Climate Audit

sonicfrog

Steve – just curious. What years do the other lines stand for? The green, blue and yellow lines also has a spike, though slightly later than 2007.

Steve- added.

I agree that the situation will be even clearer in a week, but IMO Connolley is already certain to win his bets.

Steve McIntyre

On June 11, Andrew Revkin reported that 11 of 14 expert teams predicted that this summer’s melt would be at least as extraordinary as last year’s.

Fourteen research teams studying the impacts of warming on the Arctic Ocean have issued independent projections of how the sea ice will behave this summer, and 11 of them foresee an ice retreat at least as extraordinary as last year’s or even more dramatic.

Another busted flush.

Joe Crawford

Steve,
I am not sure how much trust I would put in Dr. Comiso’s algorithm (and code) for processing the data. If you take a close look at days 150 to 155, all six years appear to have the same sine wave type dip/peak in them. While possible, I would think it highly improbable, at least without further investigation and some rational physical explination.

Gary

What’s with that big dip around day 150-153 occurring in most of the recent years?

Tony Brown

I would just like to point out that in my post #9 ‘unprecendted’ is being spelt in the British fashion…

Tony B

Jon

Fourteen research teams studying the impacts of warming on the Arctic Ocean have issued independent projections of how the sea ice will behave this summer, and 11 of them foresee an ice retreat at least as extraordinary as last year’s or even more dramatic.

Of the predictions I looked at, most seemed attributable to qualitative claims about multi-year rather than single year ice. Interestingly I saw very little of ‘estimators’ to project melt trends… but then maybe its my fault for reading press-release style documents.

If knows of such estimators, please post links to actual analysis.

mbabbitt

I guess this is what climate science has come to: one side hoping for the magically symbolic full ice melt to prove their end-of-days apocalyptic point and the other side hoping that the catastrophists don’t get a chance for a propaganda win. I guess it all about the PR show today. I too am on the side of not giving the Catastrophic AGWs this win but isn’t this pathetic?

Steve McIntyre

#14. I don’t disagree with your sentiment. However, now that we’ve got on our eye on the data and can watch the exit polls in near real time, I don’t see any harm in watching; I like watching basketball games and elections as well. Who would have thought that watching paint dry ice melt would attract an audience ?

Jesper

What is the y-axis on the plots (daily dIce? — It’s cut off).

The apparent singularities near day 152 (~June 1)and 182(~July 1) are curious.

The day 152 oscillation looks like it might be an artifact of abruptly switching from May to June climatological values somewhere in the ice computations. If May 31 values are normalized against May averages (~May 15 conditions), the anomalies will be consistently negative, while if June 1 values are subtracted from June means (~June 15 conditions), anomalies would appear consistently high. In combination the two effects would produce a lo-hi oscillation which reverses sign on June 1.

Steve: Added – daily melt (million sq km). I don’t know how much weight can be put on some of these daily oscillations; that’s not the point of the post. The salient issue is that we’re in the heart of the high melt period and if 2008 falls much further behind 2007, I don’t see any chance of it catching up. That doesn’t mean that there isn’t a long term trend, just that we can probably predict this year’s race in a day or two.

mbabbitt

Steve, I wasn’t putting your efforts down in any way. I was just lamenting how sad it is that part of perservering in this ongonging debate is participating in looking at real data exit polls to find the truth — wow, data — not models, fears, and hopes . I highly revere your efforts. I wish I had half of your skills, persistance, and dedication. You should be very proud to be such a voice of reason and an anchor of sanity in this very profound issue. Sometimes I don’t think that you appreciate what an impact you (and some visible others) are having on this AGW issue.

snip – while you are being complimentary to me, you’re discussing policy which I try to avoid.

Jeff C.

Steve, interesting way to plot and review the data. The next ten days should be interesting as the melt rate does seem to peak in this period. Assuming, of course, we have good data. I agree with the other comments that the behavior between days 150 and 155 (or so) looks very suspicious. Based on this plot, the Arctic added ice for the first few days and then lost ice for the next few days at high rates every year from 2002 through 2007. If one of my test engineers brought me data that looked like this, I’d tell them to go remeasure it and be more careful this time.

MrPete

Steve, a way to possibly make the graph be “fake 3d” and more visible:

Try switching from a rainbow of colors to a gradient set:
oldest year: pure gray
newest year: bright red

Don’t know what color definitions R allows, but here’s a set for you:
#808080
#957C7C
#AA7171
#c06060
#D54747
#EA2727
#ff0000

Alan S. Blue

JerseyBoy, “one year ice”, which is most of what’s there, is treated as roughly constant thickness. There’s a separate tracking for ice volume, but (IIRC) it isn’t as ‘live’ because they have to do post-processing.

But the ice melt in 2007 was severe enough that the majority of ice under discussion is single-year ice. Multi-year ice gets compressed and gets to grow (without increasing surface area really) so it is harder to melt.

Steve McIntyre

Here’s a graphic (Also posted in the thread now) that may show my guess a little more clearly. Here I’ve shown a smooth of the average melt (lowess with f=.2) for the average of the period with online daily information at the website linked here, 2007 and 2008. I realize that high melt periods can vary a bit, but sometimes in a pennant race one team builds an insurmountable lead and can run the clock. My guess is that 2007 already has an insurmountable lead over 2008. The vertical red dotted line shows today’s julian day.

Maybe I’m wrong about this; I haven’t studied Arctic sea ice beyond plotting a bit of data. However, I’m pretty sure that my estimate will outperform the estimates of the 11 learned institutions that projected record meltback only 3 weeks ago. And to my knowledge, NONE of them have resiled from their predictions yet.

Steve McIntyre

#27. I’m using daily data available here http://www.ijis.iarc.uaf.edu/seaice/extent/plot.csv which seems to show a sharper divergence from 2007 than NSIDC is showing on your graph.

SteveSadlov

I would love to see definitive proof that there is or ever has been more than a small amount of NH sea ice that is greater than 5 years “old.”

I would also love to have someone point out, with scientific rigor, the difference between so called “multiyear ice” and, ice that is thick due to wind and current driven compressive events (or sequences of such events).

Alan Bates

Re #10 by Tony Brown

“I would just like to point out that in my post #9 ‘unprecendted’ is being spelt in the British fashion…”

Not according to my Concise English Oxford Dictionary you aren’t! Unless you are a Scot or a Welshman with a whole new dictionary!

Alan – and English to the core.

Tony Brown

http://www.peakoil.com/fortopic41356-0-asc-0.html

I think some of you guys need to get over to peak oil-they are getting very excited about the ice melt. We are all doomed so anyone living in London or NY needs to get out within the next three years or reap the consequences…

However, there are some very interesting links to arctic web cams, daily ice comparisons and volcanic activity. Well worth a look if you can stand the angst (Ps don’t mention where you’re from…)

Tony B

Tony Brown

Alan #35

Its a fair cop. (although my grandfather was part Welsh…)

Tony B

pdm

#29 lucia,
did the ice melt or did it just move away from the pole due to wind? The way I read your comment it seems it didn’t actually melt, it just moved.

Tony Brown

James #40

Please post that photo on the peakoil discussion thread given in my post #37 But you might need to move house…

Tony B

Tony Brown

http://www.john-daly.com/polar/arctic.htm

My last post tonight! I was searching for ice data relating to the Skates first emergence at the North Pole in 1959. The link takes you to an excellent article about the subject which looks back at the North Pole over the last 50 years or so. Basically, back then the ice doesn’t seem to have been nearly as thick as certainly I imagined it was, and the amount of clear water even in winter is considerable. The reasons for this are clearly explained. I forgot how good this web site was!

Tony B

Phil.

Re #41

did the ice melt or did it just move away from the pole due to wind? The way I read your comment it seems it didn’t actually melt, it just moved.

It got blown away from the pole through the Fram strait to warmer regions where it melted.

Jon

Re 48:

NSIC shows the same story Steve is showing. A week from now it will be clear whether 2008 breaks down the steep line down that 2007 followed or if it continues to follow the historical trend.

The difference is that NSIC has taken to editorializing without supporting figures and calculations. Of course 2008 “may” turn out similar to 2007, but their statement is so patently PR. I’ll repeat my request for someone to point to a mathematical estimator that will predict a rapid melt.

Phil.

Re #49

NSIC shows the same story Steve is showing. A week from now it will be clear whether 2008 breaks down the steep line down that 2007 followed or if it continues to follow the historical trend.

The difference is that NSIC has taken to editorializing without supporting figures and calculations. Of course 2008 “may” turn out similar to 2007, but their statement is so patently PR. I’ll repeat my request for someone to point to a mathematical estimator that will predict a rapid melt.

To say the above you clearly didn’t read the site.
As for mathematical analysis check out the figure below produced by NSIDC which estimates this year’s minimum based on this year’s sea ice composition and previous year’s melt data:

kuhnkat

Phil,

a reasonable person looking at your graphic will say that 2007 was exceptional and is likely to have little predictive power for 2008!!! 8>)

Of course a REALLY reasonable person would ask you to detail the physical conditions (air temp, albedo, soot, water temp, currents, winds, tsi) which would lead you to believe that 2008 will be a repeat of 2007.

I’m not that reasonable, but, I will ask anyway. Besides funny numbers, whatcha got???

Phil.

Re #53

a reasonable person looking at your graphic will say that 2007 was exceptional and is likely to have little predictive power for 2008!!! 8>)

Of course a REALLY reasonable person would ask you to detail the physical conditions (air temp, albedo, soot, water temp, currents, winds, tsi) which would lead you to believe that 2008 will be a repeat of 2007.

I’m not that reasonable, but, I will ask anyway. Besides funny numbers, whatcha got???

Perhaps, however that reasonable person would realize that that wasn’t the point being made!
It’s not my graphic by the way it’s from the NSIDC as I indicated.
Your reasonable person would also doubtless realize that only one of the previous 25 years would provide a model which would lead to a minimum extent that was greater than 2007. Average (over the last 25 yrs) melt conditions of first year ice would lead to ~0.5 million sq km smaller extent than the 2007 record. Not too difficult really.

Phil.

Re #56

#53. Phil can you give a URL to a discussion of the graphic shown?

Sure Steve, it’s fig 4 about halfway down here

AndyW

It was interesting to note the description on the NSIDC website why the monthly extent can be lower than the daily calculated extent. Looking at their main graph it does seem to be following a more historical pattern than the sharp decline seen last year at this time. I’d still not be prepared to bet my house at the end of this week though, Aug 1 would be my betting timeslot. I have a feeling it will be just higher than last year.

The site quotes various estimates for ice extent being about and below last years, with most about the same. Obviously things are different this year as the ice is melting is different regions from last year, an explanation of that would be good to have if anyone has a link.

I was interested to read that the ice started melting earlier this year and that could be a main factor along with the atmospheric conditions. No mention of SST’s so I still don’t know the weighting of melt for sunlight and water temperature, perhaps the main driver is the solar direct melt.

Nylo

Most of the new ice is tipically in hotter latitudes. That’s the reason why it melts faster. If there was a really different behaviour on the way they two kind of ices melt, we should be observing runaway trends in both directions: a year of multi-year ice destruction should lead to another one, and a year of multi-year ice creation should lead to another one. However one looks at the graph and it is imposible to predict whether there will be more ice or less ice the next year by looking at the predecing ones, and even by looking at the peak ice area extent the previous March, in spite of the general trend, so the quality of the ice must not be very influential. For example, 1980 had more ice in September than the year before, which means that multi-year ice was created. One would have expected again more ice the next year, as multi-year ice survives better than new ice. But that didn’t happen, 1981’s minimum was 0.6M km2 under 1980’s. The same thing can be seen for 1983 and 1984; 1992 and 1993; 1994 and 1995; 2001 and 2002. What decides if the ice melts or not is not as much the age of the ice as the temperature of the water. This year’s melting faster than 2007 is not because of the bad quality of the ice, but because there is ice in lower latitudes, because there is more sea ice area.

