« The wet side of Greenland | Main | Petermann calves again »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Seke Rob

Thanks for the update, Neven. CAPIE... shock and horror.

Nansen, Arctic ROOS, an FS fav, is not staying behind... pretty well mirrors the profiles that others show, and record low for time of year [area], or tracking last years closely. http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic

Seke Rob

Whilst MASIE and NSIDC was off till just now, the extent for their part fell 152K of which 94K left Hudson... not much banking left there at .25 million.

Seke Rob

Think this line needs revision... not for area, and does for extent?

Like I wrote above, this could be caused by holes in the ice pack that get counted for area, but not for extent.
Neven

The holes within the ice pack get counted for sea ice area, but holes often are not big enough to meet the 15% threshold for extent, and thus do not get counted. In other words, every grid cell that has 15% or more of its surface covered with ice, will be counted as 100% ice.

Did I get that right?

Seke Rob

Well, what you intend to tell is that something is mathematically amplifying the area-extent gap [it's broken up and rotten]. >15% for a grid cell is 100% ice [for extent, right]. Before I'm getting really confused [], let me hold on to what's explained here: http://nsidc.org/cryosphere/seaice/data/terminology.html

crandles

Am I confusing things if I say:

The area of a hole does not get counted for area but will get counted for extent if the cell is still over 15% concentration.

(I think suggesting the hole area is a subtraction from area is a bit confusing: i.e. suggest we have counted the area and now we are going to add a negative area. huh??)

Neven

Yes, crandles, you are confusing things. :-P

Or maybe I'm just confused.

Seke Rob
The area of a hole does not get counted for area but will get counted for extent if the cell is still over 15% concentration.

An A++, as me having a degree in Double Dutch [from 15 years working there], it requires at times to unclog ;>)

AmbiValent

Ice is cracking more and more in the NWP, but doesn't appear to be leaving it by Parry Channel or Lancaster Sound. Is that normal, or can we expect a change?

Artful Dodger

Definition of sea-ice cover (extent and area)

  • "The area of sea-ice cover is often defined in two ways, i.e., sea-ice "extent" and sea-ice "area." These multiple definitions of sea-ice cover may sometimes confuse data users. The former is defined as the areal sum of sea ice covering the ocean (sea ice + open ocean), whereas the latter "area" definition counts only sea ice covering a fraction of the ocean (sea ice only). Thus, the sea-ice extent is always larger than the sea-ice area. Because of the possible errors in SIC mentioned above, satellite-derived sea-ice concentration can be underestimated, particularly in summer. In such a case, the sea-ice area is more susceptible to errors than the sea-ice extent. Thus, we adopt the definition of sea-ice extent to monitor the variation of the Arctic sea ice on this site."
  • http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    Seke Rob

    Interesting paper found at the AGW Observer [great] resource http://agwobserver.wordpress.com/ grouped by topic in left side bar, a direct effect to loosing sea ice:

    Impact of rapid sea-ice reduction in the Arctic Ocean on the rate of ocean acidification – Yamamoto et al. (2012) [Full Text]

    Espen Olsen

    The Chinese Cutter and research ship Xue Long is on her way to the Pole, she is just entering Bering Sea these days, she also is also bringing a whole bunch of international researchers.

    http://www.marinetraffic.com/ais/default.aspx?mmsi=412863000&centerx=-173.7331&centery=63.05487&zoom=10&type_color=9

    Artful Dodger

    Hmm, Espen maybe they're planning to rescue the Shell Oil Drilling Rig that has run aground in Dutch Harbor, Alaska?

    http://www.latimes.com/news/nation/nationnow/la-na-nn-shell-discoverer-drifts-20120715,0,888755.story

    idunno

    Bastile Day!

    CT drops 150k, to 5.10, anomaly now 1.992.

    Neven

    And CAPIE takes another plunge, 2012 now ahead of other years by at least 5 percentage points, the graph looks weird!

    I feel like I'm not getting something...

    L. Hamilton

    And speaking of area, CT this morning drops below the annual minimum for 7 earlier years. More than 400k below 2007, the 2nd-lowest, on this date. And now barely short of a -2m anomaly.

    http://img.photobucket.com/albums/v224/Chiloe/Climate/sea_ice_N_min_to_date.png

    DMI showing less separation, but 2012 is about 50k below 2010 for this date.

    FrankD

    @Neven >I feel like I'm not getting something...

    But why? This is exactly what we expect to see as Arctic Sea Ice melts out. FOr instance, there's about 1.5 million sq km's of lower (~65%) concentration ice on the Pacific side that registers as 1.5 m sq km's of extent but only 1 m sq km of area. There's 6.5% of CAPIE right there.

