Passage supposedly closing up)? Today it’s been moving through Lancaster Sound, heading into Baffin Bay. Lancaster Sound, from which the report you link to was posted, is here:

    http://en.wikipedia.org/wiki/Lancaster_Sound

    Check that out against the NSIDC map you’ve linked to and you will see that it is recorded as clear blue, and is a very long way from the more marginal areas of the NW Passage!

    Do some basic research, I’d suggest!

  144. Steven Talbot says:

    Further to my last comment in response to S Goddard’s claims of an eyewitness account to the NW passage closing up, this post shows the ice map for the area they were in on the same day that SG quotes from (i.e., yesterday) –

    http://awberrimilla.blogspot.com/2008/08/baffin-bay.html

    The location SG quoted from, Lancaster Sound, is to the top left. The green ‘islands’ are thick first-year ice, the triangles are icebergs, the white is described as clear water. You can see that there is no way the passage is ‘closed up’ at this point, as claimed.

    In a later post from the same day, the blog is celebrating having completed the NW passage:

    http://awberrimilla.blogspot.com/2008/08/pond-inlet.html

    Humph – a bit of basic research there. Though maybe the notion isn’t actually to research the truth of claims that can be made?

  145. Evan Jones says:

    The difference today is interesting. Note that in the 2007 version the NW passage is open.

    It also seems as if the main area of the 2008 image has gained more ice percentage towards the center of the sheet than last week.

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=18&fy=2007&sm=08&sd=18&sy=2008

  146. Patrick Henry says:

    icefree,

    http://nsidc.org/data/seaice_index/images/daily_images/N_daily_extent_hires.png

    Looking at that map, I’m having a difficult time seeing that “half of the ice” is gone, and I’m also having difficulty seeing how whatever is going on up there affects my life. Reading the logs of the Berrimilla, it sounds like summer in the High Arctic is bitter cold. Arctic ice obsession is yet another imagined crisis for people with empty lives.

  147. iceFree says:

    the north west passage has been open many times in history. It’s a non event , ho hum time for bed.

  148. Evan Jones says:

    As for a small boat getting stuck in “shortlived, thin new ice”, well, I guess that must prove the world is cooling and that AGW is a scam, eh?

    Northern sea ice depends on too many variables other than temperatures. Precipitation, dirty snow, Arctic Oscillation weirdness (like last year), etc., etc.

    For the nonce I will settle for the satellite measurements of the lower troposphere. That shows a flat (to slighly cooling) record for the past decade (or 7 years if you prefer to skip the El Nino/La Nina episodes of 1008-2000).

    I would prefer accurate surface measurements but I won’t trust any of them at least until the NOAA/CRN system goes online later this year.

  149. Patrick Henry says:

    Steven Talbot,

    Had you actually read the Bermilla blogs, you could have saved yourself quite a bit of embarrasment. The Northwest Passage is a complete route over the top of Canada which the Bermilla is currently undertaking. They are taking the shortest and fastest route across the NW Passage. What part of They got out just in time. Resolute is freezing in behind them is it that you don’t understand?

    Friday’s new ice from the Canadian Ice Service
    http://awberrimilla.blogspot.com/2008/08/yesterdays-new-ice.html
    The purple is the new ice from the evening 13th and morning 14th, and the first I’ve seen this year.

    Sunday’s new ice from the Canadian Ice Service
    http://awberrimilla.blogspot.com/2008/08/lancaster-map-again.html
    Purple is new ice. ‘E’,’L’ &’G’ also has new ice

    Today’s NSIDC map shows the southern leg of the NW passage now blocked, which they came through a few days ago.
    http://nsidc.org/data/seaice_index/images/daily_images/N_daily_extent_hires.png

    If you have ever watched a body of water freeze, you would know that it starts out as thin ice and later gets thicker. Given that it is mid-August, someone of at least moderate intelligence might be clever enough to realize that the Arctic summer has just ended and the freeze is just beginining. They described the new ice as being The worst kind to a small boat. Tricky concept for someone who likes to shoot their mouth off before they think.

  150. Phil. says:

    Patrick Henry
    “They described the new ice as being The worst kind to a small boat. Tricky concept for someone who likes to shoot their mouth off before they think.”

    And it is, however they appear to have misread the ‘eggs’, there’s no new ice shown on the charts, in fact the ice shown off Resolute (M) is indicated as 2/10 coverage, old ice, predominantly of ‘Land origin’!
    The purple ice is either ‘big ice floes’ or ‘vast ice floes’ of ‘multi year ice’ or ‘thick first year ice’.
    You can find the codes for the eggs at: http://www.natice.noaa.gov/egg_code/index4.htm

  151. wattsupwiththat says:

    Paul K check your email

  152. dash says:

    For anyone still interested in pixel-counting, there’s a very clear-looking series of polar ice maps available from the University of Bremen:

    http://www.iup.uni-bremen.de:8084/amsr/amsre.html

    Image archive pages and folders here:

    http://www.iup.uni-bremen.de:8084/amsredata/asi_daygrid_swath/l1a/n6250/

    It does appear that Steven Goddard’s Register piece was rather hastily and carelessly put together.

  153. Steven Goddard says:

    Anthony –

    Your explanation of methodology is basically accurate. First, the software removes all non-ice pixels from the UIUC image. It then removes any pixels outside the radius of the earth. Then it converts all remaining pixels to a single color. Finally, it counts all the pixels of that color. Using a single color to represent all >15% concentrations is the same technique which NSIDC uses, and makes the generated videos easier to visualize.

    I generated a graph showing the daily percentage gain in ice extent during August vs. last year, and was surprised to see that the trend has a positive slope and is thus increasing. The average over the month is 30% gain.
    http://spreadsheets.google.com/pub?key=pj0h2MODqj3jAfVXaaKHbCQ&oid=5&output=image

    This graph uses only conservative, unadjusted pixel counts – which tend to underestimate the actual gain. I will try to keep it updated through the remainder of the season.

    REPLY: Steve, perhaps to add strength to your argument, you can run the same method on the maps at the URL in the previous comment? Replication is always good.

  154. Steven Goddard says:

    Phil –

    Are you suggesting that the crew of the Bermilla imagined the new ice they were struggling to get through?

    dash –

    Why not count pixels using the same map sets which I did, and then make your judgment? Several people here keep making the same mistake of comparing vs. different data sets. One of the primary points of my article was to show that the NSIDC graph is significantly inconsistent with the UIUC maps. All you are doing is repeating my assertion.

    No one has made any attempt to demonstrate that I measured the UIUC maps incorrectly, or that the UIUC maps are incorrect. Until you do so, your criticisms are illogical and empty.

  155. dash says:

    Steven Goddard: I believe you have counted the pixels correctly, even if you didn’t eyeball the ‘10% difference’ correctly (more like 13.5-14%).

    But according to the NSIDC’s website, sea ice extent is not simply derived from a pixel count, but from the number of pixels multiplied by their ice concentration:

    “The values for ice area are obtained by summing the concentration of ice within each pixel over the entire ice extent. For example, if a pixel’s area was 600 km² and its ice concentration was 75%, then the ice area for that pixel would be 450 km²”

    http://nsidc.org/data/seaice_index/derivation.html

    Have you done this? And may I also draw your attention to this sea ice extent graph from Japanese researchers – it looks very similar to that of the NSIDC, no?

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

  156. Steven Talbot says:

    Patrick Henry,

    I posted a response to you but it appears to have been eaten by the spam filter.

    Please figure out where the boat was, Lincoln Sound, and its direction. It was a great distance away from where you suggest the NSIDC image shows the NW Passage to be blocked, and heading away. Yes, it was experiencing some ice (“shortlived”), as they stated, but no, this was not an eyewitness account of the passage being blocked where you, or Steven Goddard, had suggested.

    Your attempts to sneer at my intelligence are purposeless. Try to address the points rather than resorting to the ad hominem.

    You have posted a ink to NSIDC which you suggest shows the passage blocked. Here is a link to a much more detailed image, which suggests otherwise –

    http://www.iup.uni-bremen.de:8084/amsr/arctic_AMSRE_visual.png

    So what? – either way. It’s of no consequence to me at all whether or not the passage is navigable, it was only of interest to me whether or not the yacht’s report was truly eye-witness evidence of the passage being blocked where claimed.

  157. Steven Goddard says:

    Anthony,

    Good suggestion. The Bremen maps, NSIDC maps and NSIDC graphs are in close agreement. I’m not disagreeing with their internal consistency. The NSIDC/Bremen maps and area calculations use an adjusted projection (an attempt at equal area,) which involves a fair amount of math to generate.

    By contrast, the UIUC maps use a natural, unadjusted projection – essentially what an astronaut flying over the North Pole on a cloudless day would see. The published UIUC maps also have much higher resolution of the Arctic Basin than either NSIDC or Bremen maps.

    I can measure the UIUC maps directly, and make reliable estimates of relative areas from map to map. There is a radial distortion in these maps, but as I have explained before, this distortion causes my estimates of larger ice areas to be conservative – on the low side.

    Because the Bremen/NSIDC maps are intentionally distorted using a more complex algorithm, there is no simple way to determine why they appear to disagree with the UIUC maps. I will assert that if the UIUC maps are correct, and if I counted the pixels properly, there is the possibility of a mathematical error in the NSIDC calculations. There is an inconsistency.

    Either NSIDC is in error, UIUC is in error, or I have measured/interpreted the UIUC maps incorrectly. Like I said, no one has made any attempt to demonstrate any one of those three possibilities. I am fully aware that my numbers are inconsistent with NSIDC. That is exactly the point!

    Do we trust our own eyes? The area growth on August 11vs. last year was clearly greater than the 10/12/14% (whatever) which the August 11 NSIDC graph indicated.

    If someone wants to prove me wrong, they need to demonstrate that either the UIUC maps are incorrect, or that I am not interpreting them correctly. I would greatly appreciate it if people would stop parroting that my numbers are inconsistent with NSIDC. I know that, and that was one of my primary reasons for writing the article.

  158. Phil. says:

    Steven Goddard (07:03:07) :
    Phil –

    Are you suggesting that the crew of the Bermilla imagined the new ice they were struggling to get through?

    No I’m saying you imagined it yourself, they never said they were ‘struggling to get through’ new ice, in fact they never mentioned new ice at all! That was their support guy, Pat in Nome, who as I pointed out apparently misread the egg charts.

    dash –

    Why not count pixels using the same map sets which I did, and then make your judgment? Several people here keep making the same mistake of comparing vs. different data sets. One of the primary points of my article was to show that the NSIDC graph is significantly inconsistent with the UIUC maps. All you are doing is repeating my assertion.