Jon

Phil #55,

I’m having trouble seeing a statistical estimator there.

Please see my original request at #12, I’ve read this page (contrary to your suggestions otherwise):

Of the predictions I looked at, most seemed attributable to qualitative claims about multi-year rather than single year ice. Interestingly I saw very little of ‘estimators’ to project melt trends… but then maybe its my fault for reading press-release style documents.

If knows of such estimators, please post links to actual analysis.

lucia

erhmm.. “beat the pros” not “been the pros”. (Some day I’ll learn not to click submit so quickly.)

Steve McIntyre

#67. I think that the difference is even more than that. 2008 is a leap year and “July 2” is one day ahead this year. My comparisons are based on “julian” days and compare July 2, 2008 to July 3, 2007. A 0.25 day interpolation would probably be most appropriate but any consistent treatment is fine.

John L, can you give me an exact URL for your data version for daily sea ice?

kuhnkat

RE: #55

“Your reasonable person would also doubtless realize that only one of the previous 25 years would provide a model which would lead to a minimum extent that was greater than 2007. Average (over the last 25 yrs) melt conditions of first year ice would lead to ~0.5 million sq km smaller extent than the 2007 record. Not too difficult really.”

Actually, NONE of the previous 25 years match the parameters of this year well enough to give reasonable predictive power.

Here is a quote from the story you referenced:

“Lindsay cautions that sea ice conditions are now changing so rapidly that predictions based on relationships developed from the past 50 years of data may no longer apply.”

Jon

Phil, I’ve now found what I was looking for:

Drobot, S.D., 2007: Using remote sensing data to develop seasonal outlooks for Arctic regional sea-ice minimum extent. Remote Sensing of Environment, 111, 136-147,doi:10.1016/j.rse.2007.03.024. [PDF]

AndyW

#65 I think your bets might be a touch high. The spread seems quite consistent apart from the very low values that you have which might mean you have a large quantity of conservatives in your group and a couple of radicals.

Looking at the quickscat picture for today

The area to the north of the geographical north pole (and here I use Webpage North of course, the most important north on the entire planet) seems pretty “wet” compared to dry. In fact we seem to have a small body of water in there, but perhaps more improtantly a very large area of mush. Or is this an observational effect which will disappear in the morn?

With the open waters appearing to the north of Canada expect the North West passage to be open this year with media comments on “first time in the history of mankind” to follow.

Steve McIntyre

I took a look at this data another way, this time calculating the julian day at which each year hit meltback milestones. At the 9.6 million sq km milestone, 2008 is only 4 days behind 2007. Interestingly 2006 was 3 days ahead of 2007 at this milestone, but easily caught 2006 in the back stretch. (2008 is a week behind 2006, but 4 days ahead of 2003. So there may be a little more race left in this than I’ve been arguing.

Remember that only a few weeks ago, we were hearing that the “new ice” was going to be particularly prone to melting. Impressionistically it doesn’t look like this forecasting method is holding up too well.

Ian

This stuff is all fine and dandy but until you can get Vegas to give you an over under the public will remain uninformed.

AndyW

I’ve been comparing the last quickscat montage with the mush to the top of the geo north pole wondering whether it is not really there or not and now I have compared it to 2007 ice melt and it fits in quite nicely

and

I think an overlay would show it better and I am resisting valiently to mention the battle of the bulge ! Of course I’m a complete amateur at this, but it at least it is keeping me occupied at work, cough.

lucia

Jon–

#65 I think your bets might be a touch high. The spread seems quite consistent apart from the very low values that you have which might mean you have a large quantity of conservatives in your group and a couple of radicals.

Part of the reason it looks high is we are trying to predict the cryosphere todays sea ice extent numbers here. Atmoz proposed that data set for purposes of betting.

I don’t know precisely how the seaonal summer ice extent file at the Cryosphere relate to daily Steve is looking at, but I downloaded it, but last years value in the seasonal file is higher than the minimum in the daily file.

If the cryosphere’s page represents the JAS average, that would explain why all our values are higher. We aren’t predicting the minimum, we are predicting the summer average.

Atmoz bet 7.7 million square Kilometers. I don’t think his politics influenced his method, which was use of “ordinary least squares”.

Philip_B

The Brits and the Irish have been betting on the weather for years. Currently the odds of nowhere in Ireland going over 22C in the month of July is 11 to 8.

http://www.freebettingonline.co.uk/Weather-Betting/
http://www.paddypower.com/bet?action=show_event_by_markets&chosen_ev_id=1153252&category=SPECIALS&promo=nov_JulyWeatherSpecials&crea=lnk&novelty=1

roger

All this science makes my brain hurt, but even through the fog of my pain I can recall what happened to an ill-starred ancestor of mine, who led two ships on an expedition to open the Northwest Passage, that expedition having been inspired by sealers’ reports to the Admiralty that “the Arctic is melting”. He got half-way through, was frozen in and perished with his entire crew. Sir John Franklin was his name and a little bit of reading on what prompted his trip should calm a lot of anxious minds. Now, people, back to the sines and math-y stuff that I can’t begin to comprehend!

Richard

This is obviously something fake. No scientist on earth would use MM as a unit. It is only a unit if you are an accountant which is a questionable profession. Does the author mean 10^6 square km? What are the units for the abscissa?

RM

Steve: The abscissa is julian days. As to the y-axis, you win the pseud of the day award.

lucia

Richard– I think someone needs to write a color subtraction algorithm so we can see the differences.

Steve: If someone can read the binary files at NSIDC into ASCII, we should be able to start. I can read the headers, bu I don’t know how to do little-endians and big-endians and rather grudge the necessity to learn this sort of crap to get at climate data. Nicholas will do it when he gets time, but if anyone else knows how, it would help.

BarryW

My guess is that the mantra will change from a “record year” to “arctic ice continues its decline” , ignoring the recovery from 2007.

Smokey

Re comment #4:

After reading Revkin’s article, I made my first ever comment to the NY Times:

“If global warming is occurring, can someone please explain what’s going on here: click

Today I checked back to read any responses to my question. But out of more than 150 comments [I had commented around #55], my question was nowhere to be found.

I suppose letting the general public view that graph violates the NY Times’ Agenda Policy.

Phil.

Re #90

No need to do that lucia & Steve just save them as jpgs and process them using photoshop.

AndyW

#93 Good point about the 2006 and 2007 hurricane seasons.

You might be surprised to find out that even without an atlantic hurricane yet this year 2008 has already beaten one record! Something along the lines of the first tropical storm to form in early July the most far east I think from recollection. It’s amazing how many records there are to be broken.

So 2008 is already a record breaking hurricane season. Even though nothing much has happened yet.

I still think the Artic might go way down, I think we might see some large daily loses in the next 2 weeks. Not sure if it will broach 2007 though.

Steve McIntyre

#97. here’s a file – the first 23 entries are headers and I can read them OK but I can’t get any of the data in a form that I know how to use.

ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/nasateam/near-real-time/north/nt_20080702_f15_nrt_n.bin

I don’t know why they wouldn’t have an ASCII file, but I guess all these guys are used to Fortran on Unix machines, converting little jobs into big jobs.

Real Richard Sharpe

Steve, can you give me the fragment you already have for downloading the file … the format of that file looks confusing but it does have structure … here is some portions listed by hexl-mode in Emacs:

00000120: 2030 372f 3034 2f32 3030 3800 0000 0000 07/04/2008…..
00000130: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
00000140: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
00000150: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
00000160: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
00000170: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
00000180: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
00000190: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
000001a0: 0000 0000 0000 0000 0000 fdfe fefe fefe …………….
000001b0: fefd 0000 0000 0000 0000 0000 0000 0000 …………….
000001c0: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
000001d0: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
000001e0: 0000 0000 fdfe fefe fefe fefe fefd 0000 …………….
000001f0: 0000 0000 0000 0000 0000 0000 0000 0000 …………….
00000200: 0000 0000 0000 0000 00fd fefe fefe fefe …………….
00000210: fefe fefe fefe fefe fefe fefe fefe fefe …………….
00000220: fefe fefe fefe fefe fefe fefe fefe fefe …………….
00000230: fefe fefe fefe fefe fefe fefe fefe fefe …………….
00000240: fefe fefe fefe fefe fefe fefe fefe fefe …………….

Steve McIntyre

url=”ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/nasateam/near-real-time/north/nt_20080702_f15_nrt_n.bin”
download.file(url,”temp.bin”,mode=”wb”)
handle=file(“temp.bin”,”rb”)
x=readBin(handle, character(),23);x
x
[1] “00255”
[2] ” 304″
[3] ” 448″
[4] “1.799”
[5] “39.43”
[6] “45.00”
[7] “558.4”
[8] “154.0”
[9] “234.0”
[10] ” SSMI”
[11] “13 cn”
[12] ” 184″
[13] “-9999”
[14] “-9999″
[15] ” 184″
[16] “-9999”
[17] “-9999″
[18] ” 2008″
[19] ” 184″
[20] ” 000″
[21] “000250nt_20080702_f15_nrt_n”
[22] ” ARCTIC SSMI TOTAL ICE CONCENTRATION DMSP F13 DAY 184 07/02/2008″
[23] “ARCTIC SSM/IONSSMIGRID CON Coast253Pole251Land254 07/04/2008”

These can all be interpreted according to documentation at http://nsidc.org/data/docs/daac/nsidc0081_ssmi_nrt_seaice.gd.html

Steve McIntyre

Also there’s software here which may or may not be helpful
http://nsidc.org/data/docs/daac/nsidc0081_ssmi_nrt_seaice.gd.html#software

Steve McIntyre

#107. I don’t doubt that it reads, it’s just that I don’t know how to read it in R. Can you post up your script??

Dishman

Agghh.. trying again

header : readBin(handle, “int”, 1*304, size=1, signed = FALSE, endian = “big”)
Data : readBin(handle, “int”, 448*304, size=1, signed = FALSE, endian = “big”)

Steve: Excellent progress. Now we have 448*304 numbers from 0 to 250, representing sea ice in various gridcells.

If one does the following:

Data=array(Data,dim=c(304,448))
contour(Data)

one gets something that looks like the Arctic, so we’re definitely getting pretty close. Thanks, Dishman. Further thoughts will be much appreciated.

Dishman

The file header calls out specific values that are not ice as well (as noted above).
You’ll want to exclude those from your processing.

It looks (to my untrained eye) very much like the right graph in #89.

Raw text header is:
00255 304 4481.79939.4345.00558.4154.0234.0 SSMI13 cn 184-9999-9999 184-9999-9999 2008 184 000000250nt_20080702_f15_nrt_n ARCTIC SSMI TOTAL ICE CONCENTRATION DMSP F13 DAY 184 07/02/2008ARCTIC SSM/IONSSMIGRID CON Coast253Pole251Land254 07/04/2008

Real Richard Sharpe

OK, I have got it to work as well, had to replace some double quotes with single quotes:

url=’ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/nasateam/near-real-time/north/nt_20080702_f15_nrt_n.bin’
download.file(url,’temp.bin’,mode=’wb’)
handle=file(‘temp.bin’,’rb’)
header=readBin(handle, ‘int’, 1*304, size=1, signed = FALSE)
data=readBin(handle, ‘int’, 448*304, size=1, signed = FALSE)
data=array(data,dim=c(304,448))
contour(data)

I think the next step is to associate colors with different values in there …

Real Richard Sharpe

filled.contour(data) does interesting things for me.

Now I want to rotate the data to the more usual orientation and make the ice stand out more …

Steve McIntyre

I agree about the quotes in WordPress. They are really annoying. You can get a “usual” orientation by the following:

Data=Data[,448:1]

and then plot. One of their software programs is a “mask” for the ocean, so there should be a tool to screen out the land area.

Real Richard Sharpe

OK, this (from the MatLab program) is useful …

%Find locations fo Land, coast, hole-at-the-pole, and missing data
land = find(map1 == 254);
coast = find(map1 == 253);
hole = find(map1 == 251);
missing1 = find(map1 == 255);

Also, I subtracted Jul3 from Jul2 data and plotted it and came up with areas where ice was lost, but it seems there were areas where ice was gained.

