For anyone who’s betting that 2008 meltback will exceed 2007 meltback, I think that you’ll be able to pretty much know where you stand by the end of this week and your chances are not looking good right now based on this week’s exit polls. Another Climate Audit first.
The plot below shows the daily meltback for the last 5 years. 2007 is in red, 2008 in black. Notice the surge in 2007 at the end of June and beginning of July. We’re at julian day 182 today – July 2, 2008.
The most intense melt occurred last year between day 179 (June 29) and day 184 (July 4) with 160,000 sq km meltback on day 182 (July 2) and over 200,000 sq km on day 183 (July 3). This year’s a leap year, so that July 2 is already day 182 and was only 90,000 sq km. As of yesterday, 2008 was about 510,000 sq km behind 2007 and it looks like it is losing ground day by day in the first week of July – a big melt week where it has to make time.
Today (day 183) and tomorrow (day 184) will probably tell the story. I’ll do daily updates for the next few days – it will be interesting to watch.
Both realclimate and William Connolley have recently (June 27, June 28) done posts on Arctic sea ice with neither pointing out that we were beginning a fairly critical melt week. (Connolley bet against a 2008 record.)
Update: Here’s a graphic 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 build an insurmountable lead. My guess is that 2007 already has an insurmountable lead over 2008.
425 Comments
You can collect daily info as follows for the past 7 years:
url=”http://www.ijis.iarc.uaf.edu/seaice/extent/plot.csv”
#AMSR-E sea ice concentration algorism developed by Dr. Comiso in NASA/GSFC.
daily=read.csv(url,header=FALSE)
dim(daily) #[1] 2406 4
names(daily)=c(“month”,”day”,”year”,”ice”)
daily$ice[daily$ice== -9999]=NA
daily$date<-as.Date(paste(daily$year,daily$month,daily$day,sep="-"))
daily$julian<-julian(daily$date,start="1970-01-01")
daily$mm=100*daily$year+daily$month
daily$dd=factor(daily$year)
m0=tapply(daily$julian,daily$year,min)
m0[1]=julian(as.Date("2002-01-01"),start="1970-01-01")
levels(daily$dd)=m0
daily$dd=as.numeric(as.character(daily$dd))
daily$dd=daily$julian-daily$dd
daily$ice=daily$ice/1E6
daily$diff=c(NA,diff(daily$ice))
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.
doesn’t it bottom out in September?
(I actually bet this year would not exceed last years.)
Thanks for this post–I did not realize that there were more intense melt weeks then others.
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.
Another busted flush.
The real question is if there’s a fourth of July in Trinidad.
Steve: As long as they have Carnival, who cares?
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.
re 4
Will the busting of the flush be reported or will they move on to another marvel?
What’s with that big dip around day 150-153 occurring in most of the recent years?
We had considerable debate in other postings concerning the causes of high 2007 melt levels. According to National Geographic the culprits were a warm current unprecendted sun and excesive wind. Do we know the position with any or all of those factors this year?
Tony B
I would just like to point out that in my post #9 ‘unprecendted’ is being spelt in the British fashion…
Tony B
A nice toy (insolation calculator) is here . Enter a latitude and month and you can get an idea of the insolation in the polar regions for various dates.
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.
“The accuracy of satellite derived ice concentration is usually 5% or better, although
errors of 10 to 20% can occur during the melt season.”
IPCC AR4, WG1, Ch.4.4.2
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: although your way of plotting the graphs has lots of information. The “PR” punch-line really has to do with the UK Independent’s graph. which just has accumulated loss plotted for “avg”, 2007, 2008. Their graph ends in mid-June, but your data suggests that if their plot was extended to present that we’d see 2007 breaking away from 2008 rather than the two tracking as they now appear to be in the UK Independent’s graph.
Too much information on one plot is fine for conserving space in a short-publication but makes the point less clear.
Steve: You’re probably right. This is new territory for me. Choosing a graphic for a general public takes a lot of work; my notes here tend to diarize my own thoughts and, as you know, I like the detail. I’ve tried to show how to get at the data for anyone that wants to do a prettier version,
#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 dryice melt would attract an audience ?Why strong melt should happen exactly this week to surpass 2007?
Sea ice this year could decline faster then usual later in the summer….
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.
Does the significant volcanic activity in the Arctic Ocean (as was recently reported) impact on these data? Does anyone doubt that it would accelerate melting?
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.
19 (MikeC) Yes. Andy Revkin’s quotes four experts on his DotEarth blog who claim all that extra heat is convected out of the Arctic Basin before it can rise to the surface. One, a Jamie Morison, admits only that it is ‘likely’ exported out to the other oceans. I am still doubtful, and particularly with their explanation that there is a barrier at about 400 meters where heat flux is always downward. Well it may be downward under normal circumstances, but I’m unable to understand how cooler water above can flux heat downward into hotter waters below. Yet, this is what the experts are quoted as saying in the New York Times.
==================================================
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.
Northern Hemisphere sea ice development can be followed here:
However, no-one seems to provide any narrative on the annual peaks and troughs
at the Arctic scale [or for the Antarctic]. I think a combination of a
meteorologist and oceanographer would be needed to say what the major element
have been in the year to year variations.
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
Stupid Question…we’re talking square kilometers of ice here. Is the ice thickness sufficently isotropic so that Volume of Polar Ice=Ice Surface Area x Constant?
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.
This daily-updated sea-ice extent graph at NSIDC is useful for tracking this year’s divergence from last year:
NSIDC Arctic Sea Ice Extent
It is kept up-to-date, and there appears to be an ever-so-slightly increasing divergence from last year. Of course, this could change at any time. But for now, there is some divergence.
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.
I doubt the effect would be significant. Certainly, it didn’t cause last years melt. The wind blowing the ices out of the polar regions was a very big factor.
#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.
Kim
There isn’t exactly a barrier and they aren’t saying that heat flows from cold to hot.
The temperature doesn’t vary monotonically. There is a local maximum temperature well above the sea floor. Relative to that point, heat flows up and down.
However, if you go down much further, to the tip of the volcano, heat could rise up. But we know that given the current temperature gradient, that heat is currently heating water below the point with the maximum temperature gradient.
There is some vagueness with the discussion at dotEarth, but that’s inevitable because it’s a newspaper article. A full discussion would require showing charts of isotherms and a bunch of other things.
But, basically, the discussion at dotEarth sounds pretty darn plausible.
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).
Just extrapolating the slope (-2 million km^2/month) from the graph supplied by Cyrosphere Today on the NH, it looks like come mid August when the normal melt phase ends, the km^2 will be very close to the mean, i.e. approx. zero anomaly. http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.jpg
Doing the same for the SH, extrapolating the slope (+2 million km^2/month) from the graph, at mid Sept. it looks to be about 2 million km^2 above the mean. That would be a record. http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.365.south.jpg
I did this last year and was pretty close to my estimate.
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.
Re: 34 (dscott)
In that case, your model shows a great deal of skill 🙂
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
I’m not sure about being able to tell at the end of this week, true it is diverging and will have to play catch-up but there is nothing special about this week in ice extent terms and a similar large drop might happen anytime until the start of August. Having said that the general trend seems to be a more linear reduction and last years end of June drop seems to fall into the freakish bracket.
It is quite interesting to watch though, who would have thought it!
Alan #35
Its a fair cop. (although my grandfather was part Welsh…)
Tony B
Here is an interesting photo…
#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.
We are looking good for a record SH sea ice extent in a couple of months time. I googled the news a few days ago for ‘Antarctic ice’ and got the usual the ‘ice is melting and we are all going to die’ stories. Plenty of pieces on NH ice melt but not a single story on the increase in SH sea ice.
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
http://www.navsource.org/archives/08/08578.htm
Steve #32 and others.
If you check #40 and the link above, there are lots of photos of subs in clear North Pole waters dating back to 1959. Interesting in themselves but to me the ice surrounding them doesn’t look at all thick (perhaps it’s all those subs crashhing through it)!
Seriously, this is an interesting series of photos (From the CA main site)which doesn’t seem to show thick multi season ice at all. Whether that was due to unusual weather conditions in previous months or because- for whatever reason- there isn’t usually much in that particular location only a submariner of the time could tell us.
Tony B
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
The NSIDC ice extent is now updated thru June.
Re #41
It got blown away from the pole through the Fram strait to warmer regions where it melted.
Re #4
I wouldn’t count those chickens yet Steve, check out NSIC June
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.
#41–
What Phil said. The ice was blow away, ended up somewhere warmer, where it melted.
I need to post a synopsis of the ice bets at my blog. They were no money bets, but it’s getting near the time to post a summary.
Re #49
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:
Day 183 (starting at 0) in the books. This was a very big day last year and 2008 fell behind another 100,000 sq km. Now over 600,000 sq km off the pace.
7 3 2007 8.925000 2007-07-03 13697 200707 183 -0.201875
7 2 2008 9.546406 2008-07-02 14062 200807 183 -0.098594
I guess the sea ice institutions never traded commodities.
Given that 2007 was still in a stretch run, it’s hard to see how 2008 can catch up – but I’m just guessing like everyone else.
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???
I am having a bad case of cognitive dissonance … because the NSIC site says:
This does not see like it can be reconciled with what Steve McI said.
Also, I am confused by Phil’s pretty graph. For example, why aren’t the years in order? It seems to have been deliberately ordered that way to make it look like ice extents have declined for a long time, but perhaps I have misunderstood it somehow.
Steve: I showed my data source and downloading method. I’m using IARC-JAXA daily data; also I’m using Julian dates (and this year is a leap year – and every day makes a difference down the stretch.)
Re #53
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.
#53. Phil can you give a URL to a discussion of the graphic shown?
Re #56
Sure Steve, it’s fig 4 about halfway down here
Anyway you slice it, that is one confusing graphic.
However, what is interesting is these couple of paragraphs just below the graphic:
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.
Re 46:
Notice that NSIDC show ice in the the Gulf of Finland, Riga Bay and Bay of Fundy! At least the first two are actually bathing-warm this time of year. Methinks they need to do some calibrating.
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.
So I looked around and found:
http://www.agu.org/pubs/crossref/2008/2008GL033463.shtml
but
http://www.springerlink.com/content/74m3022v72r531p7/
says
quote The frequency of stable, clear-sky periods, particularly during the winter, combined with these results, suggests that the downward transfer of heat through the lower atmosphere and into the surface represents an important component of the heat budgets of the lower atmosphere and snow/ice pack over the annual cycle unquote
So you pays your money and you takes your choice.