    No one has made any attempt to demonstrate that I measured the UIUC maps incorrectly, or that the UIUC maps are incorrect. Until you do so, your criticisms are illogical and empty.

    Actually they have, you’ve been told you don’t know what projection the map is using and that by measuring directly measuring from the map you’re making an error. Those who’ve measured directly from the data and used the actual area/pixel get results that agree with the published data, JAXA and NSIDC. That you choose to ignore the criticisms doesn’t make them illogical or empty.

  159. Evan Jones says:

    Have you done this?

    I am looking at today’s comparison. Estimating by eyeball, I see a much greater area of high-percentage ice coverage area this year than last year.

  160. Steven Talbot says:

    Phil,

    Thanks for clarifying that the blog entry was not an eye-witness account but rather the backup’s misinterpretation of the ice map.

    Patrick Henry – given that has now been demonstrated, do you have the courtesy to reconsider your remarks towards me made in your post at (21:51:37) ?

    Steven Goddard,

    If you are concerned to understand what you think is a discrepancy (clearly there are some others who do not agree with you), may I suggest that you contact UIUC and/or NSIDC to ask for clarification? Surely that we be good practice in researching the matter? You’re not obliged to agree with anything they might say, but at least you would give them the opportunity to explain whether or not there is, in fact, any such discrepancy.

  161. Patrick Henry says:

    phil,

    You are again confusing “area” and “extent.” Read the paragraph you quoted from NSIDC more closely. “Area” is derived from both “extent” and “concentration.”

    The NSIDC map for today again shows that all routes through the Northwest Passage are blocked. This was not true a few days ago. Obviously this has to be due to new ice, as described in the Bermilla blog.

    They got out just in time. Resolute is freezing in behind them

    What is it they “got out” of? What is freezing “behind them?” The answers seem quite obvious, que no?

  162. Jeff says:

    Evan ones
    “I am looking at today’s comparison. Estimating by eyeball, I see a much greater area of high-percentage ice coverage area this year than last year.”

    I suspect that the “eyeballers” are concentrating on the area off of the eastern half of Siberia where there is ice in 2008 but not 2007 and not noticing the areas where there was ice in 2007 but not 2008. The choice of colors might play a role in this. Many people do not like red because it tends to dominate other colors.

  163. Jeff says:

    Steven Goddard
    “f someone wants to prove me wrong, they need to demonstrate that either the UIUC maps are incorrect, or that I am not interpreting them correctly.”

    You’re the one who claimed that NSIDC was fudging data. It’s up to you to prove that NSIDC’s data are wrong. This is just a standard [snip - stop with the ad-homs, last warning] tactic: burden of proof shifting.

  164. Mark C. Serreze says:

    Mr Goddard:

    Under our “sea ice news and analysis” site, you will see a link to frequently asked questions. It is a useful educational resource and as such I suggest that you read it. Furthermore, you might want to analyze the raw data themselves which are freely available at NSIDC.

    Mark C. Serreze

    NSIDC

    REPLY: Mr. Serreze, thank you. It is nice of you to comment here.

    Would you be willing to answer questions here on this forum? Thank you for your consideration – Anthony Watts

  165. Jeff says:

    Steven Goddard
    “The NSIDC/Bremen maps and area calculations use an adjusted projection (an attempt at equal area,) which involves a fair amount of math to generate.”

    The maps are not an attempt at equal area projections. They are Polar Stereographic projections, which have been used in meteorology for decades.

    “The Bremen maps, NSIDC maps and NSIDC graphs are in close agreement.”

    As are the NCEP maps. In fact, everything is in close agreement EXCEPT your alleged calculations.

    http://polar.ncep.noaa.gov/seaice/hires/nh.xml

    “Because the Bremen/NSIDC maps are intentionally distorted using a more complex algorithm, there is no simple way to determine why they appear to disagree with the UIUC maps.”

    It’s actually easy to do get a qualitative idea of the difference. And what you find is that there isn’t any real difference.

  166. Glenn says:

    Jeff says
    “I suspect that the “eyeballers” are concentrating on the area off of the eastern half of Siberia where there is ice in 2008 but not 2007 and not noticing the areas where there was ice in 2007 but not 2008. ”

    LOL. Where would those be, Jeff?

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=11&fy=2007&sm=08&sd=11&sy=2008

    This isn’t much different or difficult than looking at a piechart (which is also data) and seeing a 30% slice marked as 10%. There is an obvious inconsistency in either the UIUC maps, or the NSIDC chart for Aug 11, or the data.

  167. Steven Talbot says:

    Patrick Henry,

    You appear to be someone who likes to call others stupid but who is not prepared to look at presented evidence yourself –

    The NSIDC map for today again shows that all routes through the Northwest Passage are blocked. This was not true a few days ago. Obviously this has to be due to new ice, as described in the Bermilla blog.

    The NSDIC shows low resolution cells. The far higher-res Bremen image, to which I have already linked you, shows the Amundsen route as open:

    http://www.iup.uni-bremen.de:8084/amsr/arctic_AMSRE_visual.png

    Your comment please?

    “They got out just in time. Resolute is freezing in behind them”

    What is it they “got out” of? What is freezing “behind them?” The answers seem quite obvious, que no?

    This comment is written by the back-up, who is sitting somewhere warm misinterpreting an ice map, as has been carefully explained to you! To use your own words, “someone of at least moderate intelligence might be clever enough to realize” what is being said, but I guess it’s a “tricky concept for someone who likes to shoot their mouth off before they think.”

  168. Jeff says:

    “By contrast, the UIUC maps use a natural, unadjusted projection”

    Please explain this. If that projection was “natural” and “unadjusted”, why isn’t it widely used? Since it’s possible to compute the actual Earth area of the ice from NSIDC’s data, why isn’t is possible to compute the Earth area of the ice from UIUC’s maps?

  169. Steven Talbot says:

    Incidentally, Patrick Henry, to help you commenting on the difference between the Bremen image and the NSIDC image, see this from the BSIDC FAQ:

    Why do other sources indicate there is no ice where passive-microwave data shows ice?

    The passive-microwave sensor records ice in 25-by-25-kilometer (16-by-16-mile) areas, which is lower resolution than other types of satellite sensors. This means that the ice edge could be off by as much as 25 to 50 kilometers (16 to 31 miles) in passive-microwave data compared to higher-resolution satellite systems. We define ice extent as anywhere with at least 15% ice. Regions with 15% ice will look quite open in higher-resolution satellite data, but will still count as ice in the passive-microwave extent fields.

    Other reasons that passive-microwave data may show ice where none actually exists on the ground include signal variation along coastlines between land and water, the shift in albedo of actively melting ice, and atmospheric interference from rain or high winds over the ice-free ocean. In the daily extent data images, gaps (shown in dark gray in the extent map) are replaced with values interpolated from surrounding days, but temporary spurious results may occur.

    http://nsidc.org/arcticseaicenews/faq.html#data_sea_ice_index

    Your comments?

    I recommend looking at the NSIDC FAQ, as Mark Serreze has suggested. Doing some basic research is usually a more profitable process than leaping to a conclusion that suits you.

  170. Jeff says:

    Glenn
    “Where would those be, Jeff?”

    ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/nasateam/near-real-time/browse/north

    “This isn’t much different or difficult than looking at a piechart (which is also data) and seeing a 30% slice marked as 10%. There is an obvious inconsistency in either the UIUC maps, or the NSIDC chart for Aug 11, or the data.”

    Wrong.

    If you read my 12:27:53 post, you will see what I believe is the same data that UIUC uses mapped to the same projection that NSIDC uses. How does it compare?

  171. Phil. says:

    Patrick Henry (11:16:20) :
    phil,

    You are again confusing “area” and “extent.” Read the paragraph you quoted from NSIDC more closely. “Area” is derived from both “extent” and “concentration.”

    I’m not confusing area and extent not now nor earlier, it’s more accurate to say that extent and area are both derived from the same source, total area being the sum of the product of area and concentration whereas extent is the sum of area alone (for concentration greater than 15%).

    The NSIDC map for today again shows that all routes through the Northwest Passage are blocked. This was not true a few days ago. Obviously this has to be due to new ice, as described in the Bermilla blog.

    The NSIDC is rather low resolution you’be better off looking at the uni bergen map instead:
    http://iup.physik.uni-bremen.de:8084/amsr/arctic_AMSRE_nic.png
    It is not obvious that it is new ice, in fact it is old ice emerging from the strait to the north. http://ice-glaces.ec.gc.ca/prods/WIS35CT/20080819180000_WIS35CT_0003924087.gif
    That map shows today’s detail of the area around Resolute, the region marked W is the ice referred to in this comment by the Berrimilla blogger based in Nome who misread this chart.

    “They got out just in time. Resolute is freezing in behind them”

    What is it they “got out” of? What is freezing “behind them?” The answers seem quite obvious, que no?

    Yes, nothing’s freezing behind them, as I told you above he misread the chart! The ice at W is 2/10 coverage made up of 50/50 old ice and thick first year ice predominantly of land origin (icebergs).

    The color code used is intended to assist navigation decisions in ice infested water. It is loosely based on the concept of a traffic light, where green means proceed, yellow means caution and red signals danger. W is shown as green as I’m sure you can see! The presence of old ice (multi-year ice) is indicated by the colour purple (remember all that ‘new’ purple ice mentioned in the Berrimilla blog the other day?)

    The NW Passage opened for navigation on Aug 21st 2007, looks like it will be close to that this year.

  172. Steven Talbot says:

    Glenn,

    “Where would those be, Jeff?”

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=11&fy=2007&sm=08&sd=11&sy=2008

    2007 extends further into the Beaufort Sea and towards the northern tip of Novaya Zemlya (my eyeballing’s as valid as yours! ;-)).

  173. Phil. says:

    The passive-microwave sensor records ice in 25-by-25-kilometer (16-by-16-mile) areas, which is lower resolution than other types of satellite sensors.

    This refers to the SSMI imager used by NSIDC and CT (at least in the old images used by Goddard), JAXA and Uni-Bremen and CT?(new) use AMSR-E which has 6×4 km resolution.

  174. Walt Meier says:

    I see that Mr. Goddard’s article has gotten quite a discussion going. I’ve emailed The editors at The Register, but haven’t heard back, and just recently Mr. Goddard directly.

    Mr. Goddard’s approach to counting pixels is simply not the correct approach. First, let me clarify a couple things.

    1. The satellite data doesn’t directly measure sea ice area or extent. It measures brightness temperature – a measure of the amount of energy emitted by the ice. This is converted to area or extent using an algorithm.

    2. There are several different algorithms and they can yield different results in terms of absolute numbers. However, their trends and change from year to year show similar magnitudes. The Bremen AMSR data is from a different algorithm – hence it looks different.