Steve: I’ll experiment with that tomorrow. In my plot, I handled gt 250 as a group.

Steve McIntyre

I suspect that algorithm for going from the 0-250 scale to sea ice concentration area isn’t that complicated. It would be nice to be able to do it ourselves and suggestions would be welcomed.

Steve McIntyre

tim.colors in the fields package are colors that I like. Fixed the 300-byte header but didn’t bother re-uploading since it’s slight.

Thus:

library(fields)
Data=list()
loc=”ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/nasateam/near-real-time/north”
mm=”07″;dd=”03″
#mm=”03″;dd=”31″
url=file.path(loc,paste(“nt_2008″,mm,dd,”_f15_nrt_n.bin”,sep=””) )
download.file(url,”temp.bin”,mode=”wb”)
handle=file(“temp.bin”,”rb”)
header= readBin(handle, “int”, 1*300, size=1, signed = FALSE, endian = “big”)
X = readBin(handle, “int”, 448*304, size=1, signed = FALSE, endian = “big”)
close(handle)
Data[[paste(“2008″,mm,dd,sep=””)]]=array(X,dim=c(304,448))

mm=”07″;dd=”03″
breaks0=c(seq(0,250,5),251:255); M=length(breaks0)
layout(1)
par(mar=c(3,3,2,1))
mm=”07″;dd=”03″
image.plot (Data[[paste(“2008″,mm,dd,sep=””)]][,448:1], col=c(tim.colors(M-5),rep(“green4″,4)),breaks=breaks0,xlab=””,ylab=””)
title(paste(“Sea Ice: 2008″,mm,dd,sep=”-“))

The coloring needs to be tweaked a bit to match more closely.

Tony Brown

Apparently 2007 was 16% sunnier than ”average’. Therefore I’m going to go for the 2008 melt being around 16% less than 2007 (in the absence of being able to find any arctic wide sun and wind records for this summer)

TonyB

Steve McIntyre

I think that I can calculate daily sea ice area in their style – at least I get something that seems to correspond. The pixels are said to be 25 x 25 km resolution and the 0-250 scale is said to be 2.5 times concentration. Thus, simply sum the values over 0.15 (after re-normalizing) and multiply by 25*25. I got 9.69 million for July 3 and 9.60 million for July 4, which seem in the right ballpark, though they don’t match anything in particular just yet.

X=Data[[paste(“2008″,mm,dd,sep=””)]]
temp=(X>250) #land 254; coast 253; hole 251; missing 255
temp1=(X==251)
X[temp]=NA; X[!temp]=X[!temp]/2.5
25*25*sum(X>=.15,na.rm=T)/1E6 #pixels have 25 km resolution
# 9.690625 July 3
# 9.600625 July 4

Phil.

Re #129
I don’t think there’s a really good comparison year, the major factor this year seems to me to be the major breakup of the ice pack in the Beaufort sea this winter. This has led to a major melt there already, associated with this has been a shift of the transpolar drift much closer than usual to the N american coastline (and a substantial reduction in the Beaufort gyre). This means that there is strong flow of multiyear ice out of the Fram (as there was all winter).
buoys
79200 is the location of the N Pole webcam

multiyr ice

This is also consistent with the E Siberian & Laptev Sea since the flow is further away this year.

Phil.

Steve, as I understand it they both use AMSR-E data but use the two different servers to get more consistent data flow.

Phil.

Steve you could contact Georg Heygster (heygster@uni-bremen.de)
There’s also a thesis describing their algorithms etc. (in German) http://iup.physik.uni-bremen.de/iuppage/psa/2001/amsrop.html

Phil.

Re #133

The scale will depend on the projection if it’s azimuthal equidistant the scale factor is (pi/2-lat)/cos(lat) but that doesn’t seem large enough.
azimuthal equal area would be cos(pi/2- lat/2). If you can find out the projection it should be straight forward. I seem to recall that Hu is into this?

Steve McIntyre

Day 186 melt results had been revised down (less melt) when I checked just now; see note in prior post.

nevket240

http://environment.newscientist.com/article/dn14251-greenland-ice-sheet-slams-the-brakes-on.html?DCMP=ILC-hmts&nsref=news7_head_dn14251

How come this hasn’t made the headlines in the media??
Please move if in wrong section.

regards.

OldUnixHead

A couple of things:

A) RE S McI #128, 133

Given the chain of documentation discussions initiated via the NSIDC “Near Real-Time DMSP SSM/I Daily Polar Gridded Sea Ice Concentrations”
[ http://nsidc.org/data/docs/daac/nsidc0081_ssmi_nrt_seaice.gd.html ]
page, its link for “Polar Stereographic Projections and Grids”
[ http://nsidc.org/data/grids/ps_grid.html ], and that page;s link for
“DMSP SSM/I Daily Polar Gridded Brightness Temperatures”
[ http://nsidc.org/data/nsidc-0001.html ], which links to a FAQ page
[ http://nsidc.org/data/docs/faqs/ssmi_faq.html ],
I get the following impressions:

– 1) the stereo-projection info in the files you are reviewing is for a projection tangent to the earth-surface at 70 degrees latitude, rather than at 90, to minimize transformation errors of the data content at the actual surface (in the neighborhood of the circumpolar land edges); and

– 2) the text/binary data files you are trying to unwind are likely to be in the NASA-std HDF format (possibly with HDF-style metadata either contained in the files themselves or contained in separate metadata files, or both); and

– 3) there is a collection of libraries (for C and/or Fortran interfacing and for Java interfacing) and utilities, all for dealing with said HDF files (apparently including projection vs lat-long transformations and mixed text/binary data formats) –
see “the HDF Group” page [ http://hdf.ncsa.uiuc.edu/products/hdf4_tools/ ].

Sorry if the above are stating things you’ve already discovered, but trying to decode the general HDF format may be useful for a general-purpose R package, given NASA;s apparent acceptance of this format style for a chunk of its data archives, rather than doing one-offs for each of your various analyses.

It sounds like a non-trivial exerise, especially in dealing with the Polar Stereographic Projections. I don’t pretend to understand the implications of all of it, but it may give someone else a tickle to take it further. It certaily gave my mind a spin.

B) RE S McI #121: Per your observation about Lake Erie in the map graphic (and Lakes Ontario and Baikal, too):

Do these non-Polar areas in the graphics get ignored via the map-mask files when used by NSIDC internal analyses?

Regardless, does the fact that the files show real sensor “brightness” data values for those areas say anything about how we should interpret comparable data in the “real sea ice regions” (are there one or more physical biases that need to be removed) before discussing areal melting, or are the current results “good enough”?

Steve: The binary data is decoded. I don’t have the projection exactly right yet, but I’m pretty close. I’m puzzled at going from the grid ice data set to reported daily aggregates though and any insight on this wopuld be appreciated:

jae

Interesting info. on volcanos under the ice here.

Bob KC

Steve,

If you’ll look at the first sentence under the “Notes on the Data” section of this page, I think you’ll see why the numbers don’t quite add up.

John Goetz

I thought an algorism was a statement along the lines of “During my service in the United States Congress, I took the initiative in creating the Internet.” I did not know until reading that link it also referred to certain types of climatological estimations.

DeWitt Payne

Was anyone credible actually predicting complete melt this year? My impression was that complete Arctic melt was predicted by 2030 at the latest, but not this year.

Gerald Machnee

Re #150 – This year? Not quite but close. See the following by the Green Party:
http://www.greenparty.ca/en/node/3196

Arctic Ice May Melt by 2010: Scientists Scared
Submitted by Jim Harris on 16 November 2007 – 12:28pm.

The Arctic Ocean ice may completely melt by 2010 — something that hasn’t happened for more than a million years, according to Louis Fortier, scientific director of ArcticNet, a leading polar researcher.

Tom in Texas

Thanks Steve for keeping us updated with the colorful analogies.

Steve McIntyre

155. It’s hard to know what to make of the new ice- multiyear ice method right now as the calculation remains opaque. If, despite the prevalence of new ice, the meltback is not as severe as last year, then maybe the forcing is that much less than anticipated – since the new ice would otherwise have melted. 2007 continued to gain against 2006 in fairly late stages; my own hunch, and it’s little more than that, is that for a big meltback year, we’d have already had some very big melt days, which we haven’t.

markinaustin

has anyone considered the link with micro particles of coal in the ice? it seems to me that this was mentioned as a major factor in heating and that if much of the ice melted last year, this would have dissolved into the ocean…..so that this year the ice will be cleaner and more reflective.

John Lang

Just looking at some of the MODIS satellite images from the past few days. While the sea ice extent is quite a bit larger than last year and the Northwest Passage is still frozen solid, the sea ice within the Arctic Circle is very broken up. There are almost no completely solid sections even at the North Pole itself.

This 500M resolution MODIS image from today shows a great deal of broken ice and melt (note there is a lot of cloud cover today as well.)

LIA

Steve

The total cover for yesterday is now stated as 9.094063. I remebr looking a couple of hrs ago and I believe at that point, the figure was what you have put up above. With the current figure, 2007 seems to win out for 7/7 as well. Do you know why the revision seems to have taken place?

John Lang

#161 from Steve. IS that “unusual”?

I’ve been following the MODIS pics for awhile. I don’t think I’ve seen the ice inside the Circle so broken at this time of the year.

Here is the same 500M resolution from July 8th, 2006 (2007 was cloud covered for several days at this time of the year) and the ice is definitely more solid.

John Lang

To UK John, of course this is just personal anecdote, but it is does seem there is much less cloud cover this year than last. I remember being very frustrated with the MODIS satellite images last year since there always seemed to be cloud cover obscurring what I wanted to review. This year, however, one could have taken a full complete picture of the Northwest Passage and the majority of the Arctic just about every day.

We do know that low cloud cover has two impacts – first, water vapour is a very good greenhouse gas and hence leads to a warmer surface – secondly, clouds intercept sunlight from reaching the surface and hence leads to cooling.

In middle and lower latitudes, cloud cover results in a general cooling with daytime temperatures declining by more than nightime temperatures increase and the net result is cooling. Perhaps in the higher latitudes, 24 hours of sunshine combined with cloud cover actually nets out to increased surface temperatures and that is why there was such a record melt last year and less melt this year.

I haven’t seen any empirical data on this except that cloud cover has certainly been increasing in the Arctic recently, but there is no data for this year versus last year.

Steve McIntyre

Day 190 Race Report

2008 lost a little ground to both 2007 and 2006 in day 190 with another day off the clock. About 630,000 sq km behind, which would be a seemingly insurmountable 10 days at present melt rates. Plus 2007 is going into a pretty strong week – with 4 straight plus 100,000 days coming up (of which there have been none so far in 2008.) My guess (And it’s only a guess) is that 2008 will not be able to gain any ground in the next 4 days – more likely to lose some ground. It might even have to look over its shoulder at 2003, which is closer to 2008 than 2008 is to 2007.

month day year ice diff
7 9 2002 9.454688 -0.047500
7 9 2003 9.425938 -0.064218
7 8 2004 9.596406 -0.059063
7 9 2005 8.847813 -0.057656
7 9 2006 8.456719 -0.101875
7 9 2007 8.369063 -0.085937
7 8 2008 9.032500 -0.061563

Restated:
7 8 2008 9.045469 -0.048594

TAC

Rich (#169)
A further question is why, if Steve is often lamenting pressures on his time, is he working on daily updates of this…[?]

CA is a quirky place, and many threads wander into unexpected topics. I find them highly entertaining and enlightening; I learn a lot from CA.

Sure, It can seem fanciful to discuss the “horse race” between 2007 and 2008 sea ice, or the connection between “Climate Scientists and the Riemann Hypothesis,” or Steve’s “Bull Dogs Have Little Dogs” doggeral; or his fascination with FOIA requests; or his many ingenious ways to drive an additional stake through the heart of the many-times-dead MBH98.

