What Is Normal Arctic Ice Extent?

Guest post by Steven Goddard

I have been noticing in recent weeks that NSIDC extent is much closer to their 1979-2000 mean than NANSEN is to their 1979-2007 mean.  This is counter-intuitive, because the NANSEN mean should be relatively lower than NSIDC – as NANSEN’s mean includes the low extent years of the 2001-2007 period.  Those low years should have the effect of lowering the mean, and as a result I would expect the NANSEN current extent to be equal to or above the 1979-2007 mean.

(For exclusive subsets A and, B where subset A has a mean value of 14 and subset B has a mean value less than 14, then the mean of the full set AB must also be less than 14.)
http://eva.nersc.no/vhost/arctic-roos.org/doc/observations/images/ssmi1_ice_ext.png
http://nsidc.org/data/seaice_index/images/daily_images/N_timeseries.png

I overlaid the NANSEN graph on top of the NSIDC graph below, and it is easy to see how large the discrepancy is.  In fact, the NSIDC mean sits at about one standard deviation below the NANSEN mean – which makes little sense given their base time periods.  It should be the opposite way.

(Note – the NANSEN and NSIDC measuring systems are not identical, and I had to make a shift along the Y-axis to line them up.  However, the X and Y scales are identical for both graphs in the overlay image.)

NANSEN and NSIDC combined
As mentioned above, one might expect that the current NANSEN extent would actually be above the 1979-2007 mean.  But something odd happened with the NANSEN data on December 13, 2008.  Overnight it lost about 500,000 km2 of ice, as Anthony captured in the blink comparator below.
nansen_sea_ice_area2-520
Is it possible that there is still an error in the NANSEN data?  The discrepancy in the offset from the mean vs. NSIDC is rather large – nearly large enough to place California inside.  What are your thoughts?
I asked Dr. Walt Meier from NSIDC his opinion, and he replied (as always) courteously and promptly.  His answers are below:

Nansen uses a different algorithm to calculate the sea ice extent. The algorithms differ in the way combine the raw data together to estimate extent. As long as one uses the same algorithm, the stories are all the same, but the details can differ, more so at certain times of year. When there is a diffuse, broken up ice edge and melt is starting is one such time.

I suspect the Bering Sea is probably the region resulting in most of the differences. While our algorithm shows the region has mostly “ice-covered” the ice cover there is very fragmented, broken-up, and thin.

….

The other thing that’s important to mention is that I was referring simply to discrepancy between how close the current lines are to climatology. However, there is also generally an “offset” between algorithm outputs – a bias or mean difference between the algorithms that is fairly consistent throughout the record. That is why NSIDC’s climatology is different than the Nansen climatology.

The important thing to remember is that there is a good consistent record from the passive microwave data as long as you consistently use the same algorithm and the same processing. But you can’t mix and match products.

Advertisements

  Subscribe  
newest oldest most voted
Notify of
TERRY46

As always if you don’t like the outcome just move the goal post.

Aron

Blatant lies:
http://www.guardian.co.uk/environment/2009/apr/28/climate-change-poles
“In the past four years, air temperatures have increased, sea ice has declined sharply, surface waters in the Arctic ocean have warmed and permafrost is in some areas rapidly thawing.”
Note the use of alarmist language “sharply” and “rapidly”

Mike Bryant

I appreciate the answer from Meier, but I am not sure it goes to the heart of the discrepancy. What am I missing?
Mike
Also, according to CT the Global Sea Ice Anomaly is now 765,000 sq. km. which is a large enough area to contain, Virginia, West Virginia, Louisiana, Mississippi, Ohio, Maine and Pennsylvania. Oh yeah Rhode Island and D. C. will fit in there too…

Mike Bryant

Also,
I wonder if Dr. Meier would comment on why these side by side graphs are using different comparison years… 2007 in Arctic vs. 2008 in Antarctic?
http://nsidc.org/data/seaice_index/daily.html