    As long as we see ever larger areas of <100% concentration, CAPIE will go on taking a beating. As we continue to melt-spread-melt-spread, concentration will go on falling. Extent will fall less that area, and CAPIE will go on going down.

    It will never happen quite so neatly, but theoretically you could get down to a CAPIE of 15% (or 30%, I've forgotten which IJIS uses as their extent threshold). Then everything will melt out suddenly. But CAPIE has only fallen to 67%, we've still got a long way to go...

    Dave Leaton

    I'll post on area, for what it's worth and despite the known problems. We've been above the pre-existing 60-day drop total record for three consecutive days now (above a century daily average). Also, only one five-day drop total has been this high (81260) this late in the season (82760 - 2002).

    Mdoliner43

    If you look at the CT map for 7/14/2012 and the one for 7/14/2007 in the archives the difference in colors is striking. Much more of the ice cover is in the 60% range and 40% range. I think this is a non-numerical yet telling fact.

    maltose

    There are 17 days left until August 1st (day 212) and 1.1 million km^2 to get to 4 million km^2 by that date. That's 64k a day. At this rate, it's going to happen but there may be a slowdown at some point...

    Account Deleted

    Wow! Peterman crack!

    http://lance-modis.eosdis.nasa.gov/imagery/subsets/?subset=Arctic_r03c03.2012198.aqua.250m

    Espen Olsen

    Congratulations Arcticlost80; That piece of ice is +/- 110 - 120 km2. Bingo!

    Espen Olsen

    Or twice the size of Manhattan!

    Account Deleted

    Espen, thank Patrick Lockerby, he waited a long time past summer this calving.

    Espen Olsen

    It is like a virgin birth, I didn't see it come!

    Espen Olsen

    Yes Patrick, I have not heard form him lately, wonder if he is aware of this? It is at least half the size of the calving in 2010!

    Account Deleted

    This is the second largest calving Peterman glacier in the last 12 years since 2010.

    http://bprc.osu.edu/blogfiles/MODIS/2010/08/Petermann_2010_Effective_Length_Change_523x500.jpg

    michael sweet

    Canada Ice Service shows most of the ice in the Beaufort Sea as multi-year ice. They also show most of the ice on the edge of the pack in the Chukchi as being multi-year ice. Perhaps the ice is holding out in the Chukchi because of the old ice that moved in from the Beaufort Sea over the winter. How long will this old ice hold out with the very high sea temperatures in the Beaufort sea?

    Account Deleted

    Espen, I think Patrick was very upset that his predictions have not come true last summer. Because he stopped writing.

    Espen Olsen

    Some media from Greenland wants to interview me on this what is your name Arcticlost?

    Account Deleted

    Sorry, I do not want to publish your name. ( Our world is tough enough, and I have enemies.

    So you Espen can not talk about me :) It's in my best interest.

    Chris Biscan

    I was right again, the vortex has simply moved the ice in a more compacting fashion and is transporting heat into the basin off land.

    The 2012 season will finish 2.7-2.8 mil km2 on CT today.

    4.28 mil km2 on Jaxa

    and basically a record low tie on DMI.

    3300km3 on piomas


    and denialists will say 2012 wasn't as bad 2007.

    Paradoxnl.wordpress.com

    I found three definitions of Sea ice area and Sea ice extent on the internet:

    1) Definition on Wiki:
    To estimate ice area, scientists calculate the percentage of sea ice in each pixel, multiply by the pixel area, and total the amounts.
    To estimate ice extent, scientists set a threshold percentage, and count every pixel meeting or exceeding that threshold as “ice-covered.”
    The National Snow and Ice Data Center, one of NASA’s Distributed Active Archive Centers, monitors sea ice extent using a threshold of 15 percent/

    2) Definition on ijis website:
    The area of sea-ice cover is often defined in two ways, i.e., sea-ice "extent" and sea-ice "area."
    These multiple definitions of sea-ice cover may sometimes confuse data users.
    The former is defined as the areal sum of sea ice covering the ocean (sea ice + open ocean), whereas the latter "area" definition counts only sea ice covering a fraction of the ocean (sea ice only).
    Thus, the sea-ice extent is always larger than the sea-ice area.
    Because of the possible errors in SIC mentioned above, satellite-derived sea-ice concentration can be underestimated, particularly in summer.
    In such a case, the sea-ice area is more susceptible to errors than the sea-ice extent. T
    hus, we adopt the definition of sea-ice extent to monitor the variation of the Arctic sea ice on this site.

    3) Definition on NSIDC website:
    Extent is different from the total area in that if a given region has a percentage of ice concentration greater than the threshold, the entire region is considered "ice-covered."
    Total area tells how much of the region is actually covered by ice. Arctic- or antarctic-wide sea ice extent is always a larger number than area.