Yet it all makes sense once you get used to it. The edifice of Climate Science has an “Alice-in-Wonderland” quality; it demands to be respected, understood and appreciated on its own terms.

Sure, purists object when words like “rigorous” and “conservative” are turned on their heads, or when Statistics is distorted to provide whatever answer an author needs to support a current hypothesis, or when authors of refuted papers behave like Monte Python’s Black Knight, denying their wounds and boldly declaring: “Come back and fight, coward!”.

And I know that some people get angry and jump up-and-down and whine about the pathetic state of the world. But why do that when you can take a creative and pleasant tack? Steve chooses to recognize and appreciate the humor — I love Monte Python, too — and plays along with it.

It’s fun; he enjoys it; I enjoy it.

It is, after all, only Climate Science — not real like, say, the Riemann Hypothesis.

😉

Luis Dias

John Lang’s images are quite frightening. It almost seems as though it will eventually crack and the ice will just fade away instantly. I can’t wait until September so we can see what really has happened.

Joshua Matthew

The trouble with numbers Steve is they only say so much – and when we trade statistics blithely we sometimes don’t stop to look at the real world.

Compare 2008 false coloured sea ice maps with 2007 for the same julian day – the Beaufort sea area is primed to plummet

Secondly Hudson Bay is 170,000 behind last year – but that will all be made up as the Hudson Bay ice disappears.

Finally the Atmosphere is set to do its part. A deep low is forecast to persist adjacent to the Fram Strait Thursday, Friday and Saturday, pumping out lots of ice through the gap in a strong N’ly flow. Meanwhile a large high holding the promise of near maximum insolation persists over the Beaufort Sea … primed ready to melt. Later in the period strong winds strong winds set up blowing from Russia to Canada, compacting the sea ice extent

Two furlongs to go and 2008 is well back from the lead … but expect a strong late run down the outside. I’d be placing an each way bet.

Steve: Good points. 2007 made up a LOT of grond on 2006 in late July and August so there may be a lot of race left. Maybe even a mountain stage.

Joshua Matthew

Hmm … did I say three? – I meant two.

Jedwards

Phil @ #177

Realize that the length of the tail on those charts represents a 60 day drift track, so in the next 60 days you can expect those buoys to move approximately the same distance again. The Ocean Buoys map shows a slightly wider view. Basically, only the buoys wich have already entered the Fram Strait between Spitsbergen and Greenland even stand a chance of clearing the southern end between Greenland and Iceland before refreezing starts setting in come September.

UK John

John Lang 166

Thanks for that, one thing I just cannot get my head round is that I have observed that ice in direct sunlight disappears more quickly than ice in the shade.

So I don’t get how an increase cloud cover could ever increase a loss of ice, but I could see how less clouds would cause increased melt, so I was just wondering which of this year and last year had more sunshine hours.

Jedwards

Re #179 Agree that the flow does not stop, however during refreezing, the flow slows down markedly. But then, this is all part of the natural cycle of Arctic Ice formation and destruction. Unless the ice is caught up in the Beaufort Gyre, it inevitably gets flushed out via Transpolar Drift via the Fram Strait. The key to events like what occured in 2007 is whether is whether a larger than normal ammount is flushed into the N Atlantic. The flow rates as represented by the buoy data that you linked seem to indicate that though we may see higher than average ice melt, we will likely not see anywhere near the levels achieved in 2007.

See - owe to Rich

Re #181 crosspatch

Though I’m no expert, this is a blog, so I’m allowed to disagree – with the assertion that cloud can mean more melting. It is my belief, especially given 24-hour Arctic days, that it is the radiative power of the Sun which causes more melting, rather than the air temperature. That is certainly the case with snow on my lawn anyway (not that I’ve got any now, it’s a cool July but not that cool).

I see no-one has yet answered my question over why it is preferable to study ice extent rather than ice area.

Rich.

Phil.

Re #184

It seems to me that both are used, the sat. maps all show concentration, the question then is how to represent by a single number. NSIDC & JAXA, use extent whereas CT used area. My guess is that when extent is used it’s possible to compare with earlier non-satellite data whereas that’s not really possible with area (see CT for instance). If you have access to the data, as Steve has for example, it’s possible to produce either, they don’t always tell the same story, in the antarctic last year one was a record the other wasn’t (but it was close).

Steve McIntyre

Day 191 Race Report
I mentioned yesterday that 2007 was entering some active stages and that 2008 would have to play with a lot more urgency to keep up than it had being. 2008 lost about 64,000 sq kn to 2007 and is now 740,000 sq km behind (all figures IARC-JAXA daily extent http://www.ijis.iarc.uaf.edu/cgi-bin/seaice-monitor.cgi)

month day year ice diff
7 10 2002 9.382344 -0.072344
7 10 2003 9.362500 -0.063438
7 9 2004 9.534063 -0.062343
7 10 2005 8.768750 -0.079063
7 10 2006 8.359688 -0.097031
7 10 2007 8.233906 -0.135157
7 9 2008 8.973594 -0.071875

AndyW

#188

And as you mentioned befre 10-13th 2007 had some pretty big losses so 2008 will have to pull up it’s socks!

It will be interesting to to read the analysis in early August from NSIDC on why the change back to a standard loss and no longer any chance of a record, or even near record melt this year. I guess it won’t reach the media though.

Phil.

Re #189

Doesn’t the datafile you’re reading from JAXA have the ice concentration in each square since they plot that in their maps?

http://www.ijis.iarc.uaf.edu/seaice/data/200807/P1AME20080709IC0.png” alt=”Today’s ice concentration”

DeWitt Payne

Blue and black aren’t exactly contrasting colors. For that matter, red is only slightly better. Can you replot with, say, red, blue and green? Or put marks on the 2008 curve so it’s easier to tell which path it’s following.

Phil.

Re #195

In the view from cam #2 the posts are starting to collapse!

jeez

Phil, repeat after me.

Download image.
Resize to approx. 600-800 pixel width.
Use bicubic scaling for highest quality if given an option.
Upload resized image to imageshack.us
Link to image.

Phil.

Re #199

That appears to be 2004, I believe that they replace the cameras every spring, the current position is at 84.375°N 2.084°E.

STAFFAN LINDSTROEM

196 … phil….phil…phil…weak…weak…weak
Borrowing from Tony Blair versus Major April 1995 ,
What about the ice wall in the background, how high
is it, would you say?? Or am I seeing ice desert mirages???
…Am I referring to the collapsing North Pole Poles or to
your argument or both …Your choice…BTW was just over
to Wetterzentrale…The models for 15-17 July show subzero
temps for much of the Arctic basin…BTW2 In this very
province, Uppland, last night July 9 2008, Kerstinbo, a small
village some 200 KM NNW of Stockholm (half of our capital is
in Uppland…) RECORDED Sweden’s lowest temp: +0.3C…
If that weather station is in the village and not outside
we N.B. MAY get A Road Management road weather station reading of -0.5C!?
So where is infamous CO2 blanket when you need it??…

Joshua Matthew

One hypothesis is that warmer temperatures lead to the greatest change in sea ice extent (June / July / August). I used the data set Steve has nominated (sea ice EXTENT) – and took a moving average of 3 days (to smooth out data artefacts). I then selected the largest and second largest melt period for each year (separated by at least 10 days)

The mid point dates of these large melts are: 15/07/0227, 02/07/2006, 29/07/2005, 14/07/2004, 19/07/2003, 07/07/2002 and 26/07/2007, 05/08/2006, 26/06/2005, 13/08/2004, 21/06/2003 and 17/07/2002.

The surface temperature anomaly for these days shows cool to neutral conditions in the Beaufort, with mildly warm conditions (2 to 3 degrees) near the surrounding coasts. Amongst other things, this could mean:

a) large sea ice EXTENT changes are not strongly temperature related,
b) changes are temperature related, but strongly warm areas are somewhat randomly distributed, and partly cancelled by strongly cool areas.

I’m sure this approach does not qualify for the term rigorous.

In the next post I’ll look at another possible player on sea ice EXTENT: wind.

Joshua Matthew

Hmmm .. my long post was blocked .. the very short version here shows no temperature anomlaies during 2007’s major change in isea ice extent ..

tty

Re 205 The melt ponds *are* sitting on thick ice otherwise they would leak through. Ultimately they will do so, and incidentally some of the fresh water refreezes and “underplates” the ice since the salt water under the ice is typically below zero centigrade. This puzzled Fritjof Nansen no end on the Fram expedition until he figured out what was going on. The snow and surface ice was melting merrily and the ice was growing thicker and thicker! Incidentally this is a factor that is rarely if ever mentioned in connection with sea-ice melting. If the water under the ice is warmer than usual (over or very close to zero) this “underplating” won’t happen and the ice will melt much faster.

Jedwards

Interesting data point courtesy of Steve Milloy;

German Scientists Predict 2008 Arctic Ice Melt to be GREATER than 2005, but less than 2007

The ice cover in the Arctic Ocean at the end of summer 2008 will lie, with almost 100 per cent probability, below that of the year 2005 – the year with the second lowest sea ice extent ever measured. Chances of an equally low value as in the extreme conditions of the year 2007 lie around eight per cent. Climate scientists from the Alfred Wegener Institute for Polar and Marine Research in the Helmholtz Association come to this conclusion in a recent model calculation.

Based on the numbers Steve is showing us here, even this “projection”, which has been generated by “models” looks to be at risk.

DeWitt Payne

Out of curiosity, does anyone know how often Cryophere Today updates their charts? It seems like it’s been way more than a week with no apparent change.

See - owe to Rich

212 DeWitt – I don’t know for sure, but I believe it’s fairly frequently (at least twice a week), but since they had some awful glitches both last year and February this year, they only publish about a week after the event.

Interestingly, the CT NH area anomaly has crept up again by about 1/3M km^2, and is now (i.e. Jul 3 approx) about 2/3M above last year’s. That is a very similar amount to that reported by Steve for ice extent at #52 on the 2nd.

Rich.

An Inquirer

Rich, Refering to the picture in POST #196, you said that “the big white ridge in the distance (it doesn’t look very big because of the distance). . . is not a large ice ridge, but the impressive surf at the edge of the ice cap!”

Question: How do you know that?

STAFFAN LINDSTROEM

#209 Rich … aha the Beach Boys effect: Too many icebreakers
with zillions of surfers in the Arctic … Seriously, I found
a paper of Yang et al that reported of 25-90 % more runoff
from Siberian rivers in winter caused by increased snowfall,
so the Arctic has washed out the baby with the bathing water,
so to speak…That was of course blamed on GW (not necessarily
AGW IF I recall correctly) I’m surprised the Arctic summer ice
did not leave earlier…What does IPCC say about this? We have
to check!!

Phil.

Re #208

Re 205 The melt ponds *are* sitting on thick ice otherwise they would leak through. Ultimately they will do so, and incidentally some of the fresh water refreezes and “underplates” the ice since the salt water under the ice is typically below zero centigrade. This puzzled Fritjof Nansen no end on the Fram expedition until he figured out what was going on. The snow and surface ice was melting merrily and the ice was growing thicker and thicker! Incidentally this is a factor that is rarely if ever mentioned in connection with sea-ice melting. If the water under the ice is warmer than usual (over or very close to zero) this “underplating” won’t happen and the ice will melt much faster.

The ice in this location is about 190 cm thick and has started to thin (it’s first year ice).

Steve McIntyre

Day 192 Race Report
Another day in which 2008 showed no urgency whatever in its efforts to catch 2007. It fell behind another 37,000 sq km and is now nearly 790,000 sq km behind (over 10 days at current rates). It might be worth watching the race with 2002, which made up 50,000 sq km on 2008 yesterday.

month day year ice diff
7 11 2002 9.262344 -0.120000
7 11 2003 9.302188 -0.060312
7 10 2004 9.458125 -0.075938
7 11 2005 8.697500 -0.071250
7 11 2006 8.313594 -0.046094
7 11 2007 8.125156 -0.108750
7 10 2008 8.916563 -0.071562

Revised: 7 10 2008 8.917969 -0.070156

Tom in Texas

If 2008 beats 2002 in total ice remaining, doesn’t that mean we’re cooling rather quickly?