BTW: quote Steve: As long as they have Carnival, who cares? unquote Who are you and what have you done with Steve McIntyre?
JF
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):
A bunch of us made bet’s in a blog post last april (I think I may be sitting pretty! 🙂
See todays post to see the 12 bets.
Unless the wind starts blowing,the April blog estimates using fancy methods like Ordinary Least Squares, SWAG, or Bensinstasjonslokaliseringsprinsippet may been the pros at NSIDC!
erhmm.. “beat the pros” not “been the pros”. (Some day I’ll learn not to click submit so quickly.)
Yesterday’s sea ice extent (July 2nd) was 431,000 km2 higher than 2007 and the difference between the two continues to increase. The melt rate in 2008 at -86,000 km2/day was about half the melt rate of July 2, 2007 at -162,000 km2/day.
So, I guess Steve’s proposition that the daily melt rates are important at this time of year is bearing fruit.
#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?
Re #63
Yes I read your request and complied with it. I’m also having trouble finding the word ‘statistical’ in your request.
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.”
Phil (#69), I confess my words were not specific enough w.r.t. to what I was looking for. But since we understand each other now, I think my point has been made.
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]
I was just using the data sheet you linked to above.
http://www.ijis.iarc.uaf.edu/seaice/extent/plot.csv
#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.
Is it too early to start bets (or Bingo) on the excuses for less melting this year?
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.
Re (#76): I wondering if the ‘new-ice’ phenomena is actually a confounding variable effect. Certainly ice-thickness has an effect, which makes sea ice extent a volatile proxy…
But it seems reasonable those areas geographically prone to melting will be comprised of ‘one-year’ ice thus, one-year ice would seem to be more prone to melting away. This correlation would hold until you get instances of heavy melt in the prior year–which would then result in first-year ice being in many places where one would usually find multiyear ice.
This stuff is all fine and dandy but until you can get Vegas to give you an over under the public will remain uninformed.
“In the August 29, 2000 edition of the New York Times, the same NSIDC expert, Mark Serreze, said:
“There’s nothing to be necessarily alarmed about. There’s been open water at the pole before. We have no clear evidence at this point that this is related to global climate change.”
During the summer of 2000 there was “a large body of ice-free water about 10 miles long and 3 miles wide near the pole”. Also in 2000, Dr Claire Parkinson at NASA’s Goddard Space Flight Center was quoted as saying: “The fact of having no ice at the pole is not so stunning.”
“Hansen is only telling half the story. In the 1980s the same Dr Hansen wrote a paper titled Climate Sensitivity to Increasing Greenhouse Gases, in which he explained how CO2 causes “polar amplification.” He predicted nearly symmetrical warming at both poles. As shown in Figure 2-2 from the article, Hansen calculated that both the Arctic and Antarctic would warm by 5-6 degrees Centigrade. His predictions were largely incorrect, as most of Antarctica has cooled and sea ice has rapidly expanded. The evidence does not support the theory.”
From: http://www.theregister.co.uk/2008/07/03/goddard_polar_ice/
There’s more
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.
What the leap year does is counter drift of a quarter day a year in the calendar vs the seasons, so the effect of the leap year really amounts to 3/4 of a day, since without it, the seasonal effects of the solar orbit would have advanced by one quarter of a day compared to last year, half a day, the year before, a quarter before that, and just about right on before that. I just don’t think the leap year is that big of a deal to even notice.
Jon–
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”.
It makes sense that there are more intense melt weeks. We are just past the longest day of the year; if you think of a bell curve, we would be just over the peak of the bell right now, only slightly past the maximum with no noticeable decline, the longest days of the year, the most radiant heat hitting the Arctic. After this, it is all down hill. Ice melting will at first decrease slowly, as we the Arctic gets less and less light, the decrease in the rate of ice melt will accelerate, until the sun stops rising in the north, and ice starts to accumulate.
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
Day 184
For the most up-to-minute sea ice race report.
Day 184 was a very active meltback day 129,000 sq km, only a bit less than 2007’s day 184 (131,000 sq km.) This leaves 2008 about 630,000 sq km behind (about 5 days at today’s rate). 2008 is also trailing 2005 and 2006, despite the high proportion of new ice, but is ahead of 2002-2004 (years available in this format).
Day 184 Statistics
month day year ice date julian mm dd diff
7 4 2002 9.890781 2002-07-04 11872 200207 184 -0.094063
7 4 2003 9.762969 2003-07-04 12237 200307 184 -0.072500
7 3 2004 9.942500 2004-07-03 12602 200407 184 -0.065156
7 4 2005 9.399063 2005-07-04 12968 200507 184 -0.103750
7 4 2006 8.965313 2006-07-04 13333 200607 184 -0.120625
7 4 2007 8.794063 2007-07-04 13698 200707 184 -0.130937
7 3 2008 9.429375 2008-07-03 14063 200807 184 -0.129219
The paired picture at Cryosphere Today for July 3 2008 and July 4 2007 shows noticeable differences as well.
Note: This was re-stated at source on July 4 as follows:
month day year ice date julian mm dd diff
34 7 4 2002 9.890781 2002-07-04 11872 200207 184 -0.094063
399 7 4 2003 9.762969 2003-07-04 12237 200307 184 -0.072500
764 7 3 2004 9.942500 2004-07-03 12602 200407 184 -0.065156
1130 7 4 2005 9.399063 2005-07-04 12968 200507 184 -0.103750
1495 7 4 2006 8.965313 2006-07-04 13333 200607 184 -0.120625
1860 7 4 2007 8.794063 2007-07-04 13698 200707 184 -0.130937
2225 7 3 2008 9.434688 2008-07-03 14063 200807 184 -0.123906
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!
Roger#87
If you have any more information or specific references, could you post them to my thread on ‘historic sea ice’
Thanks
Tony B
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.
Can anyone help me understand these two graphics?
The world is a very confusing place.
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.
Richard Sharpe @89:
Yes I can. It’s called cherry picking and a joke on Hansen. Quite childish in fact.
My guess is that the mantra will change from a “record year” to “arctic ice continues its decline” , ignoring the recovery from 2007.
It reminds me a bit of hurricane after the big 2005 season. Not much talk about 2006 and especially 2007.
Nonetheless, despite the fevered predictions of the Arctic folks, sea ice melt in 2008 looks to me like it’s still going to be at relatively high levels compared to earlier years in the satellite period. I haven’t examined homogeneity issues in this record. If there are problems in homogeneity with relatively “simple” things like tropospheric temperature, it’s hard to imagine that homogeneity issues in the sea ice record don’t exist, merely because we haven’t heard any talk about them. If one parsed the satellite methods, I’m sure that there are some puzzles.
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.
Greenland Ice Sheet Slams The Brakes On:
Re #90
No need to do that lucia & Steve just save them as jpgs and process them using photoshop.
Are you asking for something to do the conversion in R?
This should be pretty much a case of (to convert a 16 bit item from one endianness to the other):
in a language like C (avoiding the use of shift and mask operators).
This can be easily extended to 32 bits, and all I am doing is isolating bytes and shifting them around.
However, to do color subtraction on jpegs will require converting the images to full un-compressed binary and then subtracting and there might already be modules for R for doing that.
#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.
Phil says:
For once, something Phil and I can agree with.
If you don’t have Photoshop, try The Gimp.
#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.
#100. Steve, I assume you want a solution for R?
I will look at the problem today. It looks like an interesting problem.
I have played with R before but will have to re-install it.
Maybe have something tonight after gym and dinner (out).
Steve: Yes please. The next step will of course be to interpret exactly what the data means and how to organize the gridcells.
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 …………….
OK, it seems to be SSM/I F13 style data as described here: SSM/I formats.
These can all be interpreted according to documentation at http://nsidc.org/data/docs/daac/nsidc0081_ssmi_nrt_seaice.gd.html
There should be 304*448 data values.
If I now do the following (which is something that Nicholas did with GISS binary files)
readBin(handle, “int”, 1000, size=4, signed = TRUE, endian = “big”)
I get numbers that look like something and which may be readily interpreted by someone who knows binary.
[1] 0 0 0 0 0 0 0 0
[9] 0 0 0 0 0 0 0 0
[17] 0 0 0 0 0 0 0 0
[25] 0 0 0 0 0 0 0 65022
[33] -16843010 -16973824 0 0 0 0 0 0
[41] 0 0 0 0 0 0 -33620226 -16843010
[49] -16973824 0 0 0 0 0 0 16645886
[57] -16843010 -16843010 -16843010 -16843010 -16843010 -16843010 -16843010 -16843010
[65] -16843010 -16843010 -16843010 -16843010 -16843010 -16843010 -16843010 -16843010
[73] -16843010 -16843010 -16843010 -16843010 0 0 0 0
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
#100, Steve:
It has a 304 byte header, followed by 447 rows of 304 bytes of data.
Data is scaled from 0 to 250.
251 is the pole
253 is coast
254 is land
I displayed it as a bitmap, and it clearly shows Greenland bottom center, the pole in the center, and the Japan at the top
#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??
I don’t know R. I coded it in C.
Looking at the docs on R, I think it’s probably:
header
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:
one gets something that looks like the Arctic, so we’re definitely getting pretty close. Thanks, Dishman. Further thoughts will be much appreciated.
Steve it’s an image file check here: http://www.tech-recipes.com/rx/696/how_use_open_bin_cue_iso_files
Steve: Thanks, but can you write down a reicpe. I can’t tell from this what it’s supposed to do.
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
Steve, Phil’s comment is a red herring. Dishman is correct. It is one-byte entries after the header. An endian statement should not be needed for 1-byte data.
OK, I have got it to work as well, had to replace some double quotes with single quotes:
I think the next step is to associate colors with different values in there …
Does this page on doing colored contour plots in R help?
Contour Plots of Matrix Data
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 …
Actually, this is more correct:
Wish I didn’t have to fix the quotes each time I copied the commands …
However, the header is only 300 bytes.
I agree about the quotes in WordPress. They are really annoying. You can get a “usual” orientation by the following:
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.
Day 186 Race Report
Day 186 was an active meltback day 128,000 sq km, a fairly typical day 186 during the past 6 years. (I note that day 185 results have been re-stated a little, reducing them.) 2008 gained on 2007 in day 186 and is trailing by about 600,000 sq km with another day off the clock, a bit under 5 days at today’s rate. 2008 also trails 2005 and 2006,
month day year ice date julian diff
7 5 2002 9.757344 2002-07-05 -0.133437
7 5 2003 9.672656 2003-07-05 -0.090313
7 4 2004 9.829531 2004-07-04 -0.112969
7 5 2005 9.237344 2005-07-05 -0.161719
7 5 2006 8.774219 2006-07-05 -0.191094
7 5 2007 8.704219 2007-07-05 -0.089844
7 4 2008 9.306250 2008-07-04 -0.128438
OK, this (from the MatLab program) is useful …
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.