    My interpretation of the above is as follows:

    E.g. you have 3 pixels (3 squares). Each pixel (square) has a surface of e.g. 100 km2.
    Assume (the surface of) square 1 is ice covered for 8%
    Assume (the surface of) square 2 is ice covered for 10%
    Assume (the surface of) square 2 is ice covered for 16%

    Then, based on the above definitions:
    the total 15% ice extent for the three squares would be: 8 km2 (below threshold of 15%) + 10 km2 (below threshold of 15%) + 100 km2 = 118 km2
    the total ice area for the three squares would be: 8 km2 + 10 km2 + 16 km2 = 34 km2

    Is the above interpretation of the definitions correct or am I missing something?

    crandles

    >"The 2012 season will finish 2.7-2.8 mil km2 on CT today.
    4.28 mil km2 on Jaxa
    and basically a record low tie on DMI.
    3300km3 on piomas
    and denialists will say 2012 wasn't as bad 2007."

    Optimist.

    How do you reckon on PIOMAS 3300km3 when 30 June is 1100 lower than 30 June 11 and the 2011 minimum was 4017? We are going to recover 400Km^3 compared to 2011 when the area is lower causing more energy absorbtion? Perhaps you have good reason to think it will be a lot cloudier, or is it smokier from all the fires?


    crandles

    >"Is the above interpretation of the definitions correct or am I missing something?"

    Area is either 34km^2 as you said or more likely there is an extra wrinkle not explained that says concentration under 15% isn't counted for area making the answer 16km^2.

    Extent is 100km^2 (one cell of over 15%)


    If you want to go into a bit more detail: Suppose cell 1 reports ice concentration on 5 consecutive days as:
    35% 10% 16% 8% 4%
    I think NSIDC uses a 5 day average (35+10+16+8+4)/5=73/5=14.6 Therefore does not count for extent or area.

    IJIS uses a 2 day average so for first two days (35+10)/2 > 15 so extent = 100 and area =22.5km^2

    For second and third day (10+16)/2 <15 so no extent or area

    Similarly no extent or area for subsequent pairs of days.

    (I think.)

    Dave Leaton

    More statistical esoterica: all of the daily minimum area (CT) records have been set within the last 8 years. Fifty-two percent (191) have been set in the last four years. Forty-four have been set this year so far. 2007 contains 122, 2011 = 106, 2006 and 2012 are tied at 44.

    Chris Biscan

    As the ice pack get's down to 4.5 mil km2 the remaining ice is sitting with colder waters around it and the solar influence wanes and the thickest ice is left over.

    It just get's harder to melt out the ice that goes from 3M to 2.4M May to August.

    I don't think our climate can do that yet. We will see.

    If we get a wall to wall DPA in August then yeah, it will be worse than I said up above.

    Paradoxnl.wordpress.com

    Crandles, thank you!

    You wrote:

    "Area is either 34km^2 as you said or more likely there is an extra wrinkle not explained that says concentration under 15% isn't counted for area making the answer 16km^2.

    Extent is 100km^2 (one cell of over 15%)"

    => Yes, I think you're right that the extra wrinkle not explained is that concentration under 15% isn't counted for area (and extent)making.

    That indeed (probably) the correct answer is: Extent is 100km^2 (one cell of over 15%)(and area is 16 km^2).

    Your example how you think NSIDC and IJIS calculate extent and area makes sense to me! Thank you.

    crandles

    The complexity of the process isn't in how the 15% threshold is aplied, it is in how the brightness temperature measured by the satellites is converted to the ice concentration figures. I know snow, ice of various types and in various stages and water have different brightness temperatures and it is complex and error prone to convert the signal to an ice concentration figure but very little more of that problem.

    Seke Rob

    Stats Esoterica? Counting 56 days on the historic record that had a greater/equal negative anomaly than the -1.992 that CT is now showing for the Arctic. 57 of 12248 days is 0.465% having this state. The first of these to occur Aug.19,2007. The earliest ever was Aug.12, 2011 (may have already mentioned). Scary to be 1 month ahead of anything on the Area metric.

    http://bit.ly/CTAR01
    http://bit.ly/CTAR02

    That -1.992MKm^2 anomaly of day 195 puts it at 2.95 Std.Dev. (the light blue range) of second chart. The 365 day rolling mean anomaly is nearly hitting 2 standard deviation (The yellow area).

    Bob Wallace

    "As the ice pack get's down to 4.5 mil km2 the remaining ice is sitting with colder waters around it and the solar influence wanes and the thickest ice is left over.

    It just get's harder to melt out the ice that goes from 3M to 2.4M May to August."