AndyW

#196 Phil you missed a classic instance to state, quite correctly, that due to ice melt the poles are tilting this year! That news would have gone around the world in a flash. but you have missed you 5 minites of fame by referring to them as posts.

Looking at Steve’s graphs 2008 seems to have missed out on the large spike that several years have exhibited in past years around 180-185, so it is no wonder is it wandering up to the average line currently. Looking at the Quikscat images it looks like there is a potential chance of a NE passage route this yar though around the top of Russia. I believe someone tried this last year and failed.

tty

Re 226 The northeast passage is passable (and used) every year in contrast to the northwest passage. However this year there is exceptionally much ice along the Siberian coast and it may never be possible to get through without icebreaker assistance.

UK John

Still don’t get any of this.

The whole point of all of this is looking for the “accepted” AGW signal trend in the sea ice extent/area statistics, shouldn’t we all be looking for a trend in the Winter sea ice extent numbers.

In the Winter there can be no influence from solar as there isn’t any sunshine! Looking at Summer melt will not tell you much about AGW, might tell you a bit about cloud cover and solar but that is all, the AGW signal will be overwhelmed by other natural effects.

Have we all gone mad, or is my simple logic wrong?

Joshua Matthew

A post in three parts … following up my previous attempts to explain large EXTENT changes ..

Using the large and second largest melts – I looked at wind impacts. I synoptically typed the 12 periods into HBLR, LBHR, HGLR and H patterns. (Apologies for the shorthand pattern names – their meaning is irrelvant)

Joshua Matthew

There were very strong vectory wind anomlaies for HBLR, LBHR, HGLR, and H patterns.

STAFFAN LINDSTROEM

228 tty … According to unknown author at Weather
Underground the NE passage was just open for ordinary
ships for 2 days 2007, Sept 25 and 26, not so good
behaviour of the ice commercially-wise…BTW see
and DL Tony “weak…weak…weak” Blair on You Tube…
Priceless …not iceless…Ha gotcha!!; and TB was paraphrasing
me really (1993…Oct, meeting Mark Levengood when delivering
morning papers in Stockholm…12:30… Saturday…)ML is also
viewable at YouTube!! I’ll be the last person to get there!!??
I do have an account, though…

AndyW

#237 .. you are perhaps right. SST’s anomalies are going down as well to about average. Quikscat pictures are showing more breakup so perhaps wind effects are now causing this.

I still think it will be interesting to watch.

SteveSadlov

Up until a couple years ago, I thought Cryosphere Today was respectable. Now, I consider it a complete laughing stock.

Tony Brown

Post 229

Nice one Rich! What with senoj mot on another thread and the literary efforts on The Bulldog thread its starting to feel end of term!

Here’s a challenge. We all knew you were talking nonsense. However, suppose you were to write that up seriously? Put in that the first ever attempt at surfing is to be made at the NP due to unprecedented global warming. Name a date in late September. Add in some plausible names. Incorporate one of the photos I posted. Send off the material to a heavyweight alarmist newspaper

Such is the degree of hysteria I’d lay a small bet it would get published THEN cited! How about it as an exercise in gullibility?

Tony Brown

John Lang

Great pic of the Northwest Passage from a few hours ago. It is going to be at least 4 weeks before anyone can claim the NW Passage has opened this year.

BarryW

Re 241

They don’t seem to provide the recent raw data that I can find. The charts don’t seem be be updated either for July.

TAC

Preliminary Unofficial Day 194 Race Report
Same as Day 193: 2008 is falling still further behind 2007, though it seems to have made up a little ground on 2006 and 2005.
7 13 2002 9.091406 -0.048438
7 13 2003 9.129531 -0.104844
7 12 2004 9.321094 -0.072656
7 13 2005 8.597969 -0.049062
7 13 2006 8.205000 -0.075313
7 13 2007 7.881250 -0.133906
7 12 2008 8.779375 -0.087344

Revised: 7 12 2008 8.780000 2008-07-12 14072 200807 194 -0.08671900

David Smith

I use this homemade plot to estimate the effect of insolation on melting:

The idea is that there is incoming sunlight which is partially offset by IR loss to space. I arbitrarily use 180 w/m2 for the IR loss. Stated differently, when insolation falls to about 180 w/m2 (the insolation around 3 September) the ice melt ends or at least approaches zero.

What this shows is that the net radiation remains high but we’re “over the hump” and accelerating downwards.

Jim Arndt

#250

Nice bell curve but I don’t think nature is that predictable.

Hans Erren

re 252:
Thats not a bell curve that’s astronomy
Try this excel sheet:

http://home.casema.nl/errenwijlens/co2/insol.zip

Patrick Henry

#255 – Hudson Bay ice always disappears by the end of July, and is close to normal now. The point I was making is that temperatures near the pole are forecast to be below freezing for the rest of the month.

http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.13.html
http://wxmaps.org/pix/temp2.html

David Smith

Re #251 Jim, to me the big unknown is how first-year ice behaves in August. It is conceivable that much of this newer ice has, so far, thinned but not melted. Come August it may thin enough to melt on a large scale. This would produce an unusual seasonal melt pattern, heavily loaded on the back end and susceptible to unpredictable late-season factors like cloud cover and wind. So, I have no real guess.

The most interesting thing to me in all of this is that the impact of consecutive years of reduced summer ice appears to be overwhelmed by the effects of wind, current and cloud cover trends.

Jim Arndt

David #258,

Yes, I see your point that is why I put out the Hail-Mary at 2003.Who knows no currents or wind to push out the ice and then we have a much reduced melt off.

Eggplant fan

An AP story about a Russian base on an Arctic ice flow needing to be withdrawn early “because of global warming” is getting a lot of press, for example at http://news.yahoo.com/s/ap/20080714/ap_on_re_as/russia_polar_station . It also says that the floe drifted 1550 miles west, but doesn’t indicate if that is unusual and might have led to the earlier than expected melt. Can that be determined from the data being examined here? According to the story it started in early September (2007 I assume) near the Severnaya Zemlya archipelago.

Tony Brown

#263

I picked this report up a few days ago from the BBC and emailed the head of the Russian Arctic and Antarctic Scientific research institute asking for a clarification but have had no reply.

From the Russian news reports this was a limited experiment that was intended to end in July anyway and the ships that picked them up had been pre booked when the experiment started last August/September. Whether it ended normally, a few days early due to ice melt, or that they had drifted a long way into more open waters is not clear from the reports.

The following link leads to the official Russian research document confirming the original short lived nature of the experiment

Click to access arctic_AARI_activities.pdf

This second link is the original BBC report-their climate science tends to be biased. This ice station being launched in that geographic position (see map in report)is all very well at the beginning of the Arctic winter but it looks rather far south for this stage of the summer
http://news.bbc.co.uk/2/hi/science/nature/7503060.stm

This later AP report tends to suggest the rescue was because of global warming but then spoilt it by mentioning a nuclear powered ice breaker had to clear the way for the rescue http://ap.google.com/article/ALeqM5jifNfzf0IHoEvLMAb6sHLjcphdVwD91TJDTG5

Are there any experts here who can confirm if the ice floe drifted too far?

Tony Brown.

Eggplant fan

Re #264

So the earlier melting is due to it moving south earlier than expected rather than the claimed “because of global warming”? The quote was attributed to Sergei Balyasnikov of the Russian Arctic and Antarctic Research Institute. Would there be any reason to make that claim other than sensationalism?

Tony Brown

#266

The stated intention of the experiment was to explore the northern part of the Laptev sea
The link below shows that location.

http://encarta.msn.com/map_701514058/laptev_sea.html

The BBC report in my post 264 has a map. The floe looks way too far over to meet their original purpose so perhaps the ice station merely drifted partially the wrong way. The only international participant was taken off in April as per schedule so we can’t ask them.

Perhaps there is more to this than meets the eye and perhaps not but it needs an arctic expert to confirm;
a) whether the original scientific objectives of the expedition had been compromised by their drift pattern
b) for an expert on currents to confirm if they ended up where they would have expected.

Tony Brown.

Tony Brown

#266

The Russians wanted to explore the Laptev sea-northen part-whose location can be seen in the link below

http://encarta.msn.com/map_701514058/laptev_sea.html

My post 264 has a map within the BBC link showing the floes actual route. It looks quite a long way from where it might have been expected to be, so it might have ended up in the wrong place.

Whether there is more of a story here than merely abandonment due to global warming is not something I have knowledge of. I guess we would need an arctic researcher to tell us if the objectives of the experiment were being met bearing in mind where the floe ended up and whether that could have caused the early curtailment.

Tony Brown.

Aaron Wells

Re #265 & 265
If you look at this map (#35) you’ll see it’s not very far from the edge of the sea ice, bear in mind that the ice is far from solid at that point.

Phil,

Looks far enough from the sea ice edge that they should not have to worry about imminent melting of their ice floe. Just follow your own link that you posted earlier Drifting Station North Pole 35 and look at the Camp temperature record (upper left graph under Operative Meteorology) and you will see that the temperature barely has gotten up to 0-degree Celsius. And no matter what the case, this link shows that the location where station 35 currently is located is in an area where the ice edge is pretty much where it normally is and is not receded northward at that point as a result of abnormal melt rate. NSIDC Sea Ice Extent 07/13/2007

Dave Gallagher

I once had a Chinese Physics professor who told our class (which was, as a group, having trouble with an easy problem) your heads are so high in the clouds, you can’t see the ground.

Russians could care less about global warming, A. they want it to be warmer B. they have never cared about the environment C. They think that we are fools to worry about it. (well not all of us, but westerners in general)

When they announce to the western media ” we had to abandon our research station due to global warming”, It’s a BIG JOKE! Then they laugh because the media doesn’t get it, pretty funny if you’re in on it.

TAC

Preliminary Unofficial Day 196 Race Report
2008 had a big day, with daily melt apparently exceeding 2007 and every other year since 2002.
7 15 2002 8.917344 -0.103437
7 15 2003 8.953281 -0.034844
7 14 2004 9.157500 -0.068594
7 15 2005 8.488594 -0.058594
7 15 2006 8.104531 -0.035782
7 15 2007 7.690313 -0.094687
7 14 2008 8.600625 -0.097813

Phil.

Re #277

Starting Ice distribution, winds/currents, air temperature history in that order would be my guess.

Philip_B

In 1981, third week August, I was in Winisk on the shore of Hudson Bay and you could clearly see sea ice a few kilometers offshore. That area is where apparently the sea ice persists longest.

I found the following quote from a paper that seems to be from the 1970s.

perhaps very small ice concentrations do usually linger in the southwest Bay into September.

The paper also has a graph showing 2 or 3% ice cover in August.

Either the melt is earlier in the year now or the satellite data doesn’t handle small(er) ice areas well.

Ryan Maue

The NH sea-ice anomaly from the Cryosphere Today Tale of the Tape spurred me to grab the simplest gridded data of ice concentration and plot it up in GrADS. This is the NCEP/NCAR Reanalysis Daily Mean Ice-Concentration data, which I assume is of decent enough quality to show it. This data is not the best resolution, 192×94 grid points covering the globe, but so what.

I plotted up the weekly sea-ice concentration for 2005 – 2008 in a 4 panel plot … Ryan Maue Weekly Sea-Ice concentration

2008 has more ice near Russia, less near Arctic Canada … FYI The hole near the North Pole is NOT real, b/c the Reanalysis grids extend only to 88.542N (it is a spectral model).

hengav

Philip_B #280

The paper seems to be dealing with a 9 year period from 1958 to 1966. Published in 1969. 2008 Hudson Bay ice is behaving pretty much on par with the diagrams. Current ice coverage would match somewhere between “Late July” and “Early August” observations from the paper.