    3. Both UIUC and NSIDC use the same brightness temperature data.

    4. Both UIUC and NSIDC use the same algorithm, but with some differences in the specifics, so the numbers aren’t perfectly matched, but there is very good overall agreement and they yield the same conclusions about changes in Arctic sea ice.

    5. People have talked a lot about “pixels”, but one needs to understand what one is talking about. There are two types of “pixels”. One is “data pixels”; this is a function of the spatial resolution of the sensor (i.e., how small of an area the sensor can resolve). For the data UIUC and NSIDC uses, the data pixels are about 25 x 25 km. The other is “image pixels”, which describes the qualities of the image.

    6. The data has to be gridded onto a projection, which yields a gridded resolution, which is also about 25 x 25 km, but varies depending on the type of projection and where the grid cell within the projection. The input data for both UIUC and NSIDC is on a 25 x 25 km grid. The UIUC grid that Mr. Goddard analyzes has been interpolated onto a different grid. I do not know the specifics of that grid, but such interpolation will change how the data looks when viewed.

    7. The data can then be conveyed in an image. The image has an “image pixel” resolution. This is generally given in dpi or dots per inch. Higher dpi means a sharper image. However it does NOT change the fundamental resolution of the data.

    8. An image is simply a way to convey data; it is not data itself. Therefor it is not proper to do analysis on the image. You need to use the data.

    9. The gridded data, when analyzed, must account for the projection in terms of the area of the grid cells. You have to sum the ice, weighted by the correct area for each grid cell. NSIDC uses a polar stereographic projection with a true latitude of 70 N. Other than at 70 N there will be distortion that needs to be corrected for, as NSIDC does.

    10. NSIDC freely distributes all the data, tools to work with the data, and the grid cell area files. So anyone can do their own analysis.

    11. NSIDC’s methods have been around for over 20 years, have been thoroughly vetted in peer-reviewed science journals, and confirmed numerous times over by independent scientists conducting the proper method.

    12. Finally, Mr. Goddard need not have wasted his time doing his image pixel counting. He could’ve simply referred to the UIUC site, which actually counts the pixels properly and creates a timeseries plot:

    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.jpg

    If you look at that plot, you’ll that because it is area instead of extent, the raw numbers are lower. However, while it’s a bit hard to make out the values real accurately, you see that this year on Aug. 11, it was ~4 million sq km, while last year on Aug. 12, it was ~3.6 million sq km (actually, since it’s a one-year sliding window, Aug. 12, 2007 is no longer visible, but that’s what it was on that data and the current range shows a similar difference). That’s a bit more than an 11% difference.

    Hopefully Mr. Goddard will have a correction posted prominently on The Register as soon as possible.

    Walt Meier
    Research Scientist
    National Snow and Ice Data Center
    University of Colorado at Boulder

  175. wattsupwiththat says:

    Thank you Dr. Walt Meier, for your input. I’ve emailed Mr. Goddard directly to make sure he’s seen your response.

    Would you be willing to answer questions about the NSIDC data and the presentations of it on this forum?

    Thank you for your consideration. – Anthony Watts

  176. Steven Goddard says:

    Walt,

    Thanks very much for your response. I just received and responded to your E-mail as well. I do understand that you have worked diligently on your methodology and that it has been throughly tested.

    I understand the NSIDC definition of “extent” as being the area of ocean with >= 15% sea ice concentration. As I explained in the E-mail, when I measure that area in the UIUC maps (marked as concentration >=15%,) I see a 30% increase over last year.

    Please advise why the UIUC concentration maps show a significantly larger increase in the amount of >=15% concentration ice, than the NSIDC graph shows.

    If you or Mark can answer that question directly, technically and convincingly – I will be absolutely delighted to publicly retract and apologize.
    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=12&fy=2007&sm=08&sd=11&sy=2008

    The UIUC maps show a clear 30% increase in extent.

    Regardless of any real or perceived inadequacies with the methodology, there should be closer agreement between the UIUC maps and the NSIDC graph. Even with any rounding errors due to pixels, or distortions in the UIUC projection, the difference is much too large. Indeed, the UIUC projection should tend to underestimate larger area relative to smaller ones, because lower latitude pixels get underrepresented on that map projection.

    Basically, I am asking you to account mathematically why counting pixels on UIUC maps shows a 30% increase, and explain what is wrong with that methodology.

    Also, do you still believe that these estimates will be realized?
    http://nsidc.org/images/arcticseaicenews/200805_Figure4.png

    And please comment on this widely published story.
    “Mark Serreze, a scientist with the National Snow and Ice Data Center in Boulder, Colorado, told Agence France-Presse in an interview on June 27, 2008, that by the end of the 2008 summer, there might, for the first time in human history, briefly be no ice at the North Pole.”
    http://en.wikipedia.org/wiki/North_Pole

    This appears to be at odds with an earlier interview from several years ago.
    Dr. Mark Serreze – There’s been open water at the pole before.
    http://query.nytimes.com/gst/fullpage.html?res=9F00E3DD1E31F93AA1575BC0A9669C8B63

    Once again, thank you for getting involved in the discussion.

  177. Steven Goddard says:

    Dr. Meier,

    To speed the discussion up, a bit more on the type of explanation I am looking for wrt the ice extent dispute. Something like this –

    1. “Pixel counting UIUC maps doesn’t work, because the map is distorted in a way which makes the area around 80N latitude too large. The distortion can be defined mathematically as follows ………”

    or

    2. “Pixel counting UIUC maps doesn’t work because the resolution per pixel is too low. The grid cell size is smaller than one pixel, so a region of four contiguous pixels may appear to represent more extent than what is measured on the higher resolution grid. This can be defined mathematically as follows……….. The reason we believe our grid size to be the correct one, rather than the one pixel grid you are using in the article, is because …….”

    Thanks again. It is very late and I need some sleep. I look forward to seeing your response.

  178. Jeff says:

    Steven Goddard
    “Please advise why the UIUC concentration maps show a significantly larger increase in the amount of >=15% concentration ice, than the NSIDC graph shows.”

    If the UIUC maps show a larger increase, where is the difference? Where do the NSIDC maps not show ice and the UIUC maps do show ice?

  179. Glenn says:

    Dr. Meier, this is apples and oranges.

    The chart you refer to appears to be sea ice concentration. It does not depict ocean area with sea ice as does the NSIDC chart and the UIUC maps under consideration, and the numbers are off by about two million sqKm.

    The UIUC maps appear to show a recognizable and significant difference in percentage between the same time last year and now that does not correlate at all well with the NSIDC graph. That is my concern, and what I understand Mr. Goddard’s concern to be as well. Simply put (hint) the NSIDC chart tells me there is about 10% more ice than last year, and the UIUC maps tell me that there is about 30% more ice than last year.

  180. Smokey says:

    Thank you, Dr Meier, for your input.

    However, Mr Goddard has posed some relevant questions.

    As they say, the ball is now in your court.

  181. dipole says:

    Earlier I complained I was having some trouble matching the UIUC map colours to the included scale. Closer examination showed that the 100% ice cover on the map isn’t well represented on the scale. But using the sum of absolute differences of RGB values as a metric and a threshhold of 50 I can get what looks like a reasonable match with the ice extent on the UIUC maps. I would have to play around a bit more before being totally happy with this.

    But with this threshhold I get a 30.5% increase in pixel count for 20080811 over 20070812, in support of Steven Goddard’s original claim.

  182. Jeff says:

    Somebody asked why I limited my ice area calculations to north of 63 degrees. The answer is that it was a lazy way of filtering the (probably) spurious concentrations near the coasts of Hudson Bay and the Great Lakes. I’ve actually looked at it both ways, and it only made 2 tenths of a percent or so difference, which isn’t significant when we’re arguing whether the true difference is 14 percent or 30 percent. Especially since it was a 0.2 percent increase.

  183. Paul K says:

    No, Smokey, the ball is not in the NSIDC court. Rather, the match is over. Steven Goddard refuses to look at the data, and those scientists who have looked at the data, have determined that Goddard’s measurement of 30% increase in sea ice extent, from August 11, 2007 to August 11, 2008, is incorrect.

    Dr. Meiers made the point quite clear:
    “8. An image is simply a way to convey data; it is not data itself. Therefor it is not proper to do analysis on the image. You need to use the data.

    9. The gridded data, when analyzed, must account for the projection in terms of the area of the grid cells. You have to sum the ice, weighted by the correct area for each grid cell. NSIDC uses a polar stereographic projection with a true latitude of 70 N. Other than at 70 N there will be distortion that needs to be corrected for, as NSIDC does.

    10. NSIDC freely distributes all the data, tools to work with the data, and the grid cell area files. So anyone can do their own analysis.”

    People on this thread did the analysis and confirm Mr. Goddard’s conclusion is incorrect…

    Why doesn’t Mr. Goddard use the actual data? What does he have against the actual reported data?

    Yesterday, I already suggested to Mr. Anthony Watts that he needs to publish a correction, but he embargoed my comment. I will attempt to make the comment again.

    REPLY: PaulK Your “request for correction” was one that you wrote with specifics about language and placement. That’s not how it works. It’s not your blog, you don’t have the priviledge to dictate terms, and you don’t get to tell me what to write. The story is written by owned by Mr. Goddard, and since then he has offered to print a correction if his questions are satisfied, I’m going to respect his right to choose what he wants to do and when.

    Right now we have some interesting questions in play, and I want to see how they are answered.

  184. Jeff says:

    Steven Goddard has created a graph of his difference calculations. The values appear to (approximately)

    1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
    24 23 28 41 19 23 20 27 40 31 32 28 32 45 42 41 26 30 21

    Can I be an optimist and hope that the latest figure of 21 percent represents a convergence towards my calculations?

  185. Jeff says:

    Glenn
    “The UIUC maps appear to show a recognizable and significant difference in percentage between the same time last year and now that does not correlate at all well with the NSIDC graph.”

    The alleged difference in the UIUC maps doesn’t correlate well with the UIUC graph either, as Walt Meier pointed out. We again get back to the issue that everything is in agreement except for Mr. Goddard’s numbers.

  186. Patrick Henry says:

    Images certainly are data. They contain location, intensity and color information.

    Why do we send satellites into space to take images of the earth? Isn’t ice data derived from “brightness” images? And at the other end of the spectrum, our understanding of microscopic objects is almost exclusively derived from images.

    Image processing is a huge field, and the base technique of almost any image processing algorithm includes counting pixels of different colors and intensities.

    The assertion that “images aren’t data” is simply not accurate. Many fields of science and engineering rely very heavily on digitized image data.

  187. Paul K says:

    Patrick Henry… In some cases, you are correct, but not in this case. The image Goddard is using, was generated from a database, not directly recorded.