    Doesn't that assume that the pack stays together and doesn't get spread around by storms?

    Plus the size of the Fram remains constant. Transport out rules.

    Chris Biscan

    Bob,


    Your right. I am just going on the idea that since 2007 we haven't gotten passed it.


    Now, of course that doesn't mean the ice pack isn't worse, all of of us here know 2012 is the worst it's been, like 2011 was the worst it's been, like 2010 was.


    However even with increased solar forcing, thinner ice, warmer rivers between 25-200m below the arctic ice, lower regional albedo's.

    The issue is the depth of cold water and thickness of ice along the CA-Greenland borders. That ice would have to be all thrusted out the Fram for the ice to drop to 3.5 mil km2 or lower.

    The GFS already shows a semi -DPA with heat crushing the russian side next week. Which would keep the MYI from flushing. Now we could get the ice line to 85N along half the ice pack by Sept, but i can't see much below 4.0 mil without a large DPA.

    crandles

    >"As the ice pack get's down to 4.5 mil km2 the remaining ice is sitting with colder waters around it"

    September certainly, second half of August quite possibly, ...

    But July, we are only just getting a decent area of water open to catch the insolation to warm them up. It just seems a bit soon to be talking about them cooling down yet.

    crandles

    We seem close on some measures eg for CT Area I have gone for 2.6 and I could be seen as alarmist so your 2.7-2.8 seems very reasonable.

    For extent I have gone for 4 (NSIDC vs record 4.3) so your 4.28 (very close to a record) is a slightly bigger gap and it certainly isn't implausible as intrade has record being less than 50% chance.

    But on PIOMAS our difference is rather large: 2200 vs 3300. Maybe I am just way too alarmist? Thats possible. However, I just don't see much reason to go for more than LY4000- last annual diff of 1100 = 2900.

    >" but i can't see much below 4.0 mil without a large DPA."
    Fair enough, I am with you there, it is just your PIOMAS figure that I struggle to reconcile.

    Chris Biscan

    I am not saying that it will be cooling down soon.

    I am saying this:

    The Arctic Ocean is shallow and deep. The thickest coldest part is over the deepest water. Which all sits mostly between 80-90N the highest latitudes.

    Also this being the oldest thickest ice, it has less brine, and it has a small melt period and zero bottom melt.

    So, it's likely we will end up with 4 mil km2, half of which is .5 meters on average the other half is 2 meters or more on average.

    eventually the climate will be able to melt that .5 meters and crash the ice, I can't see it today, unless a BOSS DPA comes.

    like 2 weeks in August at least with a 1025-1040HP over The Beaufort and a sub 1000mb slp over the laptev.

    Bob Wallace

    " That ice would have to be all thrusted out the Fram for the ice to drop to 3.5 mil km2 or lower."

    About every day I watch the ARC thickness 30 day gif.

    http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn_nowcast_anim30d.gif

    It's just my impression, but if there was less medium-thick ice crowding the Straight then a lot more of the thicker along the top of Greenland would be exiting.

    Melt out more of the 'closer to Europe' and you open the route for the thickest to exit.

    To my eyes, there's force from the left pushing on that red and yellow stuff. The thinner blue stuff is running a blocking game.

    (A drop to 3.5 mil km2 or lower, I don't know. All I'm suggesting is to not overlook transport and concentrate too much on melt.)

    AmbiValent

    I was under the impression that for Fram Strait to be blocked, the current would have to push the ice, only for it to run into Svalbard.

    But on the graphs (I especially look at the UB concentration map) the ice doesn't reach Svalbard. So it seems to be a case of the current sometimes pushing the ice towards Fram Strait, and sometimes keeping it almost in place, thereby limiting the amount of ice leaving through Fram Strait.

    crandles

    I am unsure of the effect of the deepest water so that could certainly be a factor.

    I think we have less volume of the MYI so that would be a factor towards melting more volume this year.

    >"So, it's likely we will end up with 4 mil km2, half of which is .5 meters on average the other half is 2 meters or more on average."

    Um., that would give 5000 km^3 whereas I thought we were arguing over 2200 or 3300 as our best estimate.

    For 30 June to minimum, volume declined by:
    2005 9.4 K Km^3
    2006 8.9
    2007 9.1
    2008 10.4
    2009 9.7
    2010 8.4
    2011 8.2

    For 2012 you are suggesting a decline of 7.9 K Km^3. Having already melted more volume before June 30 in later years there could well be a downward trend and 7.9 might well be reasonable.

    I not only used this measure but also used a couple of other more alarmist measures to guide my prediction which pushed my prediction upwards but maybe I shouldn't have used them despite the better fit they provided.

    I guess we will see in another 1.5 or 2.5 months.