Thanks again to Dishman for solving the binary read. In answer to Lucia’s earlier question, we can now subtract any two day’s values. Here is a plot from the original data as at July 3, 2008
and here’s a plot of the melt since March 31:
If you look closely at the top graphic, you’ll notice some “ice” in Lake Erie and Lake Ontario. I know that the water seemed cold this year, but not that cold!
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, those blue and white strips on the upper right hand edge of your graphics are because you have the header size as 304 bytes, when it is actually 300 bytes, and they say so in their document.
When I fixed that item, I got rid of those strips.
BTW, what did you use to get those colors?
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:
The coloring needs to be tweaked a bit to match more closely.
Race watchers, you will notice in June report, the way nsidc do monthly figures is a bit odd. It is not an average of the daily. To be counted, the pixel needs to have 15% ice cover every day of the month. The ices is moving. If it moves one pixels left without changing shape, the new pixel is not counted as it there were not there every day of the month. The old right will not be counted either for the same reason. There is no obvious reason to add this movement dependence. So you are right to look at the daily figures.
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
About the WordPress quotes. I did a quick search on the net and found the web site below. It is seemingly a WordPress plugin that “unfancifies” the fancy quotes.
http://www.semiologic.com/software/wp-tweaks/unfancy-quote/
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.
#119
Hi Steve,
2008 is lagging behind 2007 by a long way now but looking at where the ice is melting I see a lot more correspondence with 2004 than 2007. If you look at 4th July 2004 and 2008 and compare it then we have a good geographical comparison
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=07&fd=04&fy=2004&sm=07&sd=04&sy=2008
whereas a lot of the ice melted on the Russion coast last year.
Given that, I think 5.0 to 5.2 will be minimum.
Bet for todays loss is 150 000, probably being a bit too much enthusiastic here though. Just based it on the last 7 days spread compared to 2007.
Finally a picture so we can get a feel for things in our heads rather than just numbers
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, the Bremen website http://www.iup.physik.uni-bremen.de:8084/ refers to L1A and L1B versions from NSIDC and JAXA respectively. Do you (or anyone else) know what the difference is?
Steve, as I understand it they both use AMSR-E data but use the two different servers to get more consistent data flow.
Next problem for sea ice areas. The area for each pixel vary. Does any one know the areas corresponding to the 304×448 pixel grid?
http://nsidc.org/data/seaice_index/derivation.html#monthlymeanconcent
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
Day 188 Race Report
Day 188 meltback was only 87,000 sq km, lightly below average for day 188 during the past 6 years. 2008 lost a little ground to 2007 and 2006 in day 188 and is trailing by about 610,000 sq km with another day off the clock, a gap of about 7 days at today’s rate (but daily measured rates appear to be quite variable).
month day year ice diff(millions sq km)
7 6 2002 9.671563 -0.085781
7 6 2003 9.608750 -0.063906
7 5 2004 9.719688 -0.109843
7 6 2005 9.109844 -0.127500
7 6 2006 8.676563 -0.097656
7 6 2007 8.611094 -0.093125
7 5 2008 9.222813 -0.086718
Update (Sunday evening): Here is revised report:
7 5 2008 9.236250 -0.073281
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?
131 Steve
L1a and L1b are different processing levels of the images.
The methodology is described in some detail in the draft document found here:
http://capita.wustl.edu/Capita/CapitaReports/0305SurfColorRaffuse/Methodology%20Draft2.doc.
Day 186 melt results had been revised down (less melt) when I checked just now; see note in prior post.
Day 187 Race Results (from http://www.ijis.iarc.uaf.edu/seaice)
Another unexceptional day for 2008 – about 65,000 sq km of melt, about the recent day 187 average and losing a little more ground to 2007 (now 640,000 sq km behind – nearly 10 days at today’s unexceptional rate, but the daily rates fluctuate a lot.)
month day year ice dd diff
7 7 2002 9.580000 187 -0.091563
7 7 2003 9.557031 187 -0.051719
7 6 2004 9.671875 187 -0.047813
7 7 2005 9.004219 187 -0.105625
7 7 2006 8.625781 187 -0.050782
7 7 2007 8.529844 187 -0.081250
7 6 2008 9.170781 187 -0.065469
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.
nevket240 (#140):
Lots of possible answers: 1) The result is likely incorrect, incomplete, or misleading (most published scientific papers fall into one of these categories; the scientific process takes time to sort things out, and only rarely do brand-new discoveries stand up over time);
2) The result does not fit the current paradigm (“Global Warming” is easy to write about; scientific uncertainty and confusion is hard to understand and explain); 3) Very few journalists have the skills needed to make sense of contrary papers (and apparently they lack good sources for helping to understand such papers); 4) Journalists, like the rest of us, can be lazy (“the science is settled” + “everything that does not fit the paradigm is wrong and can be ignored” = “happiness for the lazy journalist”); 5) Possibly there is a stigma associated with enabling “deniers”?
IMHO, journalists should be careful about reporting odd and contrary findings; most stories written about such findings are wrong.
Finally, anyone who has worked with journalists will tell you that journalists make lots of mistakes. They work in a pressured environment, and try to write about rapidly evolving situations. It is a hard assignment; they often miss stories; they frequently get the facts wrong when they write.
😦
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:
Arctic insolation (80N) crossed the 500 W/m2 yesterday after peaking at 520 on June 20. The decline will steadily intensify from here on, hitting 400 on July 31, then 300 on August 17 and 200 on August 31. My reckoning is that little to no melting occurs below 200 W/m2.
Interesting info. on volcanos under the ice here.
RE: #142 – I wonder if wildfire smoke is starting to impact Arctic insolation levels?
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.
The algorism did it … and for that, the punishment will be to fly!
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.
so it is looking less and less likely that the alarmist’s projections of complete melt off will actually happen. it will be hilarious to watch how it is spun…..i am sure this can be a nice analog for what we can expect if temperatures continue to plummet.
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.
DeWitt Payne #150
I think there was a lot of confusion over the “ice-free North Pole” claim. Some ran with that as an “ice-free Arctic” prediction.
Although there doesn’t seem to be much confusion over this prediction for 2013 (link).
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.
Day 189 Race Report
The clock continues to tick off. Time is 2008’s enemy right now; as far as I can tell, we’re in the 4th quarter now and 2008 has to play with a lot more urgency if it’s to win. 2007 is quite happy to trade baskets. 2008 needs a few 3-pointers and as the insolation wanes, the 3-pointers are going to be harder and harder to come by.
Or to use another analogy, think of early July as being the mountain stage of the Tour de France. 2008 has to remain within range of LAnce Armstrong or fuggaboudit.
Day 189 was another sub-100,000 sq km day. A slight gain on 2007, but not the sort of numbers needed to catch Lance Armstrong.
639,000 sq km is a big deficit to make up once you get out of the mountain stages. Every day off the clock in the Alps makes it harder.
month day year ice diff
7 8 2002 9.502188 -0.077812
7 8 2003 9.490156 -0.066875
7 7 2004 9.655469 -0.016406
7 8 2005 8.905469 -0.098750
7 8 2006 8.558594 -0.067187
7 8 2007 8.455000 -0.074844
7 7 2008 9.082500 -0.092500
Revised:
7 7 2008 9.094063 -0.080937
Thanks Steve for keeping us updated with the colorful analogies.
#153 Nice analogy.
Looking more unlikely it will get even close at the moment, would like to see what at least the next two weeks bring though. If there is truth in little multiyear ice left from last year then even quite late you could expect the young ice to suddenly have big losses. Not sure if the multiyear pudding has been overegged as a worry though.
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.
I think a lot of people elsewhere are getting rather over excited, especially in the claims that the ice levels in the Arctic Ocean are the lowest for a million years.
I am running a thread ‘sea ice through the ages’ in which I am trying to gather together evidential data of past arctic ice conditions. If anyone can contribute information to it so I can piece together the jigsaw it would be great. In particular I am looking for a credible book where this has already been done to save me reinventing the wheel.
TonyB
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.
#155
I was exactly thinking about that. I guess we have to “wait and see…”
Thank you for the stretch runs, mr. Steve.
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.)
#160. IS that “unusual”? Arctic ice flow is quite interesting. I discussed this some time ago in connection with the Ellesmere Island ice shelves. There is an active “flow” in Arctic ice and even “old ” ice has a limited life expectancy. As I understand it, much of the Arctic ice is generated offshore Siberia presumably from fresh water from the large rivers (called “ice factories”) and then are moved by the Trans Polar Drift to the Atlantic where they melt. Some ice gets spun into the Beaufort Gyre and may stay there for a number of years before eventually getting to the Atlantic. So is broken ice in high summer “unusual”? Maybe, maybe not. Is it “occasional” or is “unprecedented”? It doesn’t appear to be unprecedented as witness the 1987 sub at the North Pole.
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?
NO idea what accounts for the revision. However, the revision seems to happen every day and in the 5 or 6 dyas that I’ve been watching, the revised melt has been a little less than the opening melt.
#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.
Last year through the great ice melt Artic skies were cloudless for day after 24 hour day, this was unprecedented. The Sunlight just melted the ice and pumped energy into the ocean.
Are they cloudless this year? Phil’s picture (130) looks like it is.
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.
so what is the word on july 8th?
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
Since the “New Ice Old Ice” thread is dormant, I am going to repost here a question I asked there.
Now that I’ve had “ice extent” (NSIDC) and “ice area” (Cryosphere Today) explained to me, I feel that the latter is a more fundamental and important measure (though they would obviously have a hig correlation). So why don’t we concentrate on ice area?
As others have said, ice volume would be even more fundamental, but pretty impossible to measure.
I should add that yesterday I stumbled across a magazine by our national environmental outfit, and they had an article which also concentrated on extent (2007 of course). I really would appreciate some comments on the merits of each of extent and area measurements.
A further question is why, if Steve is often lamenting pressures on his time, is he working on daily updates of this rather than waiting for the minimum in September, which is the only measurement which will really count? His choice of course, and he’s probably so efficient it now takes only a few minutes a day.
Rich.
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.
😉
#170. A daily update is very quick to do – about 1 minute. In order to feed the blog, I need a certain amount of material that’s quick.