    The key mistake Goddard and dipole are making, if I understand Steven Talbot and Jeff correctly, is that there is no way to show the surface of the globe on a flat screen as a continuous image. So the image people have been eyeballing, MUST have some errors in it. So where are the errors?

    Here I will attempt an explanation… (and please correct me if I get this wrong, Steve or Jeff). It is very difficult to illustrate a curved surface on a flat piece of paper, or a flat computer screen. So “projections” are used for maps, drawings or illustrations depicting the surface of the earth. Steven Goddard assumed that the projection used on Cryosphere showed the area close to the North Pole as accurate to scale, and then that this projection would not represent areas correctly at lower latitudes.

    But in reality the scientists at NSIDC were one or two steps ahead of Mr. Goddard. They used a projection accurate at 70 deg N, not at the pole (90N), as Mr. Goddard assumed. The scientists used this projection, because historically, the latitudes around 70 deg N, is roughly mid-way between summer mins and winter maxs in ice extent. (No one expected the massive ice melts nearer the Pole , that we saw last year, or this year, for another 50-100 years.) The scientists picked a projection that worked best for measuring sea ice extent, in a non-warming world. Where is 70N? It is about halfway down the West Coast of Greenland.
    https://www.cia.gov/library/publications/the-world-factbook/geos/gl.html

    If Cryosphere (UIUC) uses the same projection in the image Mr. Goddard is working with, that the NSIDC uses (and they should), then the image that you and I and Mr. Goddard are looking at, is not accurate at latitudes higher than 70N, and the higher the latitude, the bigger the inaccuracy. This isn’t because someone made an error; this is because you can’t show a three dimensional object in two dimensions without some error.

    Imagine you are looking at the North Pole, looking down at a globe, and you are trying to draw that image on a flat sheet of paper (making a projection). If the area shown on the flat image is accurate down around the mid-west coast of Greenland, then a square inch of area on the globe at that latitude will occupy a square inch on the flat image. But the globe has more surface area (square inches) than the flat image. So the area, and the number of square inches above the 70N latitude, on the flat image, will be LESS than the actual area on the globe. So the flat image you and I see, shows less area in the Arctic Basin, than the actual area on the globe. If that area is covered by ice, and included in the ice extent, then Goddard ended up measuring fewer pixels in the Arctic Basin, than he should, because he is using a flat image to measure his data. And the guys eyeballing it, are making the same mistake. Since the YOY differences in sea extent are at lower latitudes, then the difference in areas there seem bigger, and then this difference is divided by the total extent of the sea ice to get the percentage change.

    Lets try it with some approximate numbers: By getting a lower count of pixels at the high latitude, Mr. Goddard is dividing the difference of around 750k sq KM by less than the actual 5500k sq KM (which would calculate to 13.6% increase), he instead divides by a lower number of pixels equivalent to 2500 sq KM and gets a 30% increase. But his analysis is wrong, as are the eyeball estimates. The actual NSIDC data show that the YOY extent was about 13.5% different on August 11.

    OK, that is how I understand the mistake. The key issue here, is that taking measurements from a flat image representation of our planet, will always end up in error, UNLESS you adjust for the projection error. Mr. Goddard should have known that, and he should have heeded Steven Talbot and Jeff, when they pointed out the projection error.

    REPLY: That is a good explanation, and I appreciate the detail. Another question that needs to be answered is how the satellite derived data is “photographed”. As I understand it, it isn’t a single snapshot from above the pole, but a series of successive orbit slices that are patched together. Thus, how do we get from a curved rectangular view “slice” from the satellite persepctive to the raw data and to the imagery? What sort of projections errors or uninetentional distortions are introduced and are they relevant?

    While there may be errors in the analysis presented here, how would the public know which presentation from which group better reflects the reality of sea ice? Problems like this one we are discussing arise when data presentations for public cconsumption, perhaps uninetntionally, show differences that can’t be easily resolved by the layman.

  188. dipole says:

    Assuming the description of the projection used in the UIUC images as the view from 10,000 miles above the pole, I have corrected my pixel count (21:39:11) to take account of the fact that pixels further from the pole represent a slightly larger area. This increases the extent figure by about 3%, and the increase of 2008 over 2007 goes up slightly to 31.1%.

    Hope I got my trigonometry right.

  189. dipole says:

    Paul K (00:05:31)

    The key mistake Goddard and dipole are making…

    Originally I was just counting pixels. If I claimed further significance for the results it was unintentional. Where’s the ‘mistake’?

    You are claiming that the interpretation of pixel count as ice extent is invalid because of spacial distortion. Steven Goddard described the projection used on the UIUC maps, and in my last post I incorporated this into my calculation. It didn’t seem to materially affect the outcome.

    If you, or anyone else, can demonstrate I’m using the wrong projection I can have another go.

  190. dipole says:

    Hmm. I think my projection formula needs fixing. I am correcting for tilt but not change in distance from the observer. I will try again. Of course if this is the wrong projection I am wasting my time.

  191. AndyW says:

    Has UIUC been emailed to see what they say on the topic? I get the feeling NSIDC are a bit unhappy with this story being broke before the author got in touch with them first so may not be that forthcoming? Or am I reading it wrong?

    It does beg the question though that if the data is right, and having read all this I still think it is, then why are the “popular” graphs at Cryosphere seemingly so far out of kilter? A nice visiual should enable you to compare one year to the next without having to count pixels or apply algorithms or actually just junk it and go back to the data, because then it is useless.

    Maybe these are questions more for UIUC than NSIDC, so I would like to know if anyone has had any input from them?

    Talking of images, the direct NW passage is either nearly open or pretty well closed depending on what you look at.

    http://iup.physik.uni-bremen.de:8084/amsr/arctic_AMSRE_nic.png
    http://www.seaice.dk/iwicos/latest/amsr.n.ice.20080819.gif
    http://ice-glaces.ec.gc.ca/Ice_Can/CMMBCTCA.gif
    http://manati.orbit.nesdis.noaa.gov/ice_image21/D08232.NHEIMSK.GIF

    There’s not too much consistency there to me to be honest. Are all these just pretty to look at in the main?

  192. Smokey says:

    Paul K:

    No, Smokey, the ball is not in the NSIDC court. Rather, the match is over.

    Is that the same as, “The science is settled”? Or maybe, “Consensus”?

    Let’s cut to the chase here: explain this.

  193. I think the distance might be 10000km; not 10000 miles. I did an estimate based on the latitudes of regions I can make out just on the perimeter. 10000km would mean the perimeter is roughly latitude 23.

  194. Simmon says:

    “Another question that needs to be answered is how the satellite derived data is “photographed”.”

    That information is publicly available for the NSIDC data from the NSIDC web site:
    http://nsidc.org/data/docs/daac/nsidc0051_gsfc_seaice.gd.html

    and info for the original data used by Cryosphere Today (Cryosphere Today does not clearly acknowledge their data source) is available from the NCEP web site:
    http://polar.ncep.noaa.gov/seaice/support/ssmi.about.html

  195. Simmon says:

    Patrick Henry (22:54:25) :
    Images certainly are data. They contain location, intensity and color information.

    Images are data, but they’re not necessarily good data, and they do not contain location and intensity information. They contain an x,y location in image space and values for red, green, and blue. If you know the projection (and we don’t know the projection of the Cryosphere Today images) you can calculate location, but it’s not intrinsically in the image. Similarly, the RGB values don’t translate directly to intensity: they merely represent the relative strengths of three different sub-pixels on a computer monitor. Additionally, the Cryosphere Today images use lossy compression (even the PNGs contain compression artifacts) so they can’t be used for precise quantitative analysis. (Contrary to an assertion earlier in this thread, JPEG noise can not be removed.)

    REPLY: I’ve been working with computer graphics since 1983, and my company produces computer graphics as seen on the side panel to the right. Thus I’m qualified to speak to the concerns of image formats and compression since I have used them both since their inception and produce them daily.

    PNG is a 32 lossless compression image format, there are no artifacts, just like with the 8 bit GIF compression, there are no artifacts. The run length encoding schemes simply packetize pixels that are the same into tokens. It is totally lossless and without any compression or noise artifacts to convert the packets of RLE data to pixels upon display.

    http://en.wikipedia.org/wiki/Portable_Network_Graphics

    So there is no error of any kind in reading PNG pixels and translating it’s RGB component back to the numeric value of the original data as defined by the color key.

    Unlike PNG, JPEG is not perfect, and much depends on the original compression setting, known as the Q factor. See the sample photographs in this article to see what I mean: http://en.wikipedia.org/wiki/JPEG

    JPEG’s work because humans can resolve detail better than they can resolve color. NTSC and PAL color television use the same trick. They preserve the detail (luminance) while mostly compressing the color (chrominance). The weighting equation assigns the most compression to reds and blues, while preserving green, which we resolve better.

    As for removing JPEG noise from it’s lossy scheme. I’ve done it and have a program that does that that is commercially available. Granted, depending on the compression factors applied to luminance and chrominance originally there may not be a 100% perfect restoration, but noise can be removed. Most of the error is usually in color space. JPEG does not appreciably alter the shape of the photo or graphic, thus in this exercise, while there may be a margin of error for recovery, Nonetheless for a comparison of two images to determine differences in pixels, it can be done with a small margin of error. I estimate about 2% or less for these images. To prove that to yourself, create a couple of images that have varying graphical color elements of known pixel area, then save as PNG and as JPEG, then run a pixel count on each. You may find that while the colors count varies in JPEG, the luminance count varies little, and perhaps not at all depending on the compression setting used.

    But the Cryosphere today images, such as these here: http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=15&fy=2007&sm=08&sd=15&sy=2008

    are PNG and totally lossless. So I would expect accurate pixel counts between them.

    That doesn’t address the larger questions about earth’s spherical geometry and what distortions may exist, but the approach of pixel counting/comparing two images can be expected to give repeatable results with a margin of error for JPEG and exact results for PNG – Anthony

  196. Steven Talbot says:

    Mr Goddard,

    I don’t know whether or not Dr Meier will respond further to your questions, but I shall make my own observations as to whether or not you have anything to correct in your original article. Quotations are, of course, from your original.

    1. “Yet there is, however, something odd going on with the ice data.”

    You have no evidence that there is anything odd going on with the data. Your remaining case is that the visualisation of that data suggests the sea ice extent is greater than is the case. As has been made clear, NSIDC data is freely available to you. You have made no reference to it. IMV you should withdraw your implication and clarify that your concern is with visualisation exaggerating the differences in ice extent.

    2. “Their data shows Arctic sea ice extent only 10 per cent greater than this date in 2007″

    This figure has been corrected, and you should correct the statement.