    Werther

    Chris,
    Only r04c03 and r03c03 MODIS tiles show more or less unified ice in what I call a maze pattern. That’s ca. 2MK2. Healy and Polarstern slipped along this area last year, reporting a mean of 90cm thickness.
    The rest today is 5.5 MK2 patternless individual floes, clutched here and there by any low or high.
    NCEP today showed an area for -1.5dC near that of 9 August last year. I think you underestimate the latent heat balance, which is pretty much stuffed up in the last year. For me, that Dipole isn’t even necessary to get under 4 MK2 extent, 2.9 MK2 area and 3000 KM3 volume.
    We’ll see.

    Bob Wallace

    "I was under the impression that for Fram Strait to be blocked, the current would have to push the ice, only for it to run into Svalbard."

    Blocking might have been a poor word choice, I used it in the American football sense.

    Blue stuff shoving red and yellow stuff away from the opening in its rush to get out. Perhaps that works better?

    Matthew Opitz

    I have some questions about the "CICE Ice thickness Arctic Now-cast" product from the Navy.

    1. How accurate is it considered to be?

    2. Are its numbers in feet or meters?

    Something I have noticed is that there is a lot less turquoise ice this time around. Last year at this time, the turquoise ice stretched all the way up close to the ice edge. This year, there's a lot of blue and purple ice, especially on the Russian side of the Arctic.

    July 14, 2011:

    http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn/nowcast/ictn2011071518_2011071400_035_arcticictn.001.gif

    July 14, 2012:

    http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn/nowcast/ictn2012071518_2012071400_035_arcticictn.001.gif

    Matthew Opitz

    Ah, according to this:
    http://www.dtic.mil/dtic/tr/fulltext/u2/a533205.pdf
    The numbers for the CICE ice thickness product mentioned above are in meters.

    This brings me to another question:

    Would it be possible to create an algorithm that would integrate that CICE ice thickness data image, pixel by pixel, to determine overall ice volume?

    That would give us another daily ice volume measurement to compare against PIOMAS. And it would be a volume measurement where we could see exactly where the changes were occurring day-to-day.

    Bob Wallace

    Here's Matthew's two July 15 ARC images side by side.

    http://i619.photobucket.com/albums/tt275/Bob_Wall/ARCThicknessJuly1520112012.jpg?t=1342495981

    Werther

    Some developments on SIA/SIE:
    - The large floes in the Beaufort r05c02 are finally falling apart
    - Connection Beaufort-Chukchi is a matter of days/right winds
    - Loose floes in the Basin north of Frantsa Yosefa are dispersing

    Al Rodger

    Matthew Opitz,
    The ice modelling system you reference is a US Navy forecasting tool designed to allow them to drive their boats round the Arctic without crashing into unexpectedly large lumps of ice. There are people who have used the output of such models (The US Navy feeds of acronyms. Your reference names son-of-PIPS as ACNFS.) to graph Arctic ice volume but I am loathed to link here to such websites as they tend to be of the denialist ilk with such catchy post headlines as "PIPS shows Arctic Ice Thickening..." "Arctic Ice Volume Has Increased 25% since..." etc.

    As your link describes, these Navy systems are concerned with the location of the ice edge and wildly inaccurate when modelling thickness with which they always err on the thick side. "Both systems compare well against IMB ice growth trends (growing/melting), but each model over-predicts ice thickness by approximately one meter. In the comparison against airborne thickness survey data, ACNFS compared slightly better than PIPS 2.0 (0.31 m vs. 0.40 m)."
    Of course PIOMAS accuracy is not brilliant or unquestioned. However any volume calculated from PIPS may be interesting but, because modelling thickness is not a primary concern of the system, such calculations will remain stongly questionable.

    Chris Biscan

    Jaxa has a final of roughly 118,000km2 on the 16th and a prelim of 153,000km2 today.

    DMI plummeted again.

    I expected drops but the vortex is just helping flush the ice and bring warmer winds into the the weak ice.


    The models are already backing down on the extent of the vortex cold.


    The Euro almost goes to a DPA with a torch on the Russian side later on.

    But it will change.

    But still these are big losses.

    Daniel Bailey

    @ Al Rodger

    Having made navigational products for the US Navy, I can assure you that while the nautical cartographers make the most accurate products possible, there is a certain "fudge factor" built into the products they make.

    When you are dealing with billion-dollar-plus units of hardware (not even factoring the human costs into the equation), no one wants to be known as "that guy" who had a super carrier or fleet ballistic missile submarine go down with all hands lost while using one of their products.

    The ice thickness products have a large degree of relative accuracies and therefore have a great deal of precision for their intended use by naval commanders. Their absolute accuracies are not what Arctic sea ice blog watchers would like them to be.