Aron

I posted this as a public note on Facebook:
“Read this article by the Guardian’s John Vidal (a man who enjoys lying so much that he probably has bought himself a golden throne to go with his ego)
http://www.guardian.co.uk/environment/2009/apr/28/climate-change-poles
Sounds frightening doesn’t it? Let’s take a look at just one sentence of it:
“In the past four years, air temperatures have increased, sea ice has declined sharply, surface waters in the Arctic ocean have warmed and permafrost is in some areas rapidly thawing.”
Notice the use of alarming language designed to create public hysteria. Now let’s take a look at how Arctic Ice is really doing according to the National Snow and Ice Data Center (NSIDC).
http://nsidc.org/data/seaice_index/images/daily_images/N_timeseries.png
As you can see, arctic ice is in a very healthy state. It’s been growing yearly for the past few years and is almost back up to its long term average. So where did John Vidal get the idea that ” sea ice has declined sharply, surface waters in the Arctic ocean have warmed and permafrost is in some areas rapidly thawing”???
The Guardian lies every single day. It’s that simple. Their agenda is to force you to believe the world is going to end unless you change your ways and conform to their mind controlling, soul destroying socialism.
Stick your middle finger up to them. Go enjoy your lives and live the way YOU want to live.”

Paul James

“nearly large enough to place California inside”
YIKES !!!
Ring the alrm bells !!!!
Has any one told Mr Waxman that CA was moved North and then apparently melted ? Or did it evaporate ? I always have trouble keeping those two in order.

Paul

Just out of curiosity, why doesn’t NSIDC use the 1979-2007 average?

George E. Smith

Why would you accept just 15% coverage as full coverage. Maybe 85% sea ice would classify as coverage; but 15% ice coverage is open water to me.
George

ak

Shouldn’t the ice be above normal given the lull that the Sun has been in and the low(er) temperatures experienced across the region this past winter? It seems weird that given those factors it’s still below the average.

Mark T

It seems weird that given those factors it’s still below the average.
Do you mean “it seems weird that it hasn’t immediately rebounded to above normal conditions in just 1 year“?
Sounds a little weird, actually, when you put it in this context.
Mark

Cassanders

@ak
Recall that there are still heat stored in the sea. IF the “large” areas of young ice continues to thaw slower than the previous years, I would think it is reasonable to look for connections to sun and air temperature.
(But as I have stated earlier, I think the interannual variation between ice area minima is more sensitive to wind and currents than thawing, per se).
Cassanders
In Cod we trust

Following years will have to take into account the area covered with ice over Canada..:)

John Galt

We have only tracked Arctic sea ice extent for 30 years. How do we have enough of a sample to know?
It’s only 30 years. That’s a lot in human terms, but that’s a meaningless fraction of the number of years since just the last ice age.

TERRY46

Off topic but I just saw on Fox news where Antarctic ice is starting to break up and they don’t know why it’s happening but for sure of course it’s due to the warming.I didn’t think Antarctic mattered any mores???

Dr.Meir: Nansen uses a different algorithm to calculate the sea ice extent. The algorithms differ in the way combine the raw data together to estimate extent.
Perhaps I am too fool to understand, but, why the need of an algorithm?
Why not just a picture?…You know, those images taken by a thing called a camera..

RBerteig

I’m sure the choice of 1979 to some recent year as the span to average for comparison is based mostly on the fact that we don’t have any satellite-based coverage or extent measurements from any earlier than 1979. But what is the basis for assuming that this particular 20+ year period is long enough to cover all of the natural variation in the ice pack?
Isn’t it possible that recent low excursions are simply part of the natural variation and not an issue at all?
If we don’t know if the reference period is “normal”, it isn’t particularly honest to call excursions from it “abnormal” after all.

The approaching minimum will make things easier. At Boulder, new age scientists will measure ice deep from below 🙂

John H

That is peculiar that the NANSEN mean is higher than the NSIDC mean.
Makes no sense.
It’s good to have Dr. Meier accessible and so responsive.
Perhaps he stops in here on occasion.
So I have a few questions for him.
1) Why couldn’t, and hasn’t, the NSIDC adapted it’s mean from their current 1979-2000 mean to a 1979-2007 mean?
2) Will the Dr. acknowledge that in doing so would reduce the NSIDC mean?
3) With 30 years of record why are none of the most recent 9 years included in computing the mean?
4) Wouldn’t removing any 9 year period from calculating the mean make that mean a poor representation and basis for comparison?

Ed Scott

Risking the wrath of persons intent on directing the forces of Nature against me, I venture to say that the normal extent of Arctic sea ice is whatever it is at the moment. Persons with skins thinner than my Early Girl tomatoes and whose theme song might be Cole Porter’s “I’ve Got You Under My Skin,” might find fault with that observation. But as has been said, nobody disagrees that Nature is the norm.