    3. “The problem is that this graph does not appear to be correct.”

    You make the presumption that the graph is not a correct representation (and imply, IMV, that NSIDC data is thus not correct). Why did you not or do you not consider the possibility that the imaging is “not a correct representation”? It has been made clear to you that NSIDC data is accessible for examination. Your prejudiced presumption should be corrected.

    4. “Other data sources show Arctic ice having made a nice recovery this summer. “

    As Dr Meier has made clear (and others before him), “An image is simply a way to convey data; it is not data itself.” Your description of “other data sources” is highly misleading, suggesting that the issue is own of conflicting data rather than (apparently) divergent impressions of how that data is conveyed. This should be clarified, if you wish not to mislead your readers.

    5. “Additionally, some current graphs and press releases from NSIDC seem less than conservative. There appears to be a consistent pattern of overstatement related to Arctic ice loss.

    This is a serious allegation, and one that is entirely unsupported by your ‘analysis’. On the contrary, your remaining issue with the image rendition of the data amounts to a potential concern that the image is overstating the data in its visualised indication of sea ice extent. Your allegation is thus unfounded and potentially defamatory.

    Whether or not you gain an explanation that satisfies you of the visualisation apparently exaggerating differences in ice extent, the above points should be corrected if you are to show responsible concern for not misleading your readers.

  197. Steven Goddard says:

    PaulK,

    I have described in some detail the distortion in the UIUC projection, but obviously not enough. In that projection, there is nothing magical about 70 degrees or any other latitude – other than 90 degrees. The oft quoted 70 degree number relates to the projection being used by NSIDC, not UIUC.

    Starting with the ideal case of a viewpoint at infinity (no parallax) – at the pole, the image is undistorted. As you move away from the pole, the distortion increases. At the equator, the distortion would be infinite. In this case, the distortion would be = 1 / sine(latitude)

    Looking at the UIUC maps, the actual viewpoint is from a height less than infinity, so the latitude where the distortion becomes infinite is at a higher latitude than 0 – maybe ten or twenty degrees. That isn’t hugely important because there is no ice anywhere near that latitude.

    In that type of projection, a smaller circle centered around the pole (i.e 2007) is less distorted than a larger circle (i.e. 2008.) If the calculations were corrected to adjust for this, the larger circle (2008) would be adjusted upwards more than the smaller circle (2007.) This would have the effect of making the percentage gain in 2008 larger than what I reported. I did not make any adjustments to the raw pixel count, and that is why I described the methodology as conservative. The actual 2008 percentage gain is somewhat larger than I reported.

    If there is a flaw in the methodology, it is not at that point. It would have to be tied to an inaccuracy in the UIUC map or my understanding of that map. This may well be the case, and I am keen to understand where the error is.

    If Walt or Mark can explain to me where the error is, I will most certainly retract and apologize. There appears to be a discrepancy between the UIUC maps and the NSIDC calculations. I’m certain that the NSIDC calculations were done carefully and diligently.

  198. wattsupwiththat says:

    UPDATE:

    Cryosphere Today has changed their color scheme presentation very recently.

    See the new color scheme here:

    http://arctic.atmos.uiuc.edu/cryosphere/

    It appears the archived images are unchanged and remain the same color scheme. I can’t find any announcement for why this has been done, if somebody knows of one, pass it along please.

  199. Steven Goddard says:

    Anthony,

    Thanks for your follow up on the image color issue.

    In the case of “ice extent” calculations, the image color precision is not important. All pixels >= 15% are treated equally in extent calculations.

    If the calculation was for “ice area” using the images would likely introduce more error, and jpeg would probably not be the best format. I haven’t and wouldn’t attempt to do a calculation of sea ice area, using the images.

  200. Steven Talbot says:

    If there is a flaw in the methodology, it is not at that point. It would have to be tied to an inaccuracy in the UIUC map or my understanding of that map.

    In which case why are you expecting Drs Serreze or Meier, of the NSIDC, to deal with this for you? Why are you not in contact with UIUC (why, for that matter, were you not in contact with both as an obvious aspect of basic research before writing your article?). If you have determined that any inaccuracy is down to the UIUC image, why are you not correcting your article, which explicitly states that the NSIDC graph “does not appear to be correct”?

    I’m certain that the NSIDC calculations were done carefully and diligently.

    In which case, why have you not withdrawn your allegation of “a consistent pattern of overstatement related to Arctic ice loss”, referencing their recent graphs as well as their press statements?

    You seem to be suggesting that you will retract and apologise if the NSIDC can explain issues you have with a UIUC visualisation, but not that you will retract and apologise for the statements I have quoted, even though you now assert your certainty that the NSIDC has been diligent and that any inaccuracy is not to do with them!

  201. dipole says:

    I had another look at the distortion factor as noted at (01:28:52). Now I get +5% for 20080811 and +4% for 20070812 for a 2007 to 2008 increase of 31.6%.

    Still for view height of 10000 miles.

    I would still not be betting on the accuracy of my calculation, but for these images and projection it seems no big deal anyway.

  202. Steven Goddard says:

    Anthony,

    I really like the option of seeing the higher resolution color maps, and wish they had those archived from last year. The increased detail would be useful in comparing vs. satellite images.

    BTW – I need to throw in here that all my comments on this forum are my opinion only, not those of The Register.

  203. Mark C. Serreze says:

    Looking back at earlier posts, a few things caught my eye which I might be able to clarify:

    1) The north pole issue: Back in June, there was some coverage about the possibility of the North Pole being ice free by the end of this summer. This was based on recognition that the area around the north pole was covered by firstyear ice that tends to be rather thin. Thin ice is the most vulnerable to melting our in summer. I gave it a 50/50 chance. Looks like I’ll lose my own bet and Santa Claus will be safe for another year. There was indeed some coverage a some years back of an open north pole (and I was interviewed). This opening, however, was pretty clearly causes by unusual winds breaking up this ice, and not from melting out.

    2) The uptiicks/downticks in our updated time series of ice extent. The time series, as presented on our “Sea Ice News and Analysis Site” is based on 5-day averages. This is done to smooth out short-term “blips” that can occur from noise in the data (basically weather effects and surface melt effect that contaminate the passive microwave retreivals). Speciicially, for a given day N,
    the 5-day average is: (N-2 + N-1 + N + N+1 + N+2)/5 . You run into a problem for the current day (e.g., August 20 as I write this) as we don’t yet have data for N+1 and N+2. Similarly, for August 19 (yesterday), we don’t yet have data for day N+2. So, for the current day and the day before, we do a modified 5-day mean by projecting values forward by 1 or 2 day based on the slope over the past few days. This 5 averaging procedure is why the slope for the last few days or our graph shows these somewhat puzzling upticks/downticks. It turns out we actually changed procedures slightly right around August 1. Before then, the five day average for the current day was (N-2 + N-1 + N + N + N)/5, which really gave too much weight to the current day. Nothing nefarious about making the change – it was a decision based on simply trying to improve presentation of the results.

    3) Antarctic Sea Ice: There seems to be a great deal of misconception about why sea ice in Antarctica is not declining (and indeed does show an increase) and why Antarctica (except for the peninsula) is not strongly warming like the Arctic. One has to realize that a much slower and subdued response of the Antarctic to greeenhouse gas loading is something that was projected by even our earlier climate models over 25 years ago. There’s a number of reasons for this involving the much different nature of the ocean and atmospheric circulation, but the fact that it is the Arctic which is changing rapidly and not the Antarctic is no surprise. Indeed it’s exactly what we expected. The recent sea ice ingrease in Antarctica turns out be be a weak trend in a noisy time series, largely drivent by changes in atmospheric circulation. Please go to the “frequently asked questions” on our “sea ice analysis and news” site where these issue ae discussed.

    4) I see no point in further discussion about malfeasance on the part on NSIDC. We stand by our data and results. My colleague Walt Meier stated the issues quite clearly. Some do not like the story that our data tell. I don’t like the story either, but I’ll continue to tell it as I see it.

  204. Phil. says:

    There are many projections used to represent the earth’s surface on a flat surface and they all make compromises of some sort. The most common projection used for world map sis the Mercator which was designed for navigation and has the virtue that lines of constant bearing are straight lines which is great for navigation but distorts area as you go away from the equator (Greenland looks much larger than it really is). There are several projections used for views from the pole of which the polar stereographic is commonly used (e.g. by Uni Bremen and NSIDC), the projection onto a secant plane at 70ºN being chosen in this case to reduce the distortion of the Arctic.
    The CT images apparently use a different projection, Goddard claims that it’s the view that an astronaut would have from 10,000 miles above the pole (a perspective projection), simple geometry tells us that it can’t be that, if indeed it is that type of projection it’s more like 3,500 miles.
    Whichever projection is used it means a pixel on the map represents a different area depending on position and this must be allowed for in calculating the area from the map. In the absence of the actual data this is all you could do, however when the data is available it’s simpler and more accurate to use it (as several have shown here). I first pointed this out to Goddard early in this thread (on the 15th) and he repeatedly persists that it has no effect!

  205. Pamela Gray says:

    Pioneer stock like me eyeball weather. If we don’t, our crops don’t get put up and either freeze, rot or dry up in the field. So eyeballing the two Arctic images of last year and this year, this same date, we have lots more ice up there. Nuf said. Cold winter coming and soon. Don’t bother with a 3rd cutting. Irrigate like hell so what is left can be used to extend feeding till pastures run out or get snowed out and we have to start using expensive bailed hay. Plant winter wheat again now and hope to gawd it don’t freeze when it comes up like it did this year. Spring wheat won’t get in the ground in time to harvest and besides, you don’t get much for spring wheat. Winter wheat always sells higher. Buy heating fuel or cut wood now. And buy/cut at least enough to last 9 months.

  206. Phil. says:

    wattsupwiththat (07:40:23) :
    UPDATE:

    Cryosphere Today has changed their color scheme presentation very recently.

    See the new color scheme here:

    http://arctic.atmos.uiuc.edu/cryosphere/

    It appears the archived images are unchanged and remain the same color scheme. I can’t find any announcement for why this has been done, if somebody knows of one, pass it along please.

    It appears to be more than just a color scheme but higher resolution too (longer to load :( ). The ‘old’ style data can still be chosen (old (SSMI)), from comparison it seems to me that the new data is from the higher resolution AMSR-E imager as opposed to the SSMI imager (from JAXA, see also Uni-Bremen). A different color pallet is also used, most notably it appears to be logarithmic rather than linear.

  207. Simmon says:

    “But the Cryosphere today images, such as these here: http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=15&fy=2007&sm=08&sd=15&sy=2008

    are PNG and totally lossless. So I would expect accurate pixel counts between them.”