    Still, they are a useful tool.

    idunno

    Hi all,

    CT has fallen only 57k overnight, well shy of the 5M mark @ 5.046, and below the 2M anomaly, @1.964.

    crandles

    Latest estimates using data to 15th July:

    NSIDC extent minimum now projected at 3.76M Km^2 with a standard error of 0.32 so this is 1.69 standard errors into record teritory.

    Cryosphere Today area projected at 2.38 M Km^2 with a standard error of .23 so this is 2.28 standard errors into record teritory.

    Actual errors in practice are likely to be greater than these standard errors so I doubt it is sensible to claim it is over 95% certain we will get a CT area record this year.

    Bob Wallace

    If the ARCs thickness projections are close to right we're going to see a higher percentage of 2.5 - 3 meter ice heading out the Fram over the next few days. (As opposed to the more normal 2 - 2.5 meter.)

    They are predicting a higher flow from along the north side of Greenland.

    Account Deleted

    Idunno, I think the important thing is the Arctic Basin anomaly is already at 0.5 million km^2 and SIA in the AB is down to 3.1 ish million in the middle of July. We have something like 400K km^2 less in this area compared to last year.

    Bob Wallace

    Colin, that's Arctic Basin area. I'm guessing AB and Greenland Sea area and extent are the least meaningful statistics at this point in annual melt. These are areas that get most replenished from other areas. (Do I have that right?)

    If you look at the area graphs you can see recent times when amounts grew, even this far into the melt season.

    I think the most important measurement for AB and GS would be average thickness. Has the older, thicker ice gone away and been replaced by thinner ice?

    Account Deleted

    Hi Bob, my point is there is an area of 0.5 million Km^2 of open water in the AB that is able to absorb sunlight and that extra energy in the AB is likely to have an impact later in the melt season.

    Account Deleted

    As an aside Buoy 2012D is starting to show signs of bottom melt.
    http://imb.crrel.usace.army.mil/2012D.htm

    Curranstef

    Hello everybody -great blog Neven...good job everyone; especially with the recent Petermen calving- really felt like we were part of the story.

    Anyways http://weatherspark.com/#!dashboard;ws=28294 is saying that Alert in Nanavuk
    (humanity's most northerly permanent settlement)
    is @ 20C @ midday.

    The record for Alert is also 20C.

    I can only imagine what the scientists are doing up there at the moment to commemorate this occasion.

    Good luck everyone.
    Stephen

    me.yahoo.com/a/nSjChi4X3vr8X3DRw93GkY1.cerja.8nvWk-

    Bob:"Colin, that's Arctic Basin area. I'm guessing AB and Greenland Sea area and extent are the least meaningful statistics at this point in annual melt. These are areas that get most replenished from other areas. (Do I have that right?)

    If you look at the area graphs you can see recent times when amounts grew, even this far into the melt season."

    That's true but if you look at the surrounding areas you'll see there's not much left to replenish with, E Siberian sea and Chukchi are all that's left, and not much there either.

    Phil.

    idunno

    It seems to me that the Arctic Basin is critical. This plot:

    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.1.html

    In order for the total sea ice area to hit a new record low, then given that there will be some small patches of ice left in areas such as the Canadian Archipelago and the Greenland Sea, the total in the AB needs to go down to closer to 2 million than 2.5 million.

    So far, the AB has declined in area from approx 4.2M at maximum, to around 3.2M today - a loss of 1M.

    Of this, very little, as far as I can see, is made up of open water...

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

    ...comparing the main map with the CT areas map below, it would seem that there is a sliver of open water North of Svarlsbard, a small chunk North of the Laptev Sea and a tiny bite at the exit of the main channel through the Canadian Archipelago. Total = way less than the 25% supposed to be lost since the maximum.

    So the rest of this lost ice must be accounted for by the many areas of low 60%-75% concentration ice.

    I do hate to have to opine that I don't have total faith in the ability of the satellite to correctly estimate the percentage of ice concentration, but I don't. It seems to me that it's much better at drawing its outline than at colouring in.

    Occasionally this results in clear misatkes, such as that seen previously in the Canadian Archipelago:

    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.12.html

    ...and which can mean that the Arctic Basin graph wobbles up and down a lot. Over the last couple of days, for example, it has developed a small uptick, and the overall rate of ice loss throughout the Arctic has stalled somewhat. This is clearly not caused by the Arctic Basin refreezing!

    Alors, the AB has currently lost somewhere around 1M km2 of ice cover. If it loses a further 1M by around the end of August, we will have a new minimum area.

    Faites vos jeux!

    Jon Torrance

    In order for the total sea ice area to hit a new record low, then given that there will be some small patches of ice left in areas such as the Canadian Archipelago and the Greenland Sea, the total in the AB needs to go down to closer to 2 million than 2.5 million.