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.
In the end, all this stuff of the sea ice race will only bring us new ammunition against the climate alarmists, as for any later terrific prediction we will be able to say “yeah, and 2008’s arctic sea ice extent was going to reach a new record too, the statistics were clear, weren’t they”. I would say that they have shot their own foot with this absurd prediction.
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.
There is a buoy now transitioning towards the North Pole (from the Beaufort sea) that has a tremendous three year data history. Whilst the snow depth record appears suspect – the remaining data is fascinating – especially the decrease in ice depth. Of course one buoy does not a record season make.
Hmm … did I say three? – I meant two.
Re #175
The buoy Joshua mentioned is #7413 and is heading for the Fram Strait along with all his neighbors.
buoys
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.
Re #178
Yes I’m well aware of that.
Note that just because the ice refreezes it doesn’t mean that the flow stops.
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.
“So I don’t get how an increase cloud cover could ever increase a loss of ice”
It depends a lot on the time of day of the cloud cover. Clouds at night make for a warmer night. If the increase in night time low is enough to offset the lowering of the daytime high and cause a net increase in the 24 hour average temperature, a cloudy day can result in more ice melt than a clear day. It all depends on the weather in addition to the clouds. A clear, crisp day and night vs a cloudy but warmer day/night.
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.
Re #182
One of the main events this winter was the continued strong flow of multiyear ice out of the Fram Strait.
The important event of this summer is the virtual disappearance of the Beaufort Gyre and with it the recycling of multiyear ice, instead the Transpolar Drift is passing closer than usual to the N american shore flushing out more multiyear ice. The main difference in ice loss so far this year compared to last is the slow start in the Laptev and E Siberian seas, due in part I believe to the change in the flow. The major breakup of the pack-ice in the Beaufort Sea during the winter has led to a rapid melt there and in the Arctic Basin.
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.
Steve, what’s the zero crossing Julian date on your bottom chart?
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).
“24-hour Arctic days, that it is the radiative power of the Sun which causes more melting”
I was silly to use “night” in the context of a discussion of the NP in summer. Still, weather would play a strong role. Looking at Nuuk, Greenland (quite a bit South of the NP, I know), for example I see today being forecast as cloudy with a high of 8C and a low of 3C. Friday is forecast to be mostly sunny but with a high of 4C and a low of -1C. How much sunshine does it take to overcome a degree of air temperature change given the current sun angle? And wouldn’t that diminish as the angle of the sun sinks even lower?
But the North Pole looks cloudy today with a temperature over the past 24 hours of 1C. My guess is only a little melt took place there today.
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
Re #188
Steve how about producing a column for ice area?
Steve: If you can identify a source of daily info, I’d be happy to collate it and report it, but I haven’t been able to locate one. I can calculate a number from NSIDC binary info, but thus far can’t reconcile my calculation to any of the occasionally reported information. These folks put quite a bit of info out there, but the formats are messy, everything is badly explained and it always takes about 20 times longer to do the simplest thing than it should.
I’ve done a BIG scrape of NSIDC binary data – it took most of the day, but I downloaded all their daily files since 1979 and converted them into a quickly readable format and calculated daily area and extent on a consistent (though not necessarily consistent with any reports). I’ll post on this, but quickly, my impression is that , if anything 2008 area is less exceptional than extent, but I need to parse this some more. The results are also inconsistent with JARC-IAXA numbers that I’ve been reported and, needless to say, none of the agencies reconcile these differences in their online material.
#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.
The NOAA 7 day anomaly map shows a large deep cold anomaly along the Antarctic coast mostly west of the peninsula. And Cryosphere Today shows that sea ice is rapidly increasing along the peninsula from the west, and looks to completely surround Antarctica by winter sea ice max.
Does anyone know if this has happened before. I recall reading that parts of the peninsula coast are ice free all year round, but drew a blank with Google. But maybe that was a reference to the land not the sea.
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”
I thought I’d replot the data so we can see were we stand–note this data has been smoothed. “Blue” is 2008, “Red” is 2007, “Black” is the 2002-2008 average. (Thanks for helpful start in R given by comment #1; I hope I didn’t muck up the plotting.)
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.
On a more pictorial note the positoning for the “north pole” web cam seems that it might have been unfortunate. They moved the view a few days ago and the melt pool it is standing on seems to be getting some ominous dark patches spreading
I assume they represent no ice whatsoever?
Re #195
In the view from cam #2 the posts are starting to collapse!
Jon #193
I do a weekly roundup of climate science on my blog, and I’d like to use your graph. If you have any objections, can you let me know please. There’s a contact link on my blog.
Thanks
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.
The north pole image for July 10, 2005 is here , with tilted marker pole.
In the images from prior years it appears that the North Pole melt ponds stop expanding, and occasionally slightly refreeze, starting in two or three weeks. That’s a bit surprising to me.
Of course, the camera is on moving ice and may no longer be at the North Pole.
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.
This is the 3 largest difference between the extent of 2007 compared to 2008. The first 2 were in 03/19/08 – 03/20/07 (778k) and 03/20/08 – 03/21/07 (764k). By tonight, or after the revision which seems to happen in a couple of hrs, we may have the largest difference.
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??…
Re #202
I don’t have the slightest idea what you’re rambling about here, it’s just a photograph.
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.
Melt ponds could possibly be sitting on top of ice that is quite thick. In fact, it is the ice underneath, that allows the snow melt water to be retained.
Hmmm .. my long post was blocked .. the very short version here shows no temperature anomlaies during 2007’s major change in isea ice extent ..
Yesterday’s picture from camera #1 you could see that they were in fact ponds with ice underneath and some small holes were visible in the ice under the pond surface. It seems that there needs to be just the right light to be able to see that. In the picture currently displayed as the latest from #1 (Jul 10 10:38:15) you can make out those holes on the left edge of the photo and a smaller one just to the right of center at the bottom. They appear as a patch of darker blue. So that water is most likely fresh water melt sitting on top of ice. It looks cloudy and with a current temp of 0C it isn’t likely to melt a lot there today. We are going to lose more ground to 2007 today.
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.
I’d like to help Phil and Steffan out here regarding that photograph with the tilted poles. I’m sure what Steffan was referring to is the big white ridge in the distance (it doesn’t look very big because of the distance). But what Steffan doesn’t seem to realize is that this is not a large ice ridge, but the impressive surf at the edge of the ice cap!
Rich.
Interesting data point courtesy of Steve Milloy;
German Scientists Predict 2008 Arctic Ice Melt to be GREATER than 2005, but less than 2007
Based on the numbers Steve is showing us here, even this “projection”, which has been generated by “models” looks to be at risk.
Crosspatch #181 #187
Dear Crosspatch, you aren’t the first one to try and tell me clouds mean higher temps at night and therefore more summer melt of Artic ice, and then realise later what they have said, I have had “experts” from the Met office tell me this!
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.
#203
http://www.arctic.noaa.gov/gallery_np_moods.php?year=2004
Here is a link to a series of North Pole pictures. What Stephan was referring to was surely the pressure ridges in the background, not surf.
TonyB
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.
Re 197: sure, but please host the image yourself. Also, I didn’t do much more than “button pushing”.
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?
#211 UK John
It’s that when I think “arctic” I think of more than just the pole. For example: Akureyri, Iceland is currently losing nearly 8 minutes of daylight per day. While “night” there is only a little more than 3 hours, that will have extended to 4 hours in less than a week from now. So I would intuitively reach the conclusion that it would be possible for temperatures south of the pole but still considered to be “arctic” to be cold enough to re-freeze sea ice before the pole itself is. In other words, it would not surprise me to see ice increasing South of the pole while it is still melting in areas farther North that have more sunshine. And in those areas, cloud cover could play a role in how cold it gets and how much melts or freezes.
#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!!
#216 crosspatch
Today’s declination is approx 21.84 degrees so anything South of 68.16 degrees N Latitude is getting “some” night.
Formula is 23.45 * sin(( julian_day+284) * 360/366) = Declination angle (Leap Year)
Re #208
The ice in this location is about 190 cm thick and has started to thin (it’s first year ice).
any word on today’s melt?
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
I’m a long time lurker. This is my first post. (I’m an engineer, not a climatologist, so I would like to thank Steve and all of the other experts who frequent this website for providing a source the real information to counter the propaganda.)
That out of the way, I’ve been following this thread with great interest due to the alarmist fear-mongering that the Arctic could be ice-free this year. It is most interesting to see the 2008 melt at least thus far trending back from the extreme of 2007.
I’ve been looking at the Daily Arctic Sea Ice Maps on The Cryosphere Today website. Yesterday I was comparing the day-to-day images of July 7, 8 & 9 with my 13 year old daughter and we noticed that there are pretty substantial changes in the shape of the high concentration area (the purple region). Considering the scales involved, I’m having a hard time comprehending how shape of the high concentration area could be so variable. The only explanations I can come up with are that the ice is broken and shifting due to ocean currents and/or wind or cloud cover is affecting the accuracy of the instrument.
If anyone has any explanation of this phenomenon, I would like to hear it.
If 2008 beats 2002 in total ice remaining, doesn’t that mean we’re cooling rather quickly?
Jedwards #210
This was discussed also in the thread “new ice and old ice” Jul 5 #24. Interesting that although the prediction came out early this month it is getting a lot of media attention still now. One can now ask the question what is the probability of the “near 100% probability” of 2008 Arctic being below 2005?
Daryl M #222
To be fair, I don’t think anyone has predicted that the Arctic would be ice free this year. What was said that there was a 50-50 chance the North Pole would be ice free. It was noted that “From the viewpoint of the science, the North Pole is just another point in the globe, but it does have this symbolic meaning”. http://edition.cnn.com/2008/WORLD/weather/06/27/north.pole.melting/
I would suppose we could have more ice at the Arctic this year than last year (or 2005) and still have a ice-free pole ?
#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.
Re 223. Probably what we are seeing are changes in the amount of meltwater on the ice. It must be pretty hard to separate open water and meltwater pools from satellite altitude. If you compare with the Norwegian, Danish and Canadian Ice Services (http://ice-glaces.ec.gc.ca/Ice_Can/CMMBCTCA.gif, http://www.dmi.dk/dmi/index/gronland/iskort.htm and http://retro.met.no/kyst_og_hav/iskart.html) which compile their maps manually you will notice that they change much less from day to day.
Probably this is also the reason that the Cryosphere figures for sub-areas (e. g. Hudson Bay) keep going down and then up again by as much as 100000 km2 from one day to the next.