    Have you looked at those images closely? Although the image pairs are distributed as PNGs, at some point the imagery has been through lossy compression. The errors are most obvious along the ice margins, visible as a speckled pattern within 8 by 8 pixel blocks. Try selecting a single color in the color bar with the magic wand tool in Photoshop, or a similar image processing application: it doesn’t select a single horizontal line, which you would expect in a clean image.

    REPLY: Simmon, good catch. the web page presentation compresses the image size, but if you do a right click and pull up the URL for the image pair that was generated, you get the full size image which is 1709 x 856 pixels.

    I did that and copied it into my imaging program and zoomed in to look for JPEG artifacts, and indeed they are there. So yes at some point they did use JPEG.

    This begs the question as to why would they take JPEG images, then save them as PNG? It defeats the purpose of lossless compression! It makes me wonder if UIUC is truly cognizant of what they were doing graphically.

    The new color scheme images, such as this one

    http://arctic.atmos.uiuc.edu/cryosphere/NEWIMAGES/arctic.seaice.some.000.png

    Don’t appear to have been JPEG compressed. I don’t see the edge artifacts common to JPEG compression.

  208. Jeff says:

    According Steven Goddard’s graph of the differences between 2007 and 2008 sea ice areas,

    http://spreadsheets.google.com/pub?key=pj0h2MODqj3jAfVXaaKHbCQ&oid=5&output=image

    on Aug. 4 there was a 41 percent difference and on Aug. 5 there was a 19 percent difference. Since it would seem unlikely during August for the sea ice area to significantly increase from one day to the next, this implies that the sea ice area between Aug. 4 and 5 in 2008 decreased by around 18 percent. An 18 percent decrease should be visible by eyeballing the maps. Such a large change should be obvious by comparing the maps on the CT site. Why isn’t it?

    There are several other days in which the sea ice area in one year or the other would have had to have changed by more than 10 percent in a single day. Why don’t these large changes show up in “recent ice area” graph on the CT site?

    Based on the numbers on the “recent ice area” graph on the CT site, there would have had to have been at least 3 days with sea ice area decreases of more than 350000 km^2 during a 19-day period. Does this seem at all plausible? I

  209. Jeff says:

    AndyW
    “Has UIUC been emailed to see what they say on the topic? ”

    I emailed UIUC way back on Sunday about how they create their images, but I haven’t received a reply.

    “It does beg the question though that if the data is right, and having read all this I still think it is, then why are the “popular” graphs at Cryosphere seemingly so far out of kilter?”

    As has been pointed out several times, the “popular” graphs, NSIDC’s data, the map at NCEP, and the Bremen maps all agree. Only Steven’s results don’t fit. What is it about his data that makes it seem more credible?

  210. dipole says:

    Jeff (09:09:00) says:

    Only Steven’s results don’t fit. What is it about his data that makes it seem more credible?

    As I noted above, after some initial difficulties with colour distortion in the images, I was able to reproduce his results. I also checked my calculation visually by masking out the counted pixels on the original image and it looked OK.

    What contrary evidence do you have that would diminish the credibility of this calculation?

  211. Jeff says:

    dipole
    “What contrary evidence do you have that would diminish the credibility of this calculation?”

    The graphs on the CT site agree with NSIDC’s data. The sea ice map on the NCEP site agrees with NSIDC’s data. The sea ice maps on the Bremen site agree with NSIDC’s data. The large jumps in differences from day to day in Steven G.’s graph require changes in ice area that aren’t apparent by looking at UIUC’s maps, nor by looking at the maps on other sites, nor by looking at any of the graphs on the UIUC site. So far no one has mentioned another data source that agrees with Steven’s results.

    I’ve posted the actual areas that I calculated from NSIDC’s data. If someone is willing to post the areas derived from the UIUC maps, it might add something to this discussion.

  212. Phil. says:

    Pamela and others:

    Here’s map of the world which I’m sure you’re all familiar with:
    http://www.creativeforce.com/productimages/World_MER_2.gif

    ‘Eyeball’ it and tell me which has the biggest area, Greenland or Australia.

  213. Phil. says:

    Talking of images, the direct NW passage is either nearly open or pretty well closed depending on what you look at.

    http://iup.physik.uni-bremen.de:8084/amsr/arctic_AMSRE_nic.png
    http://www.seaice.dk/iwicos/latest/amsr.n.ice.20080819.gif
    http://ice-glaces.ec.gc.ca/Ice_Can/CMMBCTCA.gif
    http://manati.orbit.nesdis.noaa.gov/ice_image21/D08232.NHEIMSK.GIF

    There’s not too much consistency there to me to be honest. Are all these just pretty to look at in the main?

    Actually they’re all very consistent as well as CT and NSIDC with the exception of the canadian site which appears to have a problem in that area (it doesn’t even agree with their own local radar which does agree with all the others).

    http://ice-glaces.ec.gc.ca/prods/WIS56CT/20080818180000_WIS56CT_0003925512.gif

  214. Gibsho says:

    A lot of credibility, as far as the impartial observer (if that’s possible) is concerned, has been hung on the argument that artic ice would not revert back to 2007 melt levels. It’s looking bad at this point. What a differnce 5a week makes. I’m really rooting for the skeptics to be right overall. More science, less politics/wishful thinking.

  215. Jeff says:

    Steven Goddard’s graph contained 18 1-day intervals. In 9 of those intervals, the difference increased and in the other 9 the difference decreased. If you make the very generous assumption that the 2007 sea ice area only decreased on days when the difference increased and that the 2008 sea ice area only decreased on days when the difference decreased, i.e., for each year there was NO melt on half of the days, and plug in the percentages from the graph, you find that the ice area would have had to have decreased by 38.6 percent between Aug. 1 and 19, 2007 and 40 percent between Aug. 1 and 19, 2008. More realistically, there would be a decrease on most days so that the graph implies that about half of the ice must have melted in the 18-day period in each year. The UIUC graph shows a decrease somewhere in the 20-25 percent range for Aug. 1-19, 2008. Can someone explain this apparent contradiction?

  216. Steven Talbot says:

    Jeff,

    “Can someone explain this apparent contradiction?”

    Hmm…well, you know, the blogosphere’s “tendency to knee-jerkingly blame everything on “global warming” [alarmism] makes for an easy story – but it is not based on solid science”, to borrow the words of our investigative reporter?

  217. dipole says:

    Jeff (10:39:21) :

    There are 3 things here –

    1. The UIUC/CT images
    2. Steven Goddard’s interpretation of these
    3. Everything else

    You claim everything in (3) is consistent. I have persuaded myself (1) and (2) are consistent. If you want to argue with this I can only suggest you check it for yourself. Otherwise your criticism of (2) would be more accurately directed at (1).

    Phil. (12:50:53) says:

    ‘Eyeball’ it and tell me which has the biggest area, Greenland or Australia.

    I thought the discussion was way past ‘eyeballing’ and map projections.

  218. Jeff says:

    dipole
    “1. The UIUC/CT images
    2. Steven Goddard’s interpretation of these
    3. Everything else

    You claim everything in (3) is consistent. I have persuaded myself (1) and (2) are consistent. If you want to argue with this I can only suggest you check it for yourself. Otherwise your criticism of (2) would be more accurately directed at (1).”

    -NOT. Several people have described problems with Steven Goddard’s method. If his interpretation of the UIUC/CT maps is correct, then UIUC/CT’s interpretation of UIUC/CT’s maps is not correct, because their graphs indicate a 10-15 percent difference between the sea ice now and a year ago.

    Steven Goddard’s results require implausible changes in the sea ice area from 1 day to the next. Do you really believe that there was an 18 percent (or more) decrease between Aug. 4 and 5, 2008? That’s over 800000 km^2. When the sea ice is decreasing by an average of around 1 percent per day, what could possibly cause it to decrease by 18 percent in one day? If there is a deliberate attempt on the part of AGW supporters to trumpet the decline in arctic sea ice, don’t you think that they would have made a big deal out of the loss of 18 percent of the ice in a single day? And made a bigger deal out of the loss of half the ice in less than 3 weeks?

    Implausible results should lead to questioning the method. At least that’s my belief.

  219. dipole says:

    Jeff (18:55:09) says:

    Do you really believe that there was an 18 percent (or more) decrease between Aug. 4 and 5, 2008?

    My own pixel count gives a +1% increase between those days. Previous to that I was comparing 20080811 and 20070812 where my pixel counting gave the same approx 30% difference that Steven G. claimed.

    So I don’t support the 18%, wherever it came from. If Steven G. got that figure he should certainly check his computation. But where does this leave your case against pixel counting?

  220. dipole says:

    I hereby coin the word ‘pixicle’ for a map pixel representing >85% sea ice coverage. It appears as a username on Google, but no previous usage in this context, I think.

  221. Pamela Gray says:

    Phil, I didn’t compare Australia to Greenland, and as a teacher, I am WAY past understanding map distortion. That concept is 5th grade stuff. But regarding Arctic ice, I looked at the Arctic picture from last year, same date, compared to this year, same date. More ice. Gonna be cold this winter. This isn’t rocket science.

  222. Jeff says:

    dipole
    “My own pixel count gives a +1% increase between those days.”

    Did you also compare the corresponding days in 2007 (which might be 3-4 rather 4-5)?

  223. dipole says:

    Jeff (20:47:06) :

    Did you also compare the corresponding days in 2007

    Pixicle counts

    20070805 30947
    20070806 30967

    20080804 36278
    20080805 36574

  224. Jeff says:

    dipole,
    Based on your results, I’m guessing that Steven Goddard compared days straight up rather than offsetting by 1 day for leap year as he did for his The Register article. Otherwise, your results don’t agree with his.

  225. dipole says:

    Jeff (22:04:53) says:

    Otherwise, your results don’t agree with his.

    But they do on the two images that were used in his article. Hopefully he will be back at some stage to clarify things.

    Meanwhile perhaps we will get an independent verification or refutation from RealClimate.

  226. wattsupwiththat says:

    I should mention that there is a lot of behind the scenes communications going on between Dr. Meier, myself, and Mr. Goddard.

    Dr. Meier has been most gracious with his interactivity. I hope to have more to report soon.

  227. Simmon says:

    The basic problem here is that the UIUC/CT images are a very poor representation of the sea ice data. Between projection issues (the data have been re-projected at least twice, and the final the projection is undefined), JPEG compression artifacts (which make it impossible to do a quantitative analysis), anti-aliasing/smoothing of the texture map (look along the ice margins), and an imprecise color palette (Mr. Goddard, could you give me the rgb value you use for 30-15% ice concentration on the UIUC maps? I can’t find any pixels below 30% on the images)–it’s not appropriate to use them for data analysis.