Steven Goddard

Adolfo,
Good questions. In winter the Arctic is dark; covered with clouds, and the ice is fractured and interspersed with open water. Visible wavelength photography is out of the question, so satellites use microwave data and organisations like NSIDC calculate the extent (area of the sea with greater than 15% ice concentration.)

George E. Smith (14:19:31) :
Why would you accept just 15% coverage as full coverage. Maybe 85% sea ice would classify as coverage; but 15% ice coverage is open water to me.
===============================================
You can use any number if you apply it consistently, so use 50% or 92% or 27% it will make little to no difference over the entire time series.
Please note that at the ice edge ice must reach several centimeters thickness before becoming translucent and rise above the surface as nilas, also in rougher water frazil condense until pressure forms cakes of slush that get pushed together to produce pancake ice. So 15% Coverage does not mean 85% Ice Free, means pancake or non dark nilas covering 15% of the area.

> Mike Bryant (14:08:47) :
> Also, I wonder if Dr. Meier would comment on why these side
> by side graphs are using different comparison years… 2007
> in Arctic vs. 2008 in Antarctic?
I think it’s because those are the record years of interest in each hemisphere. 2007 was a record low year for the Arctic ice pack. 2008 was a record high year for the Antarctic ice pack (not widely covered by the MSM; it goes against their AGW religion).
I have a question for Anthony or anybody else. Where on the NSIDC website do you find plottable data for the mean and current (or other) year? Rather than rely on the cramped graph at the IARC-JAXA website, I have a nice spreadsheet graph at home of IARC-JAXA data that fills my 24″ 1920×1200 monitor. I’d like to be able to reproduce the NSIDC graphs at home. Direct URLs to the data please. I’ve tried plowing through http://nsidc.org/data/docs/noaa/g02135_seaice_index/index.html and I sea stuff about monthly data, dailiy image files, and shapefiles. I want daily text data. Help.

Jim Papsdorf

Fox News just reported on the “Unexpected Eisdicke” study Anthony posted yesterday-it was brief but they mentioned that an expected ice thickness of 2 meters turned out to be 4 meters. The Goracle has a problem handling this one !!!!!!

Neville
Harold Ambler

Speaking of ice, I’ve done some more reporting on Catlin and interviewed Kenn Borek Air v.p. of operations Sean Loutitt regarding the eventual removal of the currently underfed expedition team members.
Story here:
http://talkingabouttheweather.wordpress.com

F Rasmin

In addition to the cracks in the antarctic ice, there are now cracks in the Australia governments stand on climate change.
http://www.theaustralian.news.com.au/story/0,25197,25406747-11949,00.html

Mike Bryant

Walter,
You said that 2008 was a record high year for the Antarctic ice pack… Here is the graph from CT:
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.area.south.jpg
Maybe I’m looking at it wrong, but it appears that 2007 was the record high for SH ice…
Thanks,
Mike

Just Want Truth...

” But something odd happened with the NANSEN data on December 13, 2008 ”
Does anyone know the whereabouts of James Hansen on the date in question? 😉 jk

Just Want Truth...

NANSEN — Hansen, rhyming,…. hmmm, something’s up there… 😉

Looking at the Sea ice graphs that Anthony kindly displays on the RHS of his webpage can anyone provide a reason behind the persistent annual small “blip” in June? This appears to suggest a small regular annual increase at a time when ice is melting fastest. Quite strange.

Tom in Florida

The notorious base line of 1979 – 2000 cannot be considered normal. It is just an arbitrary reference period with no foundation for anything. We really do not know what “normal” arctic sea ice coverage is.

Mike Bryant

Another thing that seems odd… CT has the SH Sea Ice Area at 1,227,000 sq. km. ABOVE the 1979-2000 mean, while NSIDC has the SH Sea Ice EXTENT just barely meeting the 1979-2000 average. Now I know that area and extent are different but isn’t that a pretty big difference?
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.south.jpg
http://nsidc.org/data/seaice_index/daily.html
On the Arctic Roos, I know it’s the Arctic and not the Antarctic… however even though the area and extent ARE very different numbers, they are still each about the same distance from the mean…
http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic
Mike

David S

“What Is Normal Arctic Ice Extent?”
The earth spent most of the last half million years in an ice age. By that standard “normal” would mean ice covering most of Canada and much of the northern US. Under “normal” conditions the place where I’m sitting right now would be under 1/2 mile of ice. I think I prefer the abnormal conditions of the Holocene.