    Looking at your first link, it appears the current record low from last year was set without the total in the AB going below ~2.4 million so it's not clear why you think something less than 2.4 million but closer to 2.5 million than 2 million won't suffice.

    idunno

    Hi Jon,

    That would be because I have made, for the first time ever, honest, a dumb-ass mistake, as I thought that the record low was set in 2007. (For which year, I don't have any data for the AB alone) Mea culpa.

    Account Deleted

    Idunno, While I agree that the area estimates have to be taken with a degree of caution. The AB graph from CT suggest that the state of the ice is worse off than last years (either more melt ponds or small leads/opening). If this is due to an increase in small opening - then it isn't good news for the AB ice pack. I came across an EGU abstract that suggest that relatively small changes in SST (i.e. water temps of between -0.9 to 0.5 C) are associated with enhanced bottom melt in August (NB: The study and the abstract have a number of issues). I also came across a translation of a russian site that suggest there is still a fair amount of solar radiation hitting the AB in July/August

    http://meetingorganizer.copernicus.org/EGU2012/EGU2012-13880.pdf

    http://www.aari.ru/resources/m0001/Meteorology/HTML/DATA/DOCUMENTATION/radiation_marsh_eng.htm

    Neven - I am a Typepad luddite - so not sure how to shorten the links.

    Daniel Bailey

    Colin, some html posting tips are located at this page at Skeptical Science. Most will work here at Neven's.

    Kris

    Just back from holidays guasted by quite an obstacle, so it took me a while to opdate the arctic 1 July parade .

    Phil263

    On all extent graphs, (NSIDC, IJIS, NOSREX) 2012 seems to be tacking the lowest extent years, 2010,11,even 2007. However the visuals from UniBremen tell a different story. Have a look here at 2011 , and 2010 .
    The Beaufort Sea in both years was largely clear and in 2011, the Northern Route was almost open. This year, the North coast of America still has some way to go before it is completely ice free and the Northern route is nowhere near navigable. In fact, even in 2009 , even though the CAB definitely had a higher concentration, the North American coast was practically ice free at this date.

    crandles

    CT Area anomaly of over 2m

    4.862 a drop of .175

    Seke Rob

    CT Anomaly greater than -2 Million

    Werther

    Phil,
    The ’12-’11 difference is an optical illusion. It stems from extent in the Chukchi region.
    Extent in ’12 is less in the EGS, Barentsz, Kara, Laptev and ESAS. More in Chukchi and CAA.
    Still, ’11 leads with a slight margin. But the quality of the pack in ’12 is worse.
    Basis: CAD on UB 18072012

    Neven

    CT Anomaly greater than -2 Million

    Post is up, ladies and gentlemen.

    Phil (and others), remember the Concentration Maps page on the ASI Graphs website. Very handy for comparing 2005-2012 all at the same time.

    Seke Rob

    MASIE let loose of 244KKm square on the 18th and is with this again below same day of 2011. Here's a trial animation: http://bit.ly/MASIEA
    (FF, Chrome, Opera users can see, IE users only see the static latest date. Hit Esc key to stop).

    JAXA/MASIE/DMI blimp displaying the first week step change: http://bit.ly/MASDMI JAXA needs to hurry, else MASIE could overtake on the 15% concentration measure. DMI 30% concentration is probably at an in-between point... that what is just over >30% in limbo.

    Neven

    Looking at the weather forecasts, those lows are now moving in over the Canadian Archipelago, in 3-5 days from now. If this comes about, we will surely see a serious stalling of extent and area decrease?

    Twemoran

    Neven

    It certainly brought the melt season to a halt when it struck the Siberian side ;(

    The low seems to be acting like a gigantic circular saw - tearing everything up in it's path, then spitting the shards out through Fram. 3-5 more days of this and all the minimum guesses are out the window.

    Arctic Summer Storms - ASS, may kick all our predictions out the window.

    Terry

    Kris

    Looks like the Barrow webcam has been moved to a new address:

    http://feeder.gina.alaska.edu/webcam-uaf-barrow-seaice-images/current/image

    Incidentally, to me it looks like the ice concentration is fas less as 30 % as the Bremen map is suggesting.

    Ned Ward

    It's mildly remarkable to see how closely 2012 is tracking 2007 right now in IJIS.

    Here's a pair of graphs comparing July 2007 and 2012, one with the raw extent numbers and one with the anomalies.

    2007-2012 comparison graphs

    I'm sure it's just coincidence, but it's a bit spooky how similar the shapes and positions of the anomaly graphs are.