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.
Re 209, 213, 216, 218
I’m glad that at least Staffan got the joke. I sometimes put a ! instead of a :-). Just to keep you on your toes.
Rich.
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?
Re: 227
It is winter in the Sth hemisphere right now and the
penguins certainly have enough real-estate.
This thread has made much ado about the Nth hemisphere,
but little comment about Sth hemisphere. There is a
possibility that sea-ice extent will reach an all time
high, although in the last week or so the ice accumulation
rate turned south.
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)
There were no strong T anomalies for HBLR, LBHR, HGLR, and H patterns.
There were very strong vectory wind anomlaies for HBLR, LBHR, HGLR, and H patterns.
Looking at the strongest change last year – it could be argued that the change from two days prior, to two days after the 15/07/2007, was due to compaction by strong winds towards the poles, not necessarily temperature melt.
Therein lies the problem with sea ice extent – does it tell us what we need to know?
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…
Looking at the satellite pics from today, the ice is held fast against the Siberian coast (unlike last year) and there is even snow on the ground still in some parts of Siberia. I don’t see any opening along the northern top of Russia this year. The Russian ice-breakers, of course, are the best in the world and wouldn’t have a problem getting through.
Here’s two false-color sat pics covering the top side of Russia (Red is ice and snow, Orange and White is cloud cover.)
#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.
Have you noticed that Cryosphere has just completely changed the format of its maps? so now there won’t be any more comparisons of the current ice cover with earlier years, neat!
Up until a couple years ago, I thought Cryosphere Today was respectable. Now, I consider it a complete laughing stock.
I see the “Cryosphere Today” website has some of its raw data linked. But I cannot find simple time series data. For example, the data needed to plot this graph (from their site):
Am I missing it somewhere? Does anyone know where it might be posted (or the equivalent)?
Thanks,
tim
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
The comparison is still there, the link is now lower on the page. They have just re-arranged things.
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.
It is going to start getting dark at night over most (all?) of the NWP within about 4 weeks time so once they lose the 24 hour sun there, I would imagine the melt rate is going to drop considerably. It looks to me like sea ice bottoms out in area at around August 15 or so, levels off for about an month and then starts heading up in the second half of September. I would be willing to bet that if it isn’t clear within 4 weeks time, it isn’t going to be clear this year.
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.
Preliminary Unofficial Day 193 Race Report
(Apologies to SteveM: I couldn’t wait any longer)
2008 continues to fall further behind 2007, though it caught up a bit on 2006 and 2005.
7 12 2002 9.139844 -0.122500
7 12 2003 9.234375 -0.067813
7 11 2004 9.393750 -0.064375
7 12 2005 8.647031 -0.050469
7 12 2006 8.280313 -0.033281
7 12 2007 8.015156 -0.110000
7 11 2008 8.866719 -0.051250
Steve: Thanks, TAC. I’ve been at the lake. Very cold season so far,
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
Re 248
Too early to call, but 2008 is still slightly above the previous three years, of which one (2005) was the record before 2007. Weren’t the latest predictions saying 2008 would certainly be below 2005?
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.
david #250
What is your guestimate for 2008? I think around 2005 maybe a hail-mary to 2003.
#250
Nice bell curve but I don’t think nature is that predictable.
The anomalous melt in 2007 was not caused by in-situ melting. It was the result of unusual winds that pushed the ice from west to east and out into the warm waters of the north Atlantic. Temperatures are not warm enough near the pole to produce a lot of melting.
Given that those winds are not happening and NCEP forecasts below freezing temps around the pole for the next two weeks, it seems a safe bet that Arctic ice area will finish close to normal this summer.
http://wxmaps.org/pix/temp2.html
re 252:
Thats not a bell curve that’s astronomy
Try this excel sheet:
http://home.casema.nl/errenwijlens/co2/insol.zip
#253
Thanks for that link. Seems the north of Canada is looking pretty warm for the next few weeks, that should finally finish off the ice in Hudson Bay which has seemed quite persistent.
Perhaps there is still a chance of the NW passage to be possible (outside chance it’d have to be, granted).
#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
#256 PH! Exception: 2004 Still minuscule remnants of very
rotten ice first week of September!! (EC)…FYI…
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.
Hans #254,
You are using W/M2 very predictable but try that with actual temperature and not so clean and simple.
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.
Preliminary Unofficial Day 195 Race Report
Same as Day 193 and Day 194: 2008 is falling further behind 2007, though it made up a little ground on 2006 and 2005.
7 14 2002 9.020781 -0.070625
7 14 2003 8.988125 -0.141406
7 13 2004 9.226094 -0.095000
7 14 2005 8.547188 -0.050781
7 14 2006 8.140313 -0.064687
7 14 2007 7.785000 -0.096250
7 13 2008 8.692188 -0.087812
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.
Re #262
Here’s the data on the station:
Station 35
#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.
Re #262
The drift seems normal enough. What might be a bit unusual is that the floe came down east of Spitzbergen rather than to the west which is more usual. It was presumably this that shortened the expected drift time. The situation is hardly critical. They are still well to the north of Kvitöya and the ice in that area is quite dense this year.
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?
#266
I am not qualified to say whether the floe was further south than expected but it seems a surprise to have to abandon it if the ice is more extensive than in recent years.
If the Russians had to abandon the floe early they would not want to lose face and might blame other factors. #265 seems a reasonable explanation and perhaps they had merely drifted away from the area they had wanted to survey which -according to their original plans- was quite specific.
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.
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.
Buoy 35
#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.
My posts #268 and 270
I seem to have posted the same item twice-sorry, but I was told by the program that I had used an inappropriate word so redid the post without it
Tony Brown
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
Re #272
Over the ice the temperature rarely gets above 0ºC while melting (latent heat). At that location the ice is flowing over warmer water towards the open sea, probably no more than 2 months left. See here for a better image:
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.
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.
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
Hello there,
looking at the NSIDC data, the melt seems to be slower than last year (until today at least). On the other hand, the spatial distribution of sea ice seems quite different (see here for example) and one could expect the large areas of low sea ice concentrations to melt quite rapidly. Does anybody know the reason(s) behind this change in the pattern?
Re #277
Starting Ice distribution, winds/currents, air temperature history in that order would be my guess.
OK, seems logical. Could it be related then to the “la nina” we got? Btw, after some (tedious) research, the pattern looks vaguely similar to what it was (on a daily basis) in 2004 or 1993. It could be interesting to look at the melt dynamics those years (but I don’t know where to find those data…)
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.
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.
Philip: if you go the NSIDC website and look at news (july 2) you will get the beginning of an answer. Yes, it seems like the melt started earlier this year…
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).
I have been tracking this year’s arctic melt, and keeping a plot of the ice extents for 2008, 2007, and the average of 2003-2006 using the IARC-JAXA data that Steve has been tracking.
What is interesting is how closely this year is currently tracking the 03-06 average (Blue is 2008, Green is the multi-year avg):
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.
Day 197 Race Report
TAC- where are you?
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.
month day year ice diff
46 7 16 2002 8.832969 -0.084375
411 7 16 2003 8.858281 -0.095000
776 7 15 2004 9.073438 -0.084062
1142 7 16 2005 8.401094 -0.087500
1507 7 16 2006 8.079844 -0.024687
1872 7 16 2007 7.592500 -0.097813
2237 7 15 2008 8.502344 -0.100312
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.
DeWitt,
Here is my own graph of the sea ice melt rate. This is a graph of the daily melts for 2007 and 2008.
My graph above starts on April 1st, with days numbered since then. The end of the 2008 data represents today.
#279 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.
As Steve says although it looks like 2008 is going through a “good” spell of loss now it is so far behind that it will take a peak as per 2007 in Aaron’s graph to get anywhere close,.
As for the weather, I use these satellite images to show me what is happening
http://aviationweather.gov/obs/sat/intl/
Lets you track hurricanes as well 🙂 You can see the remains of Bertha plus invest 94 in
Re #285
The total area is rather closer to last year’s value at present, current anomaly = -1.31 million sq km.
Phil (#290): Your remark does not make sense. The current area is basically dead on the decadal average–as is the trajectory which is substantially more shallow than 2007. This is all the more remarkable given the preponderance of new ice.
The effects of global warming need not be monotonic, but you can’t seem to let this one go for now. Your comments read as desperation trying to pin last years anomaly on this year. Perhaps I misunderstand you.
Re #291
You certainly do misunderstand, the graph you have drawn is the extent not the area, (currently 6.535 million sq km).
SteveM (#285),
Sorry! I’m now on vacation and off-the-grid (working through a friend’s machine right now).
TAC
Hi Andy,
Here is a graph of daily ice-melt starting April 1st, between 2004 and 2008. You’re right. They are dramatically similar.
Well there seems to be some similarities at least. The maximum in the melt rate seemingly took place around days 126-131 or something like that, which is somewhat later than last year. So maybe we will have to wait something like a month before really being able to tell whether we have “another 2007” or not…
#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!!
Hi Staffan,
Here is the IARC-JAXA data for the 1st 8 days of April 2008.
4 1 2008 14072188
4 2 2008 14074844
4 3 2008 14063906
4 4 2008 13977188
4 5 2008 13925156
4 6 2008 13925156
4 7 2008 13926563
4 8 2008 13912656
According to IARC-JAXA data, there was a big drop in extent between 4/3 and 4/4. I can’t vouch for the data, but I have no reason to believe there was a problem.
Re #297
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!
#297 ok Aaron did you see the pics??????? You can
also look at Greenland snow cover July 16 2007 AND
post #240 OF Steve Sadlov …”Laughing stock” SS
knows what he is talking about… No way 80-90000 km2
disappear/melt/is compressed/compacted in one day
from April 3 to April 4 …IF it were true we would already be
halfway to Venus…some nice woman…but hell of a planet…BEST
from Sweden Solna near Stockholm with max temps 12-15 July
Stockholm Observatory 25.0C, 23ca, 25.0C, 25.0C My simple
question is: Was the observer on July 13 not informed what
the CORRECT max temp SHOULD be??? Temperature saturation??
#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.
I haven’t the faintest idea what you are talking about.
Staffan,
There was obviously a satellite glitch of some kind on April 4.
Here is a link to a comparison between Apr 3 and Apr 4, and it is clear from the Apr 4 image that the data is messed up.
Sea Ice Comparison 4/3/08 to 4/4/08
Now, here is a link to a comparison between Apr 3 and Apr 5, which shows the data corrected for Apr 5, and showing very little change in the 2 days.