OceanTwo

From a statistical standpoint, I don’t see a real problem with -2000 instead of -2007. The data in that region is fairly consistent with little variation. Since the point of contention is the recent history (the last decade), it’s nice to have a control signal outside the area in question.
In addition, if you move the end points for the 79-00 readings, the average stays fairly consistent (and flat). ‘choosing’ end points close to the 2007 minimum (or even specifically choosing a certain point) can distort the results drastically. Whenever a signal is comparing specific results to itself, you have to be very skeptical of the results and examine the comparison very carefully.
I’ve seen quite a few graphs of this ice coverage data [media bias supporting the ‘polar bears are dying’ standpoint] which don’t go much beyond 2007 – even though we have quite up-to-date daily data – disingenuous to say the least.
It’s pitiful that whenever the facts don’t fit the premise they are summarily dismissed (as the antarctic data has) on a whim. It wouldn’t surprise me to see that the sea ice coverage doesn’t matter; it’s now the volume that’s monumentally important (data which can be distorted or massaged and thrown out with a “don’t bother with the boring details, you wouldn’t understand it: our top scientists have assured us we are all going to die”)

Leon Brozyna

Out of curiosity, I went back to the original story as covered here by Anthony on 13 December:
http://wattsupwiththat.com/2008/12/13/something-is-rotten-in-norway-500000-sq-km-of-sea-ice-disappears-overnight/
NANSEN’s explanation, which was later added to the post, seems weak as it cites changes from 22 October when in fact they began 11 September. It may very well have been a small error that kept compounding, though NANSEN never said that explicitly, but ever since then I’ve been a bit dubious about NANSEN. Dr. Meier makes a good point when he says, “The important thing to remember is that there is a good consistent record from the passive microwave data as long as you consistently use the same algorithm and the same processing.”
Personally, my favored source is IARC-JAXA. The comparisons this September should be most interesting.

Heat or no heat the maximum winter ice extent has no real significance in my opinion as the arctic is constrained by landmass. An increase in the ice extent/area as opposed to the last few years at maximum melt in mid september is what we want to see. In view of the reported increase in ice thickness and changed weather/circulation patterns in the arctic as opposed to the 2007 season this looks increasingly likely in my view.
Saying that ( fingers crossed ) we dont want to see too much of a dip in temperatures in the coming years – not good for nature in general and certainly not for us human saps. The alarmists are not worth dying for just to make a point – not that we have any control over anything anyhow.

ak

Mark T said “Do you mean “it seems weird that it hasn’t immediately rebounded to above normal conditions in just 1 year“?
Sounds a little weird, actually, when you put it in this context.”
Shouldn’t be all that weird really. Snow melt in mountains can be lengthened by a strong winter and cooler spring, likewise snow melt can be quicker and earlier if the conditions are warmer over the winter and spring. In some mountainous areas, snow can last well into summer months specifically because it’s on the northern slope and away from direct sun.
True, sea ice can last up to a few seasons – it’s rare in mountains for that – so it’s not a perfect analogy, but given the homogenous nature of the Arctic sea ice compared to mountain ranges, the analogy doesn’t seem all that far-fetched and it seems like cooler winters and lower sun energy would have a more widespread effect across the whole…
How long would you postulate the “lag” be for a “rebound?”
@Cassanders, there must be meteorological and current data available to compare from year to year?

aurbo

Moving the goalposts. That’s what the media does and their information is provided by, or filtered through various so-called science organizations.
This past year we’ve been hearing about how Arctic ice is on a steady decline which will inevitably lead to an ice-free summer in the very near future. In fact, the decline in relation to recent years ended in 2007. It has been coming back rapidly this past winter and spring. So what does the media do now?
They either continue to report on the 2007 ice year, or, if they have been told by a flood of e-mails pointing to the current data which has the ice on this date above the previsous high for this date back to 2003, the previous highest in this time frame. (In fact, yesterday, Apr 28th, the extent of Arctic ice as reported by JAXA was 245,000 km² above 2003 and getting very close to the 1979-2000 mean).
So does the AGW crowd concede? You’ve got to be kidding! They simply move the goalposts. Now they say that it is the ice volume that counts and most of the present ice is 1-year ice and the Polar ice sheet is much thinner than normal. Thus the ice volume is a lot lower.
But now we have the real measurements obtained this month by the Alfred Wegener Institute for Polar and Marine Research (featured in an article on WUWT a day or two ago) that measured the ice thickness with a sophisticated electromagnetic sensing device attached to a revamped C54 (DC3) and towed about 20m above the ice. Their route crossed the Pole and they ran several legs in various directions from the Pole. (I know, all of the routes ran southward, but at different longitudes). What did they find?
The first reports said that the ice thickness at the North Pole was about twice the amount they expected to find. Since the volume varies as the cube of the thickness, what do the AGWers do now. They’ve already backed the goalposts up to the grandstands.
The answer is bait and switch. They’ll probably shift their reportage to the Catlin expedition. This is a highly publicized trek by 3 grossly unprepared publicity seekers whose equipment failed early-on and any data they may have collected will be of little, if any scientific value. Nevertheless, they will be lionized by the media, especially the BBC a co-sponsor, for their daring and acumen as they dutifully report in regard to the ice situation that “It’s worse than we thought”.
As a matter of fact, if the AGW community had an emblem, their motto would be “It’s worse than we thought”.
Reply: Small point of correction. If you vary only one axis of a solid, ie the thickness and not the width or length, (just z, not x or y) the volume change would be directly proportional or simply double, not proportional to the cube of the variance. ~ charles the moderator.