  228. Jeff says:

    It’s now been 6 days since this article first appeared. Not a single person has stated what the Earth areas of the sea ice coverages on Aug. 12, 2007 and Aug. 11, 2007 in the data set that UIUC/CT used is. Without the actual areas, how does anyone know what the difference is?

  229. Jeff says:

    Two days ago, I asked where the missing ice in NSIDC’s data is. All you have to do is compare the maps (and I’ve posted where NSIDC’s maps are) and show the area(s) where there’s ice in the UIUC/CT maps but not in the NSIDC maps. No one has done this. Why?

  230. Jeff says:

    Looking at the UIUC/CT and NSIDC ice area graphs, it looks like the ice area typically decreases this time of year by a little over 1 percent per day this time of year, with occasional spikes of a few percent. Yet according to Steven Goddard’s results, the difference between corresponding days in 2007 and 2008 changed by more than 5 percent on half the days that he measured. How can the difference change by significantly more than the ice itself changed?

  231. dipole says:

    Simmon (05:33:44) says:

    The basic problem here is that the UIUC/CT images are a very poor representation of the sea ice data…

    While I can’t disagree with any of that, I do believe they are good enough that the discrepancies cannot be accounted for just by the factors you mention.

    I can’t find any pixels below 30% on the images

    Not many for sure. I found 110 pixels in 15-30% range on 20080811, in a few scattered pockets. Is this typical of sea ice? If the maps are missing a big chunk of low concentration ice that could certainly screw up extent values.

    Jeff (06:56:20) says:

    All you have to do is compare the maps…

    Sorry, your links have scrolled into the misty past on this very long thread while I was counting pixels. The main difficulty in comparing maps from different sources is change in projection.

  232. Leon Brozyna says:

    A fine correction — about what one would expect from a science professional without an agenda to push. I’ve noticed, in comparing the NSIDC maps, that ice areas come and go; not so much the melting but from the movement of ice floes due to wind and currents. I’m still curious to see how the refreeze progresses this winter and how it’ll compare to the refreeze from the past winter.

  233. Paul K says:

    Anthony Watts: Thank you for publishing the correction, and thank you for pursuing this, and not using a hit and run approach to this story.

    Regarding some posted comments: The Northeast Passage seems to have opened today;
    http://www.iup.uni-bremen.de:8084/amsr/arctic_AMSRE_nic.png

    The Northwest Passage still shows some ice in the Viscount Melville Sound, and in the last week ice floes (50% ice) has been pushed down to block the end of the McClure Strait. Last year the NW Passage cleared on August 21; this year, the inland portion should be clear in the next few days. There is a ship in the NW Passage:
    http://www.sailwx.info/shiptrack/shipposition.phtml?call=DBLK

  234. Steven Talbot says:

    Steven Goddard,

    As one of those who has rather pursued you over this matter on this thread, may I offer my respect to you for the correction you have published.

    With regards,

    S.

  235. Jeff says:

    I second Steven Talbot’s sentiment.

    The UIUC data was misleading? Then maybe I was wrong about the method not working?

  236. dipole says:

    Yes – a good resolution of this issue.

    Jeff (19:23:21):

    The UIUC data was misleading? Then maybe I was wrong about the method not working?

    Given the absence of detail it was not unreasonable to be sceptical ;^). The 30% extent difference was also verified independently by another post on RealClimate, as well as further up this thread. Add that to the visual evidence from masking the matching pixels and it seems to me that pixel counting can be accurate and reproducible on the UIUC images.

    So this does seem to put the spotlight on the pre-2008 UIUC images. Did they discover an error in their mapping process which they silently fixed without updating the old images? Since the images are made to look like false-colour photographs it is natural for the unwary to interpret them as photographic evidence.

    I posted to this effect on the corresponding RealClimate thread, but was stuck in moderation for several hours and now the site is down.

  237. Paul K says:

    dipole, you missed the point. Mr. Goddard’s calclulated 30% increase was a sure sign that the calculation was wrong, as the sea ice extent was only higher by 10-14% than last year on August 11. A lot of the commenters here realized Steven Goddard’s result was wrong, right off the bat, because the result was contradicted by other sources.

    We still don’t know if Mr. Goddard used the correct method. A key piece of input data, the UIUC map from a year ago is different because it was apparently generated by a different method than this year. But we still don’t know if his pixel count method would have worked, using the correct input. We still don’t know the projection and method UIUC uses to generate their image, unless I missed it in all the detail above.

    But if we go to the data used to generate the maps and images, the sea ice extent has only increased about 10-14%, NOT 30%. This is the key.

    More importantly, all the general conclusions in the article above, relating to sea ice this year, have been pretty negated, including the title “Arctic ice refuses to melt as ordered”.

    Arctic sea ice area has continued to drop since this article was published on the net, and the sea ice area (a better measure than extent), is within 600,000 to 700,000 sq km of the minimum set in mid-September last year.

    Not likely to set a new low, without some unusual weather, but clearly should set the 2nd lowest NH sea ice area in the record.

  238. Steven Goddard says:

    Paul K,

    The pixel counting technique is done properly and works fine. I have found a few NSIDC maps from 2007, and when I apply the same methodology I get an exact match vs. their graphs.

    BTW – those of you concerned about the day to day variations in measurements should look at this. There is a lot of imprecision in the raw ice measurements.
    http://nsidc.org/data/seaice_index/images/daily_images/N_unsmoothed_alldata_timeseries.png

  239. dipole says:

    Paul K (22:49:45) says:

    dipole, you missed the point. Mr. Goddard’s calclulated 30% increase was a sure sign that the calculation was wrong,

    ‘Scuse me? Or one of his data sources was wrong. As he says:

    while their (UIUC) 2008 maps appear golden, their 2007 maps do not agree well with either NSIDC maps or NASA satellite imagery.

    The CT site invites you to compare images from 2 different dates. It’s not unreasonable to assume that the images are in fact generated similarly and directly comparable. It appears this is not the case, and this was the undoing of his argument, not the pixel counting.

  240. Phil. says:

    UIUC maps seem to compare very well with the high resolution images obtained with the AMSR-E imager, see below for a comparison on 8/11/07:

    http://iup.physik.uni-bremen.de:8084/amsredata/asi_daygrid_swath/l1a/n6250/2007/aug/asi-n6250-20070811-v5_nic.png
    http://arctic.atmos.uiuc.edu/cryosphere/deetest/deetmp.7104.png

    The NSIDC image looks rather similar too to be honest when one considers its lower resolution (for day earlier):

    http://nsidc.org/news/press/2007_seaiceminimum/images/20070810_extent.png

  241. William Chapman says:

    Hi Folks,

    There is no difference between the data or the way the 2008 and 2007 images were produced in the comparison images on the Cryosphere Today. The apparent differences Mr. Goddard observed between the NSIDC values and those produced comparing images from the CT are almost entirely due to the mistake of using pixel counting to compute area on severely distorted satellite projections.

    Minor differences may come from:

    (1) NSIDC uses a longer temporal averaging (around 10 days, I think) compared to the CT single-day plots.
    (2) NSIDC uses a 15% threshold data cutoff; the CT cuts off concentrations below 30% in the comparison images.

    Still, the above only account for percentage change differences of ~3%. The majority of the apparent difference comes from projection fun.

    The problem with the satellite projection is that the center of the images are distorted when compared to the periphery of the images. The distortion is such that a ‘feature’ in the middle of the image (higher latitudes) has more pixels representing it than the same feature would at the periphery of the image (lower latitudes). The ‘feature’ in this case can be thought of as the ice extent difference. I am referencing a schematic I drew that helped me understand the issue:

    http://arctic.atmos.uiuc.edu/cryosphere/equal.area.projection.png

    The satellite projected images should not be used to compute area/extent (or differences between area/extent from one year to the next). When I use the raw equal area grids as they come from NASA, before reprojecting onto my satellite projection, I get a 2007/2008 difference of around 10% in ice area around August 11 – consistent with what NSIDC is reporting. This has reduced to around 5% at present.

    If my understanding of the issue is complete, a similar difference in winter, when the ice edge was nearer to the periphery of the CT satellite images, would be misrepresented in the opposite sign (smaller) than that computed on a standard equal-area grid (ala NSIDC).

    As with any map projection comparison, there is nothing ‘wrong’ with any of the different projections, they all have their place. Problems do sometimes arise when comparisons of summary statistics computed on one are made with similar numbers from the other, as is the case here.

    REPLY: Thank you for your posting Mr. Chapman. Here are questions for you –

    1) Why don’t you answer your email contact link on the CT web page? I and other have sent queries this year, with no response. For example, I sent an email two days ago.

    2) Why do you have a quote from a politician (Al Gore) on a web page presenting science? This is a question many people have raised.

    3) What prompted the color scheme change in recent days

    Thank you for your consideration – Anthony Watts

  242. laidbackracer says:

    so now i am wondering if the incorrect UIUC 2007 maps will be corrected…. suspect they have already caused quite a few drama’s….recently saw a local doco (ABC 4 corners) which seemed to use the incorrect data as the basis of its presentation. Total sea ice since 1979 doesn’t seem to have changed much …if this data is correct??
    so its all a bit of a storm in a tea cup…but would expect a small loss of ice due to our current position in regard to total sunspot activity trend.

  243. Steven Goddard says:

    William Chapman,

    Thanks you very much for your fine explanation. The problem I (and others) had was that the side by side comparison link from your web site showed visually a large difference between August, 2007 and August, 2008. August 15, 2007 vs. August 14, 2008 is a good example.

    Pixel counting is simply a way to quantify what is processed by the brain when viewing an image. In this case, the 2008 image has 41% more ice pixels than the 2007 image. NSIDC extent graphs and UIUC area graphs both correctly show about 10% difference for that date.

    This is a substantial difference (41% vs. 10%) and misled me and numerous other people. Perhaps you should put a disclaimer on the page that the images are not proportionate and can not be directly compared? Having a side by side comparison is quite misleading in the absence of such information.

  244. William Chapman says:

    1) Why don’t you answer your email contact link on the CT web page? I and other have sent queries this year, with no response. For example, I sent an email two days ago.

    limited time – sorry.

    2) Why do you have a quote from a politician (Al Gore) on a web page presenting science? This is a question many people have raised.

    didn’t realize it was a concern for many people. All references to Al Gore have been removed.