    Ned Ward

    Eh. Actually, most of the similarity is due to the overall downward trend in extent, and the fact that 2007 and 2012 both started out in a similar position (anomaly-wise) at the beginning of the month. So the apparent similarity is mostly an artifact of temporal autocorrelation.

    Comparing the 2007 and 2012 daily drops in extent, the parallelism is not so striking, except for the last three days.

    So ... feel free to ignore my previous comment.

    Ned Ward

    Last year on July 15 I posted a "prediction" for the September IJIS daily minimum of 4.47 million km2 ... which was only 1.2% off from the actual minimum (4.53).

    Pretty good, huh? Actually, it was mostly luck. The "prediction" was just based on taking the current value (or an average of recent values) and subtracting an estimate of the remaining decrease-to-minimum based on the previous years' data.

    If we repeat that exercise now, using just the post-2007 years to predict the expected decline from this point, I get the following:

    Predicted IJIS daily minimum: 4.40 million km2
    Predicted IJIS September mean: 4.58 million km2
    Predicted NSIDC September mean: 4.49 million km2

    The confidence intervals around those are still very wide. I'm guessing that IJIS will come in second, between 2007 and 2011, but it could easily break the 2007 record ... or end up in fifth place, behind 2007, 2008, 2010, and 2011.

    Peter Ellis

    There's no point comparing the webcam pictures to Bremen, or even to MODIS. The webcam field of view is a couple of hundred metres at most. That's one pixel in the highest-resolution MODIS picture, and 1/30 of a pixel in the Bremen data.

    This planet is BIG.

    Werther

    Neven, on the lows yesterday,
    On ECMWF SLP I see mainly lows above the polar circle, yes, for next few days. But not specifically over the CAA. There’s no dipole, true, but I don’t think that will be the necessary asset to get under 2007.

    I see several configurations on ECMWF that may clutch the pack FI on the vulnerable Atlantic side. In the end of the time range (10 days), it all boils down into renewed high pressure over the Bering side.
    It doesn’t look like much transport through Fram Strait, and the 500 hPa level shows a vortex moving to the CAA, yes. But there’s not much cold on 850 hPa.

    My take is continuing melt in situ. Not the nosedive but steady down. Which is most in volume, extent is holding on through dispersion. Will it last under late summer cloud cover?
    Oh yes, today the fast ice on the Joekelbugt NE Greenland is breaking up.
    And, under the clouds, Vilkitsky Strait seems to be opening…

    idunno

    See also spectacular cracks on the section of DMI Greenland maps called "Nord". Related to the syrgy szyr spring tides?

    Neven

    On ECMWF SLP I see mainly lows above the polar circle, yes, for next few days. But not specifically over the CAA.

    You're right. The weather forecasts change a lot from one day to the next.

    DrTskoul

    Interesting figure of extent decline rate :

    Day 201 - MASIE decline rate sq.km/day

    The extent decline rate from day-to-day data, has been accelerating since March.

    Regional MASIE data - Day 201 Update

    Kalle GZ

    Petroleum News posted that Shell had to delay drilling because, despite how low the ice in the Arctic is, the Chukchi Sea managed to stay near normal, as shows here:

    http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.10.html

    Post by Petroleum News: http://www.petroleumnews.com/pntruncate/851006337.shtml

    Kalle GZ

    The ice is still melting fast even under cloudy weather conditions. I think this shows that the Arctic has a dark future ahead. The ice must really be thin and weak out there that can melt really easily. At this rate, one year with weather conditions similar to 2007 and I think we may see an extent minimum around 3 million square kilometers.

    Kris

    Kalle GZ wrote:

    Petroleum News posted

    Hmm. OTOH, the chuckchi's extend is below average, so there isn't much of difference with previous years, if there is at all. These guys aren't used to tell the truth, so we might wonder if perhaps something else could have popped up.

    idunno

    Hi all,

    A wide-ranging piece from the Guardian, doubtless coming soon to a side-bar near here:

    http://www.guardian.co.uk/world/2012/jul/22/arctic-ice-melting-oil-drilling

    Account Deleted

    Looks like we will be crossing the 3 million Km^2 mark for sea ice area in the Arctic Basin in the next day or so on CT, which appears roughly 10 days earlier than last year. What with the export out of the AB to the Baffin/Newfoundland, Greenland and Beaufort seas and horrible state of the remaining ice in the AB - the 2.5 million km^2 mark isn't that far off.

    Werther

    Day 206…
    The 3,6 MKm2 area of loose floes is starting to show this foamy look it had last year, first week of August. The red/purple on UB in the Chukchi is like that foam. High concentration slush freckled with small, thicker floes. Last year in that region it took day 216-226 to finish that off.
    It looks feasible now that, given 2012 has a ten days lead