Leon Brozyna

BTW – Steve, perhaps you could do a post to explain to everyone how the adjustments are made by IARC-JAXA in June & October. There was an explanation in a comment awhile back on an earlier sea ice post but I guess not everyone reads the comments.

This is a bit off topic, but I thought you should see it. Re the assumptions of AGW Truth!!! Look at this drivel–especially the chart.
http://www.sciam.com/article.cfm?id=limits-on-greenhouse-gas-emissions

JeffK

MarcH said…
“Looking at the Sea ice graphs that Anthony kindly displays on the RHS of his webpage can anyone provide a reason behind the persistent annual small “blip” in June? This appears to suggest a small regular annual increase at a time when ice is melting fastest. Quite strange.”
I’m glad you have seen this too. I brought this up on a previous thread a month-or-so ago bit it was never explained. I ‘seriously’ doubt the ice is actually doing what the graph is supposed to say that it is. Can someone please address this issue?
Thanks,
Jeff

PMH

OceanTwo (16:57:38) said “From a statistical standpoint, I don’t see a real problem with -2000 instead of -2007.”
Although this argument may be correct and there may not be a “real problem”, the more words required to explain the data the more the data should be questioned.

Wondering Aloud

ak
We don’t know how much of an effect the quiet sun should have, Lief thinks not that much it seems, but, it sure wouldn’t be large this quick. Thermal inertia is huge and input change small.

steve

Hey Anthony,
I know this is completely off topic but I had a discussion a while ago with a guy called Foinavon (I think) about climate’s sensitivity to CO2. I think it was to do with Hansen having nailed the sensitivity/gain, g, to 3.
One of the papers I referenced was by John Christy and was, at the time, not published, hence Foinavon was not interested. Apparently this has now been published and is available to view at http://www.nsstc.uah.edu/atmos/christy_pubs.html.
It seems quite interesting as it purports to show that CO2’s forcing has no feedback. I.e. a doubling of CO2 leads to at most 1 degree C rise in temperature or that if there is a positive feedback mechanism it must be balanced currently by an existing negative feedback. The negative feedback cannot be due to aerosols, as the IPCC claims, because of Chylek(2007)
These results puts CO2’s forcing well below all existing models and well below any disaster scenario.
Cheers
Steve

aurbo

Re my prior comment:
Charles: You’re right, I should have kept the x and y axes constant leaving any change in the z axis directly proportional to the volume.hat was I thinking?

AnonyMoose

Notice that if you look at the values of the averages for now, they both are 14 million square kilometers. But the current values are quite different. If one value is presently larger, the average should also be larger unless what they’re measuring happens to be currently affected by the differences.

AnonyMoose

Incidentally, in our current glacial averages, “normal” extent might actually be around London or Manhattan. 🙂

drmike86

MarcH & JeffK
I believe the little blip on the chart was identified as a period when they change to a different filter on the satellite. It was on a previous thread.
Mike86

Robert Wood

Aron @14:12:47
Being as the Grauniad is a British rag, that should be “… stick your two fingers up to them”.

KipHansen

Steven,
Have I missed something here?
Has there been some reasonable explanation of the data differences as seen in the blink comparator or not?
Clearly NANSEN and NSIDC are different data sets and shouldn’t be expected to exactly track, but when a public data set changes that radically – someone has some explaining to do.
Has no explanation been forthcoming in the last four months?