Sea Ice Comparison 4/3/08 to 4/4/08
Your point is valid. It looks like Cryosphere Today and IARC-JAXA use the same data.
Aaron
Re #301
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.
#301 Phil
Now I am confused. On Cryosphere Today, how is the Northern Hemisphere ice area and anomaly, as shown on the “tapes” at the top of the page (example), related to the images of sea ice extent (example)? In #292 you are indicating a difference between extent and area. I would like to understand how they are related.
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).
#304 Phil – Thank you.
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
thanks steve……i was confused by your comment that it was “back under 100,000 sq km”, until i realized that you meant today’s melt off. but now we are over 900,000 sq km from last year’s melt off correct?
Re #272
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.
It is quite normal for ice to melt sooner than later when it is moved by winds and sea currents into seas which are warmer sooner than the seas it has otherwise melted in in other locations later. The area in which the ice is now melting is known to have sea currents transporting warmer Atlantic Ocean waters into the Arctic Sea than are found in the areas farther eastwards where the ice was expected to drift in colder Arctic waters exiting the Arctic Sea. Nothing to get excited about. Cherrypicking an individual weather event to misrepresent a long term trend says more about the person selecting the misrepresentative sample than it does about the misrepresented trend.
#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
An offhand (and slightly OT) observation: In the graphs in both this thread’s original article and AW’s comment #287, I noticed a pronounced, similarly-shaped swing in the daily ice numbers for approx June 1 through June 10 in EACH of the years 2002-2008 (AW’s shows it between day 60 and 64 based on day 1 is Apr1, SMcI’s shows it between day 150-156 [based on day 1 is Jan1?]).
My question: is there an established physical explanation for those swings in that particular week of the 6-year data (or is there some measurement or procedural difference in the capture of that data — or did we just get lucky for these 6 years)?
TIA
This should go in the dictionary next to the phrase “grasping at straws”
Steve, as the numbers are revised every day after a certain number of hours (as far as I have seen always extend up, melt down), according to the info now listed 2008 hasn’t had a 100.000+ day since day 186. Yesterday settled at 96.406 sq km.
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.)
It’s been my impression that Phil has been doggedly pursuing reports of exceptional ice melting in the Arctic that are attributed to the AGW thesis, while neglecting informaton which would negate or cast doubt upon AGW as the cause in reductions in abundance of Arctic ice. 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. The single greatest challenge to the effectve progress of science is the need for rigor in preventing deception and self-deception in the pursuit of discovering the facts of science. 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.
We already have reports that there has been recent and major changes in the CHL that causes warmer water to come into contact with the ice canopy as the strong stratification of the Arctic water column has shifted to a weak stratification. These changes in the CHL are suspected to be caused to some extent by changes in freshwater runoff affecting salinity in the Arctic and sub-Arctic seas. The changes in freshwater runoff are being attributed to changes in precipitation patterns resulting from changes in cloud cover and cyclonic circulation patterns. The changes in cloud cover and circulation patterns are being attributed to changes in solar activity, sub-Arctic oceanic salinity and currents, and various Arctic surface and sub-surface conditions. The current research is describing these as shifting cyclical patterns that are not unprecedented.
Before there can be any suggestion or hint whatsoever that AGW has somehow had something to do with the time of disintegraton of this ice floe, there must be a reasonable consideration of the multitude of natural causes which can and most likely are responsible for this event. Not the least of these causes of disintegration for this ice floe is the report that it splintered in the crush of ice resulting from strong winds and sea currents in this season. This is especially true when you consider the fact that among the experiments being conducted on this ice floe were studies of changes in the CHL which contribute to the disintegration of such ice floes.
Re #314
Thanks Dave glad to see someone reads what’s posted!
My post was in response to #272 which said inter alia:
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.
Didn’t read #273?
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?
Today’s mid-day revision was a rather large one, 8.416094 -> 8.423438, which brings the daily melt down from 0.090156 to 0.083813, which meant that the gap increase by 11,094 km^2 instead of 3,750 km^2.
Phil, regarding your comment in 316:
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.
Hi, Phil and, thanks again for the chapter from the book. I’m now able to better-formulate my question. My question is, when the analysts modify their algorithms (which I think happens from time to time), do they reanalyze and restate all the historical satellite data or do they somehow graft the new algorithm onto the old without the full reanalysis? That’s what I’d like to read a paper or article about, including how the changes affected the analyses and the extent to which different algorithms have different sensitivities and give varying results.
Re #318
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.
290 (Phil)
Sorry if I missed this, but where do you get your figures from? My limited knowledge only gives me the Cryosphere Today out-of-date graphs which I can’t read to that accuracy. I think Steve mentioned he is interested in this too.
TIA,
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.
Re #321
I read it on my iTouch, the reformatted page there gives the data to 3 figs, today the NH is 6.333 for a anomaly of -1.309.
http://arctic.atmos.uiuc.edu/cryosphere/iphone/iphone.currentarea.series.html
HTH
#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
Re #322
Thanks, I think you’re a bit hard on the Russians, if they could have found a 2.5m thick floe last Sept. they’d have taken it in a heartbeat. They’d have preferred to get a year out of it but I suspect they’re quite happy to have got the ~10 months they did. I get the impression that they expected this since last fall, as I said before they know what they’re doing, they’re the experts on this type of set up. A point not made here is that they are good citizens in these projects, they don’t just abandon the site they recover everything and take it back with them, they sent the ship to the other part of the floe to recover the previously abandoned vehicle for example.
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.
Phil #325 The Russians probably made the best of a bad job but I suspect they shouldnt have gone last September but were unable to change things at the last moment especially with an international observer on board. I agree with you about their good husbandry-some say the Russians couldnt care less about climate change but I think their scientifc community does much they can be proud of
Tony Brown.
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
Re #326
The data is updated once a day whereas the graph on CT is updated more often. I would think that the .html graph can’t be read to that accuracy anyway, it’s rather imprecise so I wouldn’t worry about that, I look it up on the iPod if I want a number. The same holds for the date, as far as I know it is today’s date but again not represented that precisely on the graph. It may be worth verifying that with CT, but that’s what I’ve always assumed. The one thing that is messed up on CT is when they changed the web-page last week the daily NH image with the new palette somehow they caused it to not show the current image! However if you use the compare function you can get the correct date in the palette that we’re used to:
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, you are being incorrectly presumptuous yourself when you assume I didn’t “actually read the posts but rather respond based on your assumptions about the poster.” I read your posts in this thread. I also read many of your posts in other threads. I read everything you had to say about the warmer waters and the movements of the ice through the Arctic. Nowhere in your posts did you mention the role of the CHL and a number of other phenomena which are not necessarily related to changes caused by what the IPCC defines as Global Warming, also known as Anthropogenic Global Warming or AGW. You did, however, spend a great deal of effort to comment upon the news article first mentioned by others in regard to the evacuation of the ice flow we are discussing.
The news story about this event very prominently reported that one of the Russian scientific leaders making a comment about the evacuation becoming necessary at an earlier than usual date due to “Global Warming.” When the BBC and other media uses this term, Global Warming, it is generally understood the intended meaning is the same as the IPCC definition equivalent to AGW. As I clearly stated, you comments left an impression. The impression, correct or incorrect, is not an assumption as you commented. It is an observation about your comments in past threads and your comments in this thread which neglect any mention of some facts while promoting bias with your commentary about an ice flow experiment reported by a Russian scientist in the news media as a consequence of Global Warming or AGW. The existence of your comments are an objective fact. They are not my imagination or my subjective misinterpretation of their meanings. Your comments are here for all of us to see. For example:
Yes, I read #273. Nowhere in it do I see any mention of the CHL or any of the other such influences which can result in the Arctic ice conditions we see in the absence of AGW, Global Warming, or natural global warming. Why do you neglect the existence of such facts and information and thereby promote bias in disclosures? Were you not aware of such facts? If not, why not? Can you not agree that your omission of the influence fo the CHL upon the composition and decomposition of the ice canopy can only serve to bias judgements and conclusions made in the absence of such scientific evidence?
I DID NOT criticize you for not correcting mis-statements. I criticized you for biased corrections of mis-statements, biased selections of facts, and biased interpretations of those selected facts. You can prove me wrong by simple act of quoting one of your posts in which you unambiguously disclosed and commented upon the role and importance of CHL and other like processes in determining the composition and decomposition of the Arctic ice canopy. Where, for example, have you ever in this thread or any other thread commented upon the role of the Mediterranean Sea in determining the salinity and variance in salinity, stratification, and halocline influenced formation and deformation of the ice canopy? You did comment upon the Beufort Gyre and in part about its influence upon the Arctic ice canopy and its movements, but I cannot see how anything you said in that regard in anyway dispels the previously introduced news story about the ice floe melting due to Global Warming. Your comments only left the impression that your explanations supported the Global Warming claim of the news story about the ice floe, with nothing explicitly not in support of such a claim. Consequently, you should not be surprised if your audience makes the inference your comments support or least do not detract from such widespread claims about the evacuation of the ice floe due to Gobal Warming, regardless of whether or not you actually commented directly about AGW.
I also do not see how you got the idea someone “asserted that there was no reason to abandon the floe because it wasn’t deteriorating.” I don’t see such a conclusion in their remarks. Without remarking on the merits or lack thereof about their comments, I do see other people having a difference of opinion about the timing of conditions on the ice floe, but not about its progressive and ultimate disintegration. It’s my impression that you are misinterpreting the comments of other posters at least in part.
[….]
I have to strongly disagree with your comment, “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.” Speaking for myself, it seems very obvious the research team had to leave the ice floe when they did, early or not. In my opinion, you are being unfair by misrepresenting the comments of the posters. If they were downplaying anything at all, it was the notion and claim of the Russian scientist’s purported comment in the news story that the ice floe had to be evacuated due to melting of the ice floe by…Global Warming. I and other posters can think of many natural events which can and are very likely to be the cause of the disintegration of the ice floe at that date which do not require the existence of Global Warming, AGW, or natural global warming.
Do you have an agenda? I don’t know for a fact that you do or do not have an agenda, only you would know that for a fact. Whether you do have an agenda or not, I’m concerned only for any potential bias by yourself, myself, or anyone else which could hinder a proper investigation of the claims by the IPCC and others with respect to Global Warming and all of the proposed changes in our world governments, economies, cultures, and more. The topic of this thread, the observation of changes in sea ice as it unfolds, is just another means of bringing attention to the other side of the public relations disclosures about such events which biased coverage by the news media and governments neglect and or suppress. Many of us who are posting on this topic do have an agenda to the extent we want to expose some of the neglected scientific information to the sunshine of public disclosure. Our openly avowed agenda is to correct bias where we find it in the search for the scientific facts regarding Climate Science. One means we have for rectifying bias where we find it is to call attention to instances where a commentator remarks about some of the factors which caused an ice floe to melt at a given rate which others have attributed to Global Warming while failing to remark about other perhaps more influential factors.