    3) What prompted the color scheme change in recent days

    I added three new color schemes about 40 days ago (July 11; is that ‘recent’?). I was hoping for more detail in the images “from the satellite perspective” in the images shown on the main page. The AMSR-E data provide more spatial resolution so I switched data sources and color schemes for those home page images. IMPORTANT: The data used for all other timeseries and comparison graphics have stayed the same (SSMI) obviously, to avoid any issues with data inhomogeneity in time. The AMSR-E data source is only used for the high resolution Northern Hemisphere graphics on the main page. I hope to convert the Southern Hemisphere as well over the next month. The AMSR-E is a relatively new platform, so maybe after it has been around for 10-15 years or so, and has a proven track-record, we can switch the timeseries and other data over entirely to that platform. I have included links to the old SSMI images on the main page for those who prefer them or want to compare current conditions to historic conditions (prior to the AMSR-E launch).

  245. dipole says:

    William Chapman (07:27:26) says:

    Hi Folks…

    Thanks for your input, and especially the confirmation of the consistency of your image sequence. Sorry to have questioned this. I have an amateur interest in map projections and had tried to allow for the distortion as described elsewhere on this thread.

    Anyway you have given me an incentive to have another look – the exact quantification of the error in the original article remains an interesting puzzle and it would be nice to tie up the loose ends. Is there a pointer somewhere to the exact projection used?

  246. dipole says:

    Uhoh. Escaped italics.

  247. Paul K says:

    Anthony, Mr. Albert Gore Jr. hasn’t held a political office for almost eight years. He has been active in advancing scientific efforts for over 30 years. Why isn’t his view on scientific developments appropriate, especially in view of his Nobel prize? None of the (sometimes dubious) sources of information you post here have a Nobel, that we know of. In fact, we don’t know anything about the author of the current article under discussion.

    Are you saying that science information sites shouldn’t have anything related to various policies discussed? This would be a serious limitation on analyzing alternatives.

    Or are you saying that former politicians who held government positions, shouldn’t be involved in activities involving government entities? If this is the case, would you object to former presidents like George H.W. Bush or Bill Clinton addressing a high school class? This also cuts off the rights of such persons from participating in many activities after their period of public service is completed.

    I don’t understand your point, nor how it is relevant to the current discussion.

    REPLY: And I don’t understand yours. The less politics involved in science the better. Gore has been, and still is a politician. See this:

    One of the most coveted speaking slots of the Democratic National Convention will go to former Vice President Al Gore. He is scheduled to appear on the final night of the convention as Senator Barack Obama prepares to accept his party’s nomination.

    Story here: http://www.iht.com/articles/2008/08/19/america/gore.php

    So shall you now try to tell me that his “coveted slot” (to quote the article) at a political convention next to the potential next president of the United States is not political work on his part but is science?

  248. Pingback: Cryosphere Today Makes Changes - Improves product, drops Gore comment « Watts Up With That?

  249. Hud says:

    Mr. Gore won a political prize, not a scientific award. Throughout history, politics has corrupted scienTISTS, but in the end, SCIENCE prevails.

  250. I have a wild suggestion…

    In looking at this, I have taken NSIDC satellite data, and made my own images. I have projected them on the globe using a viewpoint above the pole which has a tangent to the surface at about latitude 27. This gives a very close match for the land masses.

    Everything lines up very well indeed if I assume that the 2007 image is actually showing the area with ice cover at 50% or more; but the 2008 image is showing the area with ice cover at 30% or more. In particular, the NSIDC data allows me to construct an overlay of the 50% ice, or the 30% ice, and I get a very very close match with the UIUC images when I put them all together.

    Here are the numbers, comparing UIUC images assuming this projection, the JAXA report of 15% ice cover or more, and my own simple area-weighted sum of extent from the NASA data (f13 channel, ~25km grid)

    For 12-Aug-2007
    36683 pixels of ice on the UIUC image
    4469014 sq km projected area
    4201452 sq km at 50% or more, by simple count of NASA data
    5057390 sq km at 30% or more, by simple count of NASA data
    5679174 sq km at 15% or more, by simple count of NASA data
    5421094 sq km at 15% or more, reported by JAXA

    Some small differences are to be expected from processing differences; but basically the UIUC projected area lines up well with the extent of 50% ice, and JAXA lines up with the extent of 15% ice.

    For 11-Aug-2007
    47822 pixels of ice on the UIUC image
    5882718 sq km projected area
    4612971 sq km at 50% or more, by simple count of NASA data
    5783576 sq km at 30% or more, by simple count of NASA data
    6462289 sq km at 15% or more, by simple count of NASA data
    6291563 sq km at 15% or more, reported by JAXA

    This time the UIUC projected area lines up well with the extent of 30% ice, and JAXA still lines up with the extent of 15% ice.

    Furthermore, I have projected the NASA data onto bitmaps, projected to align with UIUC images, and I get close agreement with the images using 50% for 2007 and 30% for 2008.

    I’m going to stick my neck out and predict that the UIUC archived images are actually showing the extent of ice at 50% or more, and that the recent UIUC images are showing the extent of ice at 30% or more.

  251. Oops. On my previous comment, the 11-Aug-2007 above the second set of figures should be 11-Aug-2008

  252. dipole says:

    Duae Quartunciae (16:10:29) says:

    I’m going to stick my neck out…

    Excellent piece of calculation DQ. If you are correct there should be a sudden and physically implausible jump in pixel count at some point in the UIUC image sequence.

  253. I’m going to pull my head back in…

    dipole had an excellent way to test my hypothesis, and it doesn’t seem to work.

    Basically, there is no discontinuity I can see. The data in 2007 is closer to the 50% ice extent; but this may just be a co-incidence. I’m now really curious; but it could be a problem in my own maths. I’ll double check… no assurance I can come back to this soon, sorry.

  254. hunter says:

    Your explanation of your reconciliation of the two data sources is being used by AGW believers as an admission that inspite of the facts, there is as large a melt this year as last.

  255. dhogaza says:

    That’s because there has BEEN as large a melt this year as last. Remember that cold winter you guys were all excited about, and the recovery of arctic ice that y’all were going on about? Remember you guys being so excited that ice extent at the beginning of the melt season was greater than in 2007?

    Yes, the ice extent up to this point in 2008, is about 5% higher than in 2007. But since there was more ice to begin with, the amount that’s melted this year is about the same as last year (thus far).

    Nothing odd about that. You don’t even need algebra to figure it out, nor multiplication and division. Simple subtraction is sufficient.

    Or just look at the NSIDC graph…

  256. Demesure says:

    “Or just look at the NSIDC graph…”

    You’re correct dhogaza, we should just look at graphs instead of relying on AGW experts’ predictions. For 2008’s minimum extent, they mostly (9 out 17 teams) predict a lower or equal extent compared to the 2007 low record.
    And that’s a “prediction” just … 2 months ago (June report) :
    http://www.arcus.org/search/seaiceoutlook/report_june.php

  257. Phil. says:

    Demesure (06:58:19) :

    You’re correct dhogaza, we should just look at graphs instead of relying on AGW experts’ predictions. For 2008’s minimum extent, they mostly (9 out 17 teams) predict a lower or equal extent compared to the 2007 low record.

    “Of the 17 responses, all suggest that the extent will remain lower than the historical average (i.e., mean 1979–2000 September values) of 7.0 million square kilometers.”
    All correct.

    “Five (5) responses suggested a less dramatic loss than in 2007 (i.e., 4.3 million square kilometers) and closer to the pre-2007 long-term trend of approximately 10% loss per decade.”
    Too conservative, already passed.

    “Five (5) anticipate a repeat of the dramatic loss of 2007.”
    They look like being on the money.

    “Four (4) suggest a loss even greater than that experienced in 2007.”
    They could still be right but it will be close to the wire.

    “Three (3) give more detailed reports on regional trends.”
    Didn’t play the game.

    Reading the more detailed report yielded this . The current value is 5.4 and still dropping so the more conservative predictions have already been passed, it looks like it will end up close to the average of the predictions. Not bad since most of the melt has occurred since those predictions were made.

  258. Phil. says:

    Sorry image tag didn’t work, I was referring to this graph:
    http://www.arcus.org/search/seaiceoutlook/report_june.php

  259. Rick says:

    http://nsidc.org/images/arcticseaicenews/200807_Figure1.png
    http://nsidc.org/images/arcticseaicenews/20080717_Figure1.png
    http://nsidc.org/images/arcticseaicenews/20080801_Figure1.png

    Interesting…as long as the sea ice matched the “median” line, it was called the “median”. As soon as the ice was less than that line, it is called “normal”.

    So is the melting sea ice below “median” or is it below “normal”? And is it “normal” to be below “median”?

    Wouldn’t it be more scientific to be more constant with their statistical terms?

    Links to the maps where I noticed the change.

  260. Fred Houpt says:

    I just read this in today’s English internet Der Spiegel:

    http://www.spiegel.de/international/world/0,1518,574815,00.html

    I had to laugh. I read this part: “Many, in fact, have begun talking about an Arctic Ocean completely free of ice. The way things look now, such a scenario is not in the immediate future, but a sad, new record may indeed by set this year: The tiniest quantity of Arctic ice since scientists first started taking measurements. The Arctic melting season has another three weeks to go before the flat angle of the sun’s rays mark the onset of winter.”

    I find the unsupported claim that the Arctic has not seen such a small ice coverage outrageous. No reference to any study whatsoever. In my view, this is the opinion of the writer but it comes across sounding as if based on known facts.

    Another thing that I noted missing was the fact that the waters all along the passageways are often shallow and littered with hull-destroying boulders just waiting for a deep draft tanker to troll by. These waters have NEVER ever been dredged and to think that they will raises the big question of who will pay the astronomical price? Canada? Not bloody likely. I do not know how the ice breakers make their way through without hitting rocks? Perhaps they are double hull lined, go slowly and have advanced sonar to make sure they avoid trouble?

    Finally, what I found ridiculous is the premise that at the last days of August, when this current ice free/ice thin environment has shown up will somehow in future years provide a workable shortcut for big ships? I mean, what are they thinking? The ships would have to enter the northwest passage at a time when the vast majority is still ice covered and wait until it breaks up. If it takes till the end of August, what are the ships going to do? Wait it out up in northern Hudson’s Bay? It makes no sense to start planning to use the passage because as I say it is a far too short period and the lane ways are not dredged. For me the use of the NWP is still a far off dream.

  261. Pingback: Timothy Birdnow » Correction on NSIDC Graph

  262. Pingback: Arctic Ice Growth, 2008 - How Much? « Watts Up With That?

  263. Hello says:

    I dont know about you guys but i dont belived in this post.

    Here are a NASA link which shows that sea ice retreated enough to create open waters all the way around the northern ice pack.

    http://www.nasa.gov/mission_pages/IPY/multimedia/ipyimg_20080909.html

    You articles shows sattelite images but are they real? Do you have real links for that images ? Can you post them here in order to confirm your theory ?

  264. We are currently, according to Cryosphere Today data, at last year’s record low sea ice measurement. Any further melt, at all, will result in a new record.

Comments are closed.