DeWitt Payne

Steve McIntyre,

An update of the melt rate chart would be informative. I think the key is going to be when we see a peak in melt rate. About the only way that 2008 can catch up with 2007 at this point is for the peak to be delayed and/or broadened compared to the average. That could still happen if the cloud cover goes away.

Aaron Wells

My graph above starts on April 1st, with days numbered since then. The end of the 2008 data represents today.

Phil.

Re #285

The race report seems to go online about 11 pm Eastern. A very slight gain for 2008 over 2007, but trading baskets in the 4th quarter isn’t going to do it (and I think that this is the 4th qtr equivalent for the melt season). 2008 is about 890,000 sq km behind 2007 – and it’s hard to catch up because 2007 will continue to put strong numbers on the board next week. Even if 2008 had an entire week of 100,000 sq km days, it would still be about 750,000 sq km behind. And 2008 has shown little appetite so far for putting a string of 100,000 sq km days together. Maybe we’ll be surprised.

The total area is rather closer to last year’s value at present, current anomaly = -1.31 million sq km.

Phil.

Re #291
You certainly do misunderstand, the graph you have drawn is the extent not the area, (currently 6.535 million sq km).

Aaron Wells

Yes I noticed the similarity in geographical area between 2004 and 2008, can someobody do a plot between those two years for daily ice melt for a comparison.

Hi Andy,

Here is a graph of daily ice-melt starting April 1st, between 2004 and 2008. You’re right. They are dramatically similar.

STAFFAN LINDSTROEM

#287 Aaron … how can 80-90000 km2 melt on April 4 2008 I don’t think
compression can do it either SO check UIUC [CT] Compare Daily Sea Ice that date
….YES It’s the ICECAKEMONSTER … I better save that image…NOW!!

Phil.

Re #297

I can’t vouch for the data, but I have no reason to believe there was a problem.

From the data you posted there obviously was a problem with the data at that time. The extents on the 5th and 6th were identical to 8 sig figs, that just doesn’t happen in the real world!

Aaron Wells

#299

Staffan,

I have no idea what you are ranting about. First of all Steve Sadlov’s comments were made about Cryosphere Today, not IARC-JAXA Information Services. And second of all, you are making way too big of a deal about a single day’s data. Most likely, multi-day melt information got lumped into one day instead of distributed over several. Notice the drops to zero before and after. Big deal. Maybe I should have smoothed the data.

My simple
question is: Was the observer on July 13 not informed what
the CORRECT max temp SHOULD be??? Temperature saturation??

I haven’t the faintest idea what you are talking about.

Phil.

Re #301

It looks like Cryosphere Today and IARC-JAXA use the same data.

Cryosphere today uses SMMR/SSMI data: “In order to maintain a consistent data source for the last part of the period, all data from October, 1978 through December, 1998 are from the SMMR/ SSMI sources. Data from previous versions of this data set were replaced by SMMR and SSM/I data from Oct. 1978 – Dec. 1990.”
The scanning multi-channel microwave radiometer (SMMR) is on board the Nimbus-7 satellite and special sensor microwave imager (SSM/I) is on board the DMSP-F8 satellite, this data is also used by NSIDC (they recently changed to a new version of this scanner): “The DMSP F13 satellite that has been central to our Arctic sea ice analysis for the past several years is nearing the end of its mission. As is standard data practice, we have transitioned to a newer sensor, in this case the DMSP F15. The DMSP F15 has the same type of sensor as the DMSP F13.”

JAXA uses: “the Advanced Microwave Scanning Radiometer for EOS (AMSR-E) is a modified version of AMSR that flew on ADEOS-II. NASDA provides the instrument for flight on board NASA’s Earth Observing System (EOS) Aqua platform. AMSR-E is indispensable for Aqua’s mission, which is dedicated to the observation of climate and hydrology.”

So it appears that while the data channels show very similar results they aren’t from the same sources, I’ve noticed signal dropout such as you showed above both on JAXA and CT but not at the same time which to supports the equipment data.

Phil.

Re #303

The extent sums the area of all the pixels that show an ice concentration of greater than 15% whereas the total area is the sum of the pixel area times the concentration, area is necessarily less than extent but there is no fixed relationship between them. You could have a case where the ice wasn’t fragmented and area and extent were very similar, or a large extent of very broken uo ice at low concentration, in which case the area would be very much less than the extent. Right now extent is ~8.5 and area is ~6.5 so the average concentration is ~75%, in the winter it was about 14/14.8 = 95%. A record in one does not mean a record in the other, last SH winter was a record for one but not the other. I assume that the use of extent harks back to the shipping days when that was all could be measured, area tells you a little more about the nature of the ice (in my view).

Steve McIntyre

Day 198
Another day off the clock. 2008 trading baskets with 2007, losing a little ground and back under 100,000 sq km.

month day year ice diff
47 7 17 2002 8.756250 -0.076719
412 7 17 2003 8.782031 -0.076250
777 7 16 2004 9.029375 -0.044063
1143 7 17 2005 8.311250 -0.089844
1508 7 17 2006 8.013750 -0.066094
1873 7 17 2007 7.498594 -0.093906
2238 7 16 2008 8.416094 -0.090156

Revised:
2238 7 16 2008 8.423438 -0.082812

Phil.

Re #272

Phil,

Looks far enough from the sea ice edge that they should not have to worry about imminent melting of their ice floe. Just follow your own link that you posted earlier Drifting Station North Pole 35 and look at the Camp temperature record (upper left graph under Operative Meteorology) and you will see that the temperature barely has gotten up to 0-degree Celsius. And no matter what the case, this link shows that the location where station 35 currently is located is in an area where the ice edge is pretty much where it normally is and is not receded northward at that point as a result of abnormal melt rate.

Nevertheless the floe that the station is on was 5 km long and 3 km wide and 1.5m thick when the station was set up. Now it is only 600 meter long and 300 meter wide, which is why they’re evacuating ~1 month earlier than planned.

AndyW

#294 Thanks Aaron for your graph. It does seem to have got insync more with 2004 over time so we will see if it follows from now to the end of August to see what happens. If the lack of multiyear ice does have an effect then it may be that August has larger losses compared to earlier years I guess. We shall see.

Thanks to Phil also for his easy to understand description of the difference between area and extent, makes good sense.

Regards

Andy

yorick

Nevertheless the floe that the station is on was 5 km long and 3 km wide and 1.5m thick when the station was set up. Now it is only 600 meter long and 300 meter wide, which is why they’re evacuating ~1 month earlier than planned.

This should go in the dictionary next to the phrase “grasping at straws”

Dave Dardinger

Folks,

I don’t know that Phil is making any particular point. He’s just stating the facts as it were. But I do wonder just what caused the reduction in size of the floe? IF it were only 1.5m think to begin with I’d expect it can’t be melting. I expect it comes from splintering of the floe from collisions (or cracking from wave action.)

Phil.

Re #314

I don’t know that Phil is making any particular point. He’s just stating the facts as it were. But I do wonder just what caused the reduction in size of the floe? IF it were only 1.5m think to begin with I’d expect it can’t be melting. I expect it comes from splintering of the floe from collisions (or cracking from wave action.)

Thanks Dave glad to see someone reads what’s posted!
My post was in response to #272 which said inter alia:

Looks far enough from the sea ice edge that they should not have to worry about imminent melting of their ice floe………….
you will see that the temperature barely has gotten up to 0-degree Celsius

Clearly Aaron was mistaken and I corrected that in #273 & #308.
Dave, the floe is both breaking up and melting, in fact as part of the clean-up the support ship went to another part of the floe, which had separated and was now about 2km away, to recover one of the vehicles which had been abandoned there when the floe split. As a result of the break-up/melting they’d already lost their air support (no runway anymore). The fact that the floe was moving over warmer water (as I pointed out in #273) indicated that it probably wouldn’t last to the scheduled shutdown date. This had been presaged by the difficulty that they’d had in finding a suitable floe to site the station on last fall, they appear to have picked a thinner one than usual as a result.

Re #315
Thanks Mr. Patterson for confirming that you don’t actually read the posts but rather respond based on your assumptions about the poster.

I cannot discern even a shadow of such rigor in this instance when there has not been so much as a mention of the Cold Halocene Layer (CHL) and its potential contribution to the disintegration of this ice floe.

Didn’t read #273?

neglecting, deliberately or non-deliberately, contrary facts and information constitutes half-truths. Half-truths constitute a deception when they serve to deceive an audience.

In fact my post was to correct such mis-statements, I notice that Patterson has failed to come up with any of the contrary facts which he claims that I’ve ignored.
You might think from his posts that I made some statement attributing the breakup to AGW, which of course I didn’t, all I did was rebut some posts that asserted that there was no reason to abandon the floe because it wasn’t deteriorating.

Re #312
yorick, I haven’t the slightest idea what you mean by ‘clutching at straws’, I guess you didn’t read the posts either, care to elaborate?

Aaron Wells

Phil, regarding your comment in 316:

Clearly Aaron was mistaken and I corrected that in #273 & #308.

What’s clear to me is that you think that the fact that the ice floe was smaller than they calculated was evidence of warmer than expected waters. It’s clear to me that their planning was wrong, not that their planning was right and their floe melted faster due to warmer waters. My point that I had made earlier to and that you referenced, was that the area that they were in, it is clear from various charts, are still in near-100% ice density, and also not in an area where the normal ice extent had receded inward from normal.

So that fact suggests to me that the waters they are currently in are not significantly anomalously warmer than they should have anticipated, and that they just calculated wrong. Your so-call correction only made the point that the ice floe had melted significantly. That in no way “corrected” my earlier point.

Phil.

Re #318

What’s clear to me is that you think that the fact that the ice floe was smaller than they calculated was evidence of warmer than expected waters. It’s clear to me that their planning was wrong, not that their planning was right and their floe melted faster due to warmer waters. My point that I had made earlier to and that you referenced, was that the area that they were in, it is clear from various charts, are still in near-100% ice density, and also not in an area where the normal ice extent had receded inward from normal.

No the floe was smaller because they had to compromise in the choice of the initial floe because of the lack of suitable candidates last September, that is not ‘bad planning’, they’ve done 35 of these stations, they know what makes a good one! Greater than 90% ice concentration does mean that the floe is not melting (from below, you ignored my rebuttal of your comment on the significance of air temperature, #273). Being thinner means that the floe would be more susceptible to mechanical breakup, which actually happened. The reference to warmer water was in particular concerning the water that was about to be entered which would significantly reduce the lifetime of the already diminished floe, hence the decision to close down the station a month early.

Sergei Balyasnikov, a spokesman for the Arctic and Antarctic Research Institute, said when the station was deployed:
“The unit was set up on a 3.3km-wide and 5km-long ice-floe with an average ice thickness of 1.5m.
An ice-floe suitable for the unit could not be located until September 18, when the Akademik Fedorov scientific-expedition ship, following in the path of a Russian ice-breaker, discovered an area with a two-year build up of ice.”

Note that they usually like to select a site with 2.5m thick ice.

What we’ve had here and elsewhere on the web is a bunch of posters who want to downplay the fact that the station had to be left early and are prepared to ignore the facts in order to do so. When called to order they then have the nerve to accuse their critics of having an agenda! I wouldn’t have posted about this if it hadn’t been in response to mis-statements and omissions, what was it Patterson said: “Stating facts and/or pseudo-facts “as it were” and neglecting, deliberately or non-deliberately, contrary facts and information constitutes half-truths. Half-truths constitute a deception when they serve to deceive an audience.”, sounds right.

See - owe to Rich

320 (Phil)

Re all this ice floe back and forth, I think Phil has been pretty straightforward here, even though I may have (thought I) detected a slightly AGW slant in earlier postings. We shouldn’t mind if he is auditting the auditors, fairly.

Anyway, I think what it all proves is the Russians couldn’t model well enough how fast and where their floe would go. Poor modelling – where have I seen that in climate science?

Go with the floe!

Rich.