To avoid misunderstandings and recriminations about faulty assumptions, let me simply ask you the following direct questions, letting you speak for yourself.
Do you have an agenda? If so, what is that agenda? My agenda with respect to this subtopic about the ice floe is to bring attention to the way in which major influences upon the formation and deformation of the ice canopy have not even been mentioned much less considered properly.
Do you presently support or not support the published IPCC conclusion that Global Warming, defined as AGW, is an already scientifically established fact? I do not.
Do you or do you not suspect the comment of the Russian scientist reported in the news media regarding the evacuation of the ice floe being necessary due to Global Warming is valid? I suspect such a comment is invalid due to the previously unmentioned natural influences.
Were your remarks about the melting of the ice floe intended in anyway to support an hypothesis in which the timing of the disintegration of the ice floe was due to the IPCC definition of Global Warming?
Are those questions and my own responses a fair enough means of quickly and directly dispelling previously alleged misconceptions and misunderstandings, so we can more pleasantly move forward towards discovering how this year’s Arctic ice changes and the melting of the ice floe under discussion may relate to the IPCC and other predictions of catastrophic losses of the Arctic ice canopy?
Re #328
Unfortunately they apparently are so wound up their conspiracy theories that they don’t read the scales of the graphs.
Phil,I don’t think that rex’s point had anything to do with the scale of the graphs. It was the change in the anomaly graph along the bottom. I didn’t see any difference in the scales. Did I miss something?
BTW:
You wrote that in response to my previous post. I want to make it clear that I have never accused you of having any sort of agenda. I have only disputed your statements. You making that accusation lumps me into your broad brush characterization and I wish you would be more careful in the future.
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 .
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
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.
NSIDC have released a mid month update (not sure if this is usual for them or whether they wish to explain the divergence with the 2007 plot since the start of July)
http://nsidc.org/arcticseaicenews/
It’s an interesting read especially about snow cover and why the young ice is thicker than they expected this year. They are suggesting that their could be some big days ahead due to the thinness of the ice still and also the fracturedness of it. I think people on here have commented on that as well, that the ice does not look as uniform. If the ice is thin and fractured this could lead to a different melt pattern to last year with August producing more loss than the norm. We are still past summer peak though in August as a graph of daily melt in 2007 shows.
Looking at the Danish site
although there is no large extent of open water near it currently there is a nice uniform spread of small areas all along it. I’d hazard a guess that a NE passage is a lot more favourable than a NW passage this yar even with the large expanse of open water to the north of Canada/Alaska.
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.
Re: Russian iceflow
It looks to that the primary cause of the early abandonment in 2008 was the exceptional melt in 2007, which resulted in an unsuitable iceflow being selected and that they appeared to have set out late and drifted in the wrong direction. South of Svaalbard is quite a ways from the pole.
Thanks, Phil. It’s nice to hear the details and background on this.
#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.
Re: 328, 333, 338: Phil, Luis is 100% correct.
Both the comments and the posting refer to the anomaly graph at the bottom:
I don’t see how rescaling the entire graph would change the shape of the anomaly to no longer have a positive slope. (If so, that’s quite the interesting bug in the graphing algorithm!)
For other readers, we’re discussing the posting here with these two (Old then New) graphs:
#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. 😉
#342
It seems that the main difference is after april 2007, where both anomaly graphs depart. I doubt that this is an “effect of a simple vertical scale change”, it does not compute at all.
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)
Re: 341-343
I remember that there was an explanation given for the change on the lines of “we have found a minor bug in our program” which is probably true. If so it demonstrates nicely that program errors are only found if/when they cause unexpected (unwanted?) results, in this case a record ice area. There was a similar incident with HadCRU which found a long-standing error in their routine for computing average temperatures when it resulted in a record temperature drop in January 2008.
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.
346: LOL
Re the race, the one I’m really watching is 2008 v 2005, with 2008 holding a small and quite assailable lead in ice extent. It’s not just the ice that’s shrinking, it’s my fingernails too!
Rich.
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.
I don’t have access to Navy submarine data, and I am unsure how that would help me in my sea-ice studies.
RE: “It’s an interesting read especially about snow cover and why the young ice is thicker than they expected this year. ”
This is yet another reason I question the seemingly simple minded notion of “(so called) multiyear ice is thick, more saline, and hard to melt.” Ice does not need to be around multiple years to become thick. It can happen rapidly via snowfall or compressive events. I am also even somewhat skeptical of the “multi year more saline” notion.
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!!!!
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! 😉
Re #332
Phil.,
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.
If you would be so kind as to actually consider the images posted by MrPete and/or at http://mikelm.blogspot.com/2007/09/left-image-was-downloaded-from.html you will note that there is a distinct difference in the plot of the anomaly. This is not simply attributable to scaling as you aver above. Even if one were to attribute the difference to an innacurate plotting of the zero line in the anomaly, there remains the fact that the nature of the anomay curve has changed.
Clearly as presented in the two graphs the numerical value of the southern hemisphere ice area anomaly for March through September of 2007 changed dramatically. I request and welcome your reconsideration of your comment in #332. If the commentary of the blogger was unfounded please demonstrate how a change in scale would alter half of the anomaly data but not the other half. If in fact you acknowledge the alteration of the anomaly values presented in the graphs then please be so kind as to state so here.
#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
Arctic insolation is now clearly declining and the decline is accelerating. This upated plot shows current insolation minus what I figure is the breakeven value (180W/m2) at which insolation is approximately balanced by IR loss and ice melt is near-zero.
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
“I’d be very surprised if 2008 doesn’t lose some ground in the next 5 days.”
Me too looking at the forecast for Qaanaaq (Thule) Greenland. The forecast is for low temps down to -5C by Monday. Not much melting going on at that temperature.
# 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?
Here is a nice weather forecast for Qaanaaq. Monday night low -5c.
Interesting. I have some experience with our own personal ice sheet — on our driveway here in the country. Once it gets more than 5-10cm thick, it is VERY difficult to melt it, even on purpose. We can have a week of 70f (21c) days, which melts the top layer temporarily… but freezing nights bring back the ice, harder and slicker than ever. If the daytime surface melt can’t go anywhere, no long-term melting is seen. Only at the edges, or where I punch a hole through, will the ice shrink.
Everybody’s all in happy agreement here, Monday’s low is just two days and a degree short.
===================================================
Getting caught in the spam filter. I’ll try something simple.
Pretty strong adjustment overnight in NH ice.
07,18,2008,8254844
Makes the daily change -0.083125
Steve: It’s an interesting bias, isn’t it. While we’ve been watching, I think that the overnight adjustment has always been the same direction.
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.
Re #359
Steve, that’s why it’s always good to take a look at the associated image, for example when the figure for Day 199 went up there was a wedge of a ‘swath’ missing up in the Chuckchi Sea region so I expected a fairly substantial revision and so it proved. I suspect the first prelim data is a ‘bot’ product and then later is revised following inspection.
Re #351
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.
Re #346
I did!
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.
Re #340
It did, see #341
“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.
Ryan your post#348 and my reply #352
Al Gore was correct in his belief that the flow of water from Jakobshavn Isbrae (20 million gallons a day) as being sufficient to supply NYC. However he omitted to say that it would also have been possible in the 1920’s through to the 1940’s and during the MWP.
Amongst the many studies of this glacier (and others in Greenland) is a particularly good one (dated 2007) linked below-which I took to be a thesis. This is very clear, straightforward and relies on historical observations plus photos rather than complex theories-just the sort of study I prefer!
Click to access Herrington_paper.pdf
I think the author makes a very telling observation which I would heartily endorse, coming from the historical rather than theoretical side of things. It read as follows;
‘When I opened my first glaciology text book at first I couldn’t understand the inscription, but now I do-it read
“There is nothing new except what is forgotten.”
Tony Brown
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.
Grim tone in YouTube video about 2008 North Pole melt :
As it is looking more likely that the North Pole will not be “ice free” will we see as much time spent in the media and in videos in YouTube circulating mea culpas for wrong predictions?
YouTube video link did not attach in previous comment :
While ocean currents may play an important role in ice melt in the Arctic Ocean, I doubt this is the case in Hudson Bay, as it is mostly enclosed. Wind (northerly or northeasterly) would appear to play an important role, as the final ice melt occurs toward the southwest shore (see my link from 2 days ago).
You can clearly see this in the current CT image. There is a band of ice close to the south/southwest shore perhaps a 100 Ks wide, while the rest of the bay is ice free. This band of ice is surprising far from the ice further north, at least a 1,000 Ks.
Satellite image of Hudson Bay ice in late spring showing ice breakup toward the eastern shore.
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.
The current northern sea-ice extent is 11% greater than last year.
If this trend continues, there will be no northern icecap melt at all in the summer and instead the sea ice will start extending south (making it to the equator in only 50 years.) The southern sea-ice is also showing a rising trend as well and will likely reach the equator 2 years later than the northern sea-ice.
My simulated “ice-growth” model has provided a complete 99% verification of this fact.
“My simulated “ice-growth” model has provided a complete 99% verification of this fact.”
Using one or two standard deviants?
Any questioning of the model or the science is nonsense and should not be published anywhere.
#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….
Re #349
You’re right to question it since it’s a figment of your imagination, first and foremost multiyear ice is less saline not more.
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
Proof Phil please. It may be intuitive tha old ice is less saline BUT ….
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.
Re #377
Not only intuitive but good physical chemistry, there is this thing called Google too.
Sadlov’s assertions about the irrelevancy of the term ‘multiyear ice’ are just a measure of his ignorance, you’ll notice that he makes these ‘hit & run’ posts periodically but never backs them up! Historically mariners have wanted to know which floes are multiyear because not only is it thicker but it is stronger and harder for the icebreakers to clear.