Tony Brown

#318 320 and others

I made a variety of posts on this subject #264 267, 268, 270 referring back to them for the maps would be useful. I think we are actually dealing with bad planning and human fallibility here. Reference to the previous posts will give the details of the aims of the expedition.

To bring this up to date, firstly the original details from the Russian agency are given below 1) The expedition set off a month later than planned on a floe already known to be in serious danger of cracking.

They intended to drift across the North Pole but were 500 kilometres short when they started drifting back into warmer waters (item 2)
Item 3) is a quote from the only International observer-who left in April-the link is to an interview he gave to Nature.

All in all the floe was badly chosen and unstable from the start, and perhaps because they set off late the currents didn’t take them where they were expected to go. Consequently they ended up much further south and west than intended (according to their 2007 plan-previously posted). This was an expensive operation which if it went wrong because of a series of human mistakes is embarrassing, and the director would want to save face by blaming other factors. I am not saying this IS the correct explanation, but certainly this trip didn’t pan out as per the original stated details.

Whilst the weather played a big part last September it didn’t seem to be the major cause of the failure in 2008

References;
1) Sept 2007 “The last stage in organization of a new Russian polar research station is coming up. This year all the works have been complicated by the abnormally warm weather caused by the cyclone in the northern parts of the Arctic Ocean.

According to Vladimir Filatov, the representative of MVK, at the moment the vessel “Akademik Fedorov” continues to drift with the ice-floe. At 2 p.m. they plan to de-berth the vessel of the ice-floe where the researchers have already put down and put off for Murmansk. The expedition’s debarkation and discharging is being done at emergency because it sleets and strong wind blows. Thus, the helicopter works from time to time when the weather is more or less ok.
Because of the bad weather conditions – squally wind and almost zero visibility the de-berthing of “Akademik Fedorov” can be deferred to a later date. Moreover, there is a danger of the ice-floe cracking. In spite of that, at the moment almost the whole expedition team has put down and is settling. Thus, it is really hard to predict any further actions at the ice-floe because the abnormally warm weather can totally ruin it. In this case the team will be immediately evacuated.”

2) “During the course of winter, the ice floe will drift in the Arctic Ocean and across the North Pole. During the drift, a variety of measurements carried out at the station will provide information about current climate change.”

3) “We set out in September 2007 from Severnaya Zemlya, an archipelago off the Russian north coast. From there the journey went northwards until we reached 86.5º latitude — less then 500 kilometres from the North Pole. In spring we drifted back on a southwesterly course across the Arctic Ocean. When I left the camp on 10 April, ice conditions were still good enough for an AWI aircraft to be able to land on the ice-floe. Before the premature break up this week we only experienced three smaller cracks in the ice. But the camp was eventually abandoned when it was just 80 kilometres north of Svalbard, because the ice had become too thin.”

http://www.nature.com/news/2008/080716/full/news.2008.956.html

(Interview with the sole international participant)

Tony Brown

See - owe to Rich

323 (Phil)
Many thanks, I wasn’t aware of that.

But I am a bit worried about the data. Your link gives a total area of 6.333m km^2, but the standard CT graph, for about July 10 currently, reads about 6.1m km^2. So first, your link may not be any more up to date than the standard one, and second, that difference of 0.2m km^2 in the values is somewhat disturbing – unless your link is even more out of date (say by 2 days) than the standard one.

Is this something we should email the CT folks to clarify?

Rich.

rex

notice delay in updating/changing NH data at crucial times

this has happened many times already SH data
http://mikelm.blogspot.com/2007/09/left-image-was-downloaded-from.html
Fortunately people keep records these days

Steve McIntyre

Phil, I take it that you haven’t been able to locate digital information on are other than a current reading available on iPhone. Pretty odd to say the least.

Phil.

Re #328

notice delay in updating/changing NH data at crucial times

this has happened many times already SH data
http://mikelm.blogspot.com/2007/09/left-image-was-downloaded-from.html
Fortunately people keep records these days

Unfortunately they apparently are so wound up their conspiracy theories that they don’t read the scales of the graphs.

Steve McIntyre

Phil, I’m getting tired of people bandying around the word “conspiracy”. I don’t think any post along these lines has made any useful contribution. Please stay away from such language in the future. Posts using such language by anyone will be subject to deletion. Feel free to request deletion of offending posts .

Phil.

Re #334

No problem Steve, I was just astounded at a post complaining about CT rescaling a graph because the plot had maxed out so the scale was changed from -3->16
to -3->17, with the final comment to the effect that ‘it’s a good job that someone’s watching these tricky guys’!

Regarding the CT data they appear to archive everything up to last Dec but I don’t know where the current data is. I’d ask William Chapman, I’ve found him to be responsive in the past.

Phil.

Re #337

Today’s CT data when plotted vs a year ago which I attached to #329 shows the large area of fragmented ice which if it follows the trend of last year gives plenty of scope for extensive area reduction during the next two months. As I pointed out to Steve the area is following a trend closer to last year’s than the extent is which means that we have a more dispersed ice field than last year. If you look at the regional areas all but the Laptev Sea and the East Siberian Sea are at or close to last year’s values (due as I’ve said above to the unusual Transarctic drift so far this year). Consequently we’re likely to see an unusual ice pattern in a couple of months time, I expect there to be a large residual in the Laptev and E Siberian seas unlike last year.

Luis Dias

#336

Phil, from my (poor) understanding, the question was not about the scale between 16 to 17 million km, but between the anomaly graphs, which were quite different. From the first, the anomaly is much more positive than the second one. I’m sure this will have a decent rational explanation, but speaking without knowledge of such, it surely sounds weird to me (but I’m nobody), specially considering that the total ice graph didn’t seem to change its values.

RomanM

#341 Mrpete

A Photoshop comparison indicates that the difference is a quantum decrease of approximately 10^6 km^2 in the anomaly curve at April 2007. This drop also seems to straighten the the area curve above it at the same place. If Phil is correct, then this must be a previously unknown side-effect of a simple vertical scale change in a graph. 😉

Aaron Wells

Day 199 Race Report
2008 gained a little on 2007 in early returns, but remains over 900,000 sq km behind. Day 198 revisions reduced yesterday’s melt by about 10,000 sq km. Once again 2008 failed to break 100,000 sq km. Three weeks from today 2007 was at 5.5 million sq km. To catch up, 2008 would have to average over 130,000 sq km day every day for the next 3 weeks when it’s only had an occasional 100,000 sq km day to date.

month day year ice diff
48 7 18 2002 8.671094 -0.085156
413 7 18 2003 8.707969 -0.074062
778 7 17 2004 8.970156 -0.059219
1144 7 18 2005 8.205313 -0.105937
1509 7 18 2006 7.942188 -0.071562
1874 7 18 2007 7.427188 -0.071406
2239 7 17 2008 8.330313 -0.093125

Another fairly large upward mid-day revision of the ice extent. The new updated extent for 7/17 is:

7 17 2008 8.337969 -0.085470

I difference of 7656 km^2

Today’s melt rate graph (in km^2) (days starting April 1)

Real Richard Sharpe

Re: 338.

Phil, can you define unusual ice patterns?

Perhaps you mean, more ice than last year?

From those two graphics of the NP, it certainly seems like there is lots more ice there this year.

Ryan Maue

Just a question: but in Al Gore’s speech on climate change, he makes a claim that I can’t seem to verify. I sure wish there were footnotes.

The climate crisis, in particular, is getting a lot worse – much more quickly than predicted. Scientists with access to data from Navy submarines traversing underneath the North polar ice cap have warned that there is now a 75 percent chance that within five years the entire ice cap will completely disappear during the summer months. This will further increase the melting pressure on Greenland. According to experts, the Jakobshavn glacier, one of Greenland’s largest, is moving at a faster rate than ever before, losing 20 million tons of ice every day, equivalent to the amount of water used every year by the residents of New York City.

I don’t have access to Navy submarine data, and I am unsure how that would help me in my sea-ice studies.

Jedwards

I’m going to apologize in advance about this post. Steve, if its too far off-topic please delete. However since we been talking Arctic Ice, and someone mentioned submarines, and since we can’t forget the poor “endangered” Polar Bears; I just have to post this story from 2003:

Polar Bear Attacks Submarine!!!!

As the new Seawolf class sub USS Connecticut surfaced in the ice pack between the North Pole and Alaska on April 27 [2003], a polar bear chomped on its rudder, then attacked it.

After a rigourous examination of the unprecedented details of this story, I can say with 95% confidence that polar ice break up caused by Nuclear Submarines has a robuse correlation with dental decay in Polar Bears! 😉

Tony Brown

#348

climate audit had a good thread on this back in 2006!

http://www.climateaudit.org/?p=554

Go to #45 where there are lots of references relating to Greenland in general. The Jakobshavn isbrae is reckoned to be the worlds fastest glacier. The point is made a number of times that even during a thirty year period of exceptionally cold weather from the 1960’s the glacier still retreated, so other factors rather than juist temperature (which must be one factor)need to be examined.

Tony Brown

Steve McIntyre

DAy 200
2008 has a modest gain on 2007, but loses a titch to 2005 (with which it is a closer race.) 2007 has 5 days averaging around 100,000 sq km up, so gains will be very hard in the next week. I’d be very surprised if 2008 doesn’t lose some ground in the next 5 days. 2005 has a few big days coming up so 2008 will likely lose some ground there in the short term, but could easily overtake 2005 at some point.

month day year ice diff
49 7 19 2002 8.581719 -0.089375
414 7 19 2003 8.622188 -0.085781
779 7 18 2004 8.871563 -0.098593
1145 7 19 2005 8.102188 -0.103125
1510 7 19 2006 7.875000 -0.067188
1875 7 19 2007 7.363281 -0.063907
2240 7 18 2008 8.237031 -0.100938

STAFFAN LINDSTROEM

# 355 crosspatch … I beg to differ WO (Weather online)
is no lower than 0C on Monday, DMI no lower than ca -1.5C
So what is your source?

kim

Everybody’s all in happy agreement here, Monday’s low is just two days and a degree short.
===================================================

AnInquirer

Crosspatch & MrPete,
I wonder about your focus on temperature. Is ambient temperature the major driver in melting Arctic ice? Although I have heard a few claims to the contrary, my general impression is that Arctic ice is primarily melted by warmer waters underneath it. Therefore, the ambient temperature would be much less of a factor.

Phil.

Re #351

A continuing ‘just the facts’ contribution in this thread would be much appreciated. I find your dismissal of the discussion to be in error, as nowhere is the C-word brought up at http://mikelm.blogspot.com/2007/09/left-image-was-downloaded-from.html. A word I will note does not appear in this thread until it was introduced by you.

As I said originally my reaction was to the posting (to paraphrase) -see as soon as we approach a significant point these guys change the data it’s a good job we’re keeping an eye on them. That’s the mentality I described with the now banned phrase and I see no reason to change my mind. The idea that any adjustment indicates some form of chicanery is not a POV that finds favor with me particularly with satellite data. The UAH have had to make corrections to their algorithms (I rather like the more etymologically correct ‘algorism’ though 😉 ), as have RSS recently. As Steve has noted on here the JAXA update their preliminary data daily, corrections are frequently needed to account for new satellites or the drift of old ones (this happened with MSU recently) this often necessitates retrospectively correcting older data because the drift wasn’t seen at first. Sometimes changes to algorithms, graphing etc. introduce inadvertent errors (e.g. RSS, CT last week.

crosspatch

“I wonder about your focus on temperature. Is ambient temperature the major driver in melting Arctic ice? ”

It is my understanding from my reading that the ocean water can be colder that 0C and fresh water from precipitation and runoff (river water, etc) will freeze because it floats on the saltier water and there is little mixing. In a comment concerning melt ponds at the pole in a thread here someone also mentioned that water from melt caused by air temperature/sunshine can re-freeze when it floats on the ocean water which maybe be several degrees below 0C.

I have no information that the ocean water temperature at the pole is above 0C.

Sean Egan

Ryan Maue
Do a google for “ice artic Rothrock Polyakov SCICEX”. The sub data is published under the name SCICEX, but published date is 5 years old. IceSat should also give thickness, but there was a manufacturing fault, and they only turn it on from time to time.

Gene II

YouTube video link did not attach in previous comment :

tty

Re 365

There is an excellent paper on Jakobshavns isbrae with full historical data here:

http://www.geus.dk/publications/bull/nr14/index-uk.htm

It is well worth reading but unfortunately anglosaxon climate scientists don’t seem to read foreign papers, even if they are written in english.

crosspatch

“My simulated “ice-growth” model has provided a complete 99% verification of this fact.”

Using one or two standard deviants?

STAFFAN LINDSTROEM

#371 John Lang you don’t know how right you are!? If
you look at UIUC/CT CDSI Look at July 12 and then
July 19…Seems that 80 % and more concentration areas
have EXPANDED!! Could phil. or any other ice expert
help us out…Mark Serreze you’re welcome here!!
Steve Sadlov…is this unprecedeeeeeented??
(Sorry just got some SNF)Back in 8 hours from
other work….

Tony Brown

John#371

As all other climate scientists strive for absolute certainty with their theories why do you think only 99% is good enough? Apply for a large grant and go and do some more research.

Tony Brown

John Adams

Anyone watch the TV show Ice road truckers? Truckies driving over ice lakes to deliver heavy goods. Well looks like they will be using ships in the not too distant future.

Steve McIntyre

Day 201 Race Report
Revised day 201 melt was 86,700 sq km, losing ground to 4 of the past 5 years, including 2007, with another day off the clock.

month day year ice diff
7 20 2002 8.514688 -0.067031
7 20 2003 8.529375 -0.092813
7 19 2004 8.748750 -0.122813
7 20 2005 7.990938 -0.111250
7 20 2006 7.800156 -0.074844
7 20 2007 7.271094 -0.092187
7 19 2008 8.156875 -0.097969 – prelim
7 19 2008 8.168125 -0.086719 – revised

Phil.

Re #381

Ah Steve but 2008 is saving up a big Hail Mary play. 😉
Rather than reduce everything to a single number I like to look at the big picture (my research was laser diagnostics & imaging so no surprise).
Below is the AMSRE picture last July 19th :

and here’s today’s:

Looks like there’s quite a lot of broken up ice saved up within that extent, based on previous years you’d expect some rapid movement in the near future.

tty

Phil
If I were you I wouldn’t put too much faith in those density measures. Take a look at the area of very dense ice off northeast Greenland from about lat 77 to lat 80. That is an area of fast ice, i. e. a continuous ice sheet frozen to the coast. You can verify this by checking a satellite image, or if you prefer an expert opinion, by looking at the Norwegian sea-ice site(http://polarview.met.no/).
Now go to cryosphere and run through the 30-day view of ice conditions. You will see that area changing from dense to open and back to dense, though it has been fast ice all the time. So what is the satellite relly seeing? Probably that the snow on the ice has melted, formed melt pools, and then drained away.

Phil.

Re #384

I tend not to pay too much attention to individual days, the uni-bremen image from AMSR-E gives a higher resolution for such small structure and there is some variation in that region but it seems fairly systematic and I assume that it’s cloud effects. The large area of broken ice to the west is a consistent feature.
From today’s image:

John Lang

Re 384 and 386,

Here is a MODIS sat picture of the area from a few hours ago. There is some cloud cover but the false color image shows the visible sat picture is a pretty accurate depiction.

The ice on the coast looks solid (there are some melt ponds on the surface which might disrupt the density measures) but as you move off the coast there is a lot of broken ice.

4km resolution

250M resolution

SteveSadlov

In some of the debate here, we are seeing the elements of sea ice book cooking. Some of you are now appreciating how it’s done by the number fudging pros.

tty

Re 386
I agree one should not put too much importsnce on a single day so I suggest you do as I suggested – check a whole month. Incidentally it is Greenland to the west – not an area of broken ice.

Phil.

Re #388

Re 384 and 386,

Here is a MODIS sat picture of the area from a few hours ago. There is some cloud cover but the false color image shows the visible sat picture is a pretty accurate depiction.

The ice on the coast looks solid (there are some melt ponds on the surface which might disrupt the density measures) but as you move off the coast there is a lot of broken ice.

Yes John, there’s a nice agreement between the two systems, they both show the developing polynya along the north coast nicely. The fast ice is showing a few cracks.

Re #389 comparison with the Modis shots such as here is one way or to check out the ‘eggs’ for the area.
If you want to see the current in that area check out the buoy data:
http://iabp.apl.washington.edu/maps_daily_naresstrait.html

Re #392
tty: I do all the time, by the West I was referring to the original picture not the cropped piece, i.e. the W arctic basin.

Hans Erren

Marieke de Vries (Dutch NOS Journaal reporter) is having difficulties reaching the North Pole:
http://player.omroep.nl/?aflid=7535529
Scientist speak english with dutch subtitles

Mariekes weblog:
http://weblogs.nos.nl/buitenlandredactie/2008/07/21/nos-poolexpeditie/

We zijn er ongeveer 650 kilometer vanaf geweest. Dat is een enkeltje Amsterdam-Berlijn. Op 80 graden noorderbreedte. Dichterbij kwamen we niet. De Noordpool bleef voor ons een blik richting de noord, toen we met ons schip vast kwamen te liggen in metersdik pakijs, dat zich voor ons uitstrekte in wit en azuurblauw, kilometers ver, tot aan de horizon.

Translation

We came as close as 650 km. That’s a single journey Amsterdam-Berlin. At 80 degrees north. We couldn’t get closer. The North Pole remained for us a look into northern direction, when we got stuck with our vessel in meters thixck of packice, that steched out before us in white and azure blue, for kilometersupto the horizon

Fridtjof

A Dutchman myself, I just saw the late edition of our national news program.
Our occasional arctic correspondent, who normally covers the whereabouts of our royal family, immediately proved to have a contingency plan in case the arctic sea ice would not cooperate. The subject of her coverage: more insects on Spitsbergen than there used to be 30 years ago, probably due to global warming.

Gene II

Re : # 396

“Marieke de Vries (Dutch NOS Journaal reporter) is having difficulties reaching the North Pole:
http://player.omroep.nl/?aflid=7535529
Scientist speak english with dutch subtitles”

I see that story is kicked off with a video bite featuring Mark Serreze. He is also featured in the YouTube video I linked above in # 368 :

that gives funereal predictions of summer 2008 North Pole ice. In that video Serreze says the observation of North Pole ice does not match his “computer simulation”. Therefore, he concludes “we’re probably 20 to 30 years ahead of schedule”. He does not conclude, for some reason, that his computer simulation is wrong

Steve McIntyre

Another day of essentially trading baskets, but 2008 is falling further behind 2007 with the gap almost a million sq km.

month day year ice diff
7 22 2002 8.345469 -0.090000
7 22 2003 8.344219 -0.079062
7 21 2004 8.652188 -0.045312
7 22 2005 7.747656 -0.108594
7 22 2006 7.688281 -0.057500
7 22 2007 7.066406 -0.101250
7 21 2008 7.985313 -0.091875

AndyW

I agree with Phil think there is potential for large losses coming up. However 2007 still has 4 or 5 days of +100K reductions and as 2008 is so far behind already it’s unlikely to catch up.

nevket240

http://www.bloomberg.com/apps/news?pid=20601124&sid=alYnovfZgjk0&refer=home

I get the feeling these “people” are looking for an ‘out’ on this issue. The excuse may be justified if they were to say it has happened previously, but they do not. Almost as if it is an unnatural event.

regards

Aaron Wells

The university of Bremen still uses a linear scale. It also seems to take much longer to load the images now for some reason.

It always times out on me when I try to access the Bremen pages.

I’d expect a significant revision tomorrow, when the new data was posted there was a missing swath in the image.

In which direction did you expect the revision? It turned out to be a pretty normal upward revision of the ice extent. (about 4000km^2)

Phil.

Re #408
The university of Bremen still uses a linear scale. It also seems to take much longer to load the images now for some reason.
I see that I expressed myself poorly, what I meant was that the ‘new style’ images on CT take much longer to load than the ones with the ‘old style’ palette.
U-Bremen does take longer because it is higher res.

I’d expect a significant revision tomorrow, when the new data was posted there was a missing swath in the image.

In which direction did you expect the revision? It turned out to be a pretty normal upward revision of the ice extent. (about 4000km^2)

No real opinion since it would depend on how they treat missing data, I had noted that when the same thing happened previously there was a larger than normal revision. The AMSR-E does seem to frequently display with missing swaths (both JAXA and U-B) but they don’t last for long.

Jared

Global models continue to point towards no significant, positive tempature anomalies in the Arctic in the forseeable future. In fact, the areas that have the thinnest ice are forecasted to be the coolest. I don’t see any reason to believe 2008 will make a run at 2007 anytime soon…and it will need to do so within the next 2-3 weeks if there is to be any shot at a record. After that, the Arctic begins to cool again and melting slows dramatically.

Steve McIntyre

DAy 204
Over 100,000 sq km with a slight gain on 2007, but another day off the clock. Abut 35 days ago, they were neck and neck; now 2008 is 10 days behind.

month day year ice diff
7 23 2002 8.233281 -0.112188
7 23 2003 8.278125 -0.066094
7 22 2004 8.585156 -0.067032
7 23 2005 7.688906 -0.058750
7 23 2006 7.625000 -0.063281
7 23 2007 6.972031 -0.094375
7 22 2008 7.886875 -0.102969

Jared

Dewitt…part of the reason for that is that 2008 has had a lot more lower latitude ice left than 2007, such as in Hudson Bay, Greenland Strait, etc. Places like that are eventually going to melt all of the way, and it just has taken longer in 2008. The most important thing to look at, imo, is how much ice remains in the actual Arctic ocean. 2008 clearly has a lot more extent than 2007 in areas that are not guaranteed to melt, and that is a huge advantage, I think.

David Smith

Arctic insolation peaked a month ago. The decline since solstice has slowly accelerated and is now entered its rapid-decline phase.

Insolation at the North Pole took over a month to decline from 525 W/m2 down to today’s 450, but the decline rate has now increased to about 5 W/m2 per day. Insolation will fall to 400 around August 2 and then 200 on August 30. Around 200 the outgoing IR is just about in balance with the incoming sunlight and the melt rate is near-zero.

While a late-summer melt may push the 2008 ice minimum down to the 2007 value it is very unlikely that an insolation-adjusted value would fall to the 2007 level.

Joshua Matthew

IF the ECMWF prog is correct for Monday, Tuesday and Wednesday then strong winds will dramatically compact the broken seaice poleward of the Beaufort Sea.

My pointless forecast (because who really cares if I’m right – and does it make a difference anyway) is that by this time next week the extent deficit will be halved and 2008 will be closing in on 2007. Polynyas in the vicinity of the North pole are not out of the question.

ared

Joshua, the difference in extend is almost a million square kilometers. For the difference to be halved in three days, 2008 would need daily “melts” (extend decreases) of close to 250.000 square kilometers per day. Now that would truely be unprecedented. Even in the wildest weeks of 2007 it reached no more than a single day of 200.000…

DeWitt Payne

ared,

Joshua said 7 days, “this time next week”, not three. For seven days the melt rate would have to average 150,000 km2/day to reduce the difference in extent by half. Unlikely. For two weeks it’s 120,000 km2/day and for three weeks it’s 102,000. Longer than that the melt rate in 2007 is beginning to decrease rapidly and catching up is potentially easier but is less likely because by then 2008’s melt rate should be decreasing as well.

Joshua Matthew

Six years of historical data is nothing in meteorological terms – if we were to draw conclusion in a similar manner for global mean temperature then 1998 extremes would never be forecast.

I always exhort my younger trainee meteorologists to put statistics to one side and look at the facts. There is a very large area of fractured sea ice, and there are FORECAST sustained strong winds. “Melt” is not a significant factor – “compaction” is.