Since you’re apparently unable to use Google you might like to read the following:
http://www.eoearth.org/article/Sea_ice
http://nsidc.org/seaice/characteristics/multiyear.html
http://nsidc.org/seaice/characteristics/brine_salinity.html
http://www.tpub.com/content/ArmyCRREL/CR96_07/CR96_070017.htm
http://www.ccrs.nrcan.gc.ca/resource/tutor/fundam/chapter5/16_e.php
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
Day 202 Race Report
2008 simply can’t keep trading baskets with 2007 if it’s going to carch up. Lost a little ground in the early returns, so it will probably be a it more in the revised. Another sub=100,000 sq km day. 2008 is playing without any sense of urgency.
month day year ice diff
7 21 2002 8.435469 -0.079219
7 21 2003 8.423281 -0.106094
7 20 2004 8.697500 -0.051250
7 21 2005 7.856250 -0.134688
7 21 2006 7.745781 -0.054375
7 21 2007 7.167656 -0.103438
7 20 2008 8.077344 -0.090781
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.
#382
Or it could be otherwise. If this ice doesn’t crack open but stays put in those packs of ice, isn’t it a good frame for the new ice of 2008/2009 winter?
I’ll wait and see.
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.
well now take a look at that, for the first time the revised extend is down, melt up. A whole 156 sq.km, but still.
month day year ice diff
7 21 2002 8.435469 -0.079219
7 21 2003 8.423281 -0.106094
7 20 2004 8.697500 -0.051250
7 21 2005 7.856250 -0.134688
7 21 2006 7.745781 -0.054375
7 21 2007 7.167656 -0.103438
7 20 2008 8.077344 -0.090781 *estimate
7 20 2008 8.077188 -0.090937 *revised
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:
I wonder if the PDO cool phase that we are entering is conducive to a repeat of last year’s winds that moved lots of ice out into the Atlantic where it could melt …
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
Phil – You must undoubtedly be aware of the difficulties presented by surface melt ponds and even certain ice / snow with certain surface characteristics. What is your method of deterministically validating that what appears to be “broken up ice” is in fact broken up ice?
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.
John Lang – the north to south long shore current along the east coast of Greenland appears to be immense.
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.
Re #362,
Phil.,
Thank you for your honesty. It is good to know that you wouldn’t let the presence of an observed discrepancy in the graphed anomalies get in the way of your conclusions. The facts, while interesting, are wholly irrelevant.
Re #388
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.
Phil, your graphics are temp measures by passive microwave on aqua not ice draft? I am not familiar with this product. As no other “Nul” has stepped up, I’ll ask the dumb questions. What colour is what temp? You say the ice is broken up. Are you getting this from these diagrams above or are you looking at another product?
I have looked for your graphics using yahoo I could not find it. Is this a pay production ? If it is free, where is it.
I did find this on estimating depth of snow
Click to access seaiceatbd.pdf
and this glossy guide for school children
Click to access amsr-e.pdf
But nothing on how to read your charts.
Feel free to point me to a text on the product above and I will go away an read up.
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/
Translation
Some typo’s but you get the idea, I hope.
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.
Re #395
They are ice concentration not temperature from: http://iup.physik.uni-bremen.de:8084/amsr/amsre.html
Scroll down and you will see the current images along with their color key, I’m sorry I had cropped the images to illustrate the particular features and to get them to fit on the page. There’s also an archive of previous year’s images. It’s a product from the Japanese imaging instrument on the Aqua satellite, see also:
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
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
Re: 396 and 400
Hans, I note that our host [his choice, of course] decided to snip our earlier exchange about Ms.deVries.
Since there was nothing snipable there, I truly hope it wasn’t because I said hello to you in Dutch by way of opener. [note to Steve M: the rest of the posting was in English. There is ample evidence that both Hans and I are fully conversant in present Anglo-Saxon..: so why?]
My original comment reflected on the cognitive dissonance these well-meaning folks encounter when they come face to face with reality. By way of example: the world view [“weltanschaung”] of many of those who comment here and who by the very nature of where and how they live [urbanization being the major driving trend these days] would be inclined to look at, say, bears as somehow “cute” [ref: the Polar Bears]. Anyone who has ever been out in the bush in North America well tell you: “that ain’t so”. Likewise with the realities of the Arctic [or for that matter the Antarctic, where as of this morning the thermometer is showing around -60C].
The cognitive dissonance encountered by the likes of Ms deVries and {Dr?] Serreze is because – to put it in North American English vernacular- they are breathing their own fumes. Since they “know” that AGW is real, and the models tell them it’s real, therefore there must be melting at the Poles. Therefore, if they don’t see it, it’s simply not true. The incontrovertible fact that there is over 1,000,000 sq km more ice in the Arctic than at the same time last year, and that the Antarctic has more ice accumulation ever seen since satellites started recording things some 30 years ago, simply does not matter. A classic case of circular reasoning and self fulfilling prophecy.
The overarching concern is that these very people are on the evening news one way or the other, and that their message is what the great unwashed hear time and time again. Not what our host shows up for real on this blog on a regular basis.
If anything, it is Serreze’ delusional “conclusions” that should chill all of us to the bone. [Bad pun, I’ll grant you that]
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
I’d expect a significant revision tomorrow, when the new data was posted there was a missing swath in the image.
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.
Have you noticed that Cryosphere has changed from a linear to a logarithmic color scale with the new map format? Now most of the color change is concentrated in the 80-100% percent range while everything below 50% is essentially the same color. It makes the ice cover look much more broken up than before, which I am sure is quite unintentional…
Just for forestall any claims that I am imagining things, the old scale is still used in the comparation images below the main map, so you can compare them for yourself.
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
Re #405
Yes I’d noticed, they also have a toggle to switch between 3 different scales but that doesn’t seem to be working properly yet. I don’t like to use the log scales unless there’s a good reason, in this case there might be since there never seems to be much area of below 50%, just a very narrow band. 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.
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)
Since this thread is getting so big and getting so long to load, I am going to just post links to updated ice melt graphs:
Sea Ice Melt Rate Day 203
Sea Ice Extent Day 203
For the ice extent, I have 2008, 2007, avg03-06, and 2005. I added 2005 because 2008 is tracking it very tightly, and may give a good estimate of where 2008 may end up.
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.
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.
Today’s contribution of Marieke de Vries in the NOS Journaal was not the starting topic, (that was of course dominated by Karadžić) fastforward to minute 25:52 just after the economic news and before the weather forecast.
Todays topic:The threatened polar bear, Neil Hamilton quoted at 27:08
http://player.omroep.nl/?aflid=7535555
mariekes weblog in dutch
http://weblogs.nos.nl/buitenlandredactie/2008/07/22/bedreigd-of-doodgeknuffeld/
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.
Aaron Wells,
Since the melt rate data is so noisy, some sort of moving average seems advisable. I’m rather partial to an Exponentially Weighted Moving Average. The nice thing about an EWMA plot is that the latest data point can be included. Picking a starting value can be a little tricky, but it shouldn’t have much influence beyond the first N points where N = (2/alpha) – 1. Or you can calculate weighting factors for a given alpha and use the first N data points or data points from December, 2007. Pick an alpha as low as possible to minimize group delay, but high enough to give reasonable smoothing. Probably something in the range of 0.1 to 0.3 will work.
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
2008 may well catch up with 2007. The EWMA plot shows that the melt rate for 2008 hasn’t bottomed yet. At the same time in 2007, the smoothed (EWMA, alpha = 0.1 or N = 19) had been decreasing for several weeks. The lines will cross in a few days if the 2008 melt rate stays above 85,000 km2/day on average, as seems likely. Then 2008 will start to catch up with 2007. We’ll see.
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.
Dewitt: the last data-point is shifted by the group delay just as it would before any filter. In stock analysis, this shift is often omitted because comparing the EWMA against the unfiltered data shows whether the trend is breaking. Nonetheless, the EWMA should be shifted by group-delay when the filter is being used to smooth a dataset.
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.
Re #416
If you look at the regional breakdown in Cryosphere today you’ll see that except for E Siberia and the Laptev sea everywhere is close to it’s 2007 value. The Laptev sea has about 0.3 million sq km above its 2007 minimum, E Siberia about 0.55 million sq km above its 2007 minimum, these areas are below ~75ºN and still have favorable melt conditions. Most of the remaining ice is in the central Arctic basin (3.2 million sq km) and what happens there will determine whether a record is set or not.
As Stated in NSIDC: “An unusual area of low ice concentration is also developing near 85 degrees North latitude” so it will be interesting to see what happens there.
Yesterday’s data from buoy 07413 at Lat: 84.498 N Long: 142.108 W indicates rapid thinning in that region:
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.
Jon,
I understand that there is group delay in an EWMA plot. However the 2007 curve in the graph has the same alpha as 2008 so the group delay should be the same. I have also plotted the average of 2003 to 2006. That curve has also bottomed before 2008. The difference is small so far and could easily be noise. We’ll know in a week or two.
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…
Lots of new snow on the ice shown in this interesting pic from the North Pole Webcam. The snow cover has been on the ice for about two weeks according to the archive pics. I imagine this will affect the melt rate.
The North Pole Webcam has drifted considerably this year and is now at 84N, almost between Greenland and Svalbard.
Position (Purple line).
http://psc.apl.washington.edu/northpole/DriftTrackMap.html
The MODIS satellite pics from the past few days show the icepack within the Arctic Circle is much more solid than it was a few weeks ago.
http://rapidfire.sci.gsfc.nasa.gov/realtime/
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.
DeWitt,
you’re right, but Joshua thought the driver would be strong winds on monday, tuesday and wednesday. Until then 2007 had pretty high daily melts (on average close to 100.000), so it’s unlikely 2008 will gain much before that.
Furthermore I think you’re putting too much dependence on the EWMA. If you look at the raw melt rates of 2003-2007 for the next 7 days, you see but a few spikes, just above 100.000 km. Yes, maybe if the conditions are right 2008 will put a string a 100.000’s on the clocks, but a consistent 150.000 for a week would be beyond unprecedented. It would be a miracle, especially sice 2008 has tracked the 2003-2007 average so well these past months.
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.
revised score for day 204 is in, we have the first 100.000+ km. sq. day since day 186. No other recent (2002-2007) year had a streak that long in July, save 2006. That year the last 100.000+ day was on day 190 and there were no more after that.
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 – prelim
7 22 2008 7.883125 -0.106719 – revised
2 Trackbacks
[…] Comment on Sea Ice – the Stretch Run by kim Andy Revkin’s quotes four experts on his DotEarth blog who claim all that extra heat is convected out of the Arctic Basin before it can rise to the surface. One, a Jamie Morison, admits only that it is ‘likely’ exported out to the other … […]
[…] It’s July, Steve McIntyre reported on the current state of sea ice. So, I think it’s a good time to post the zero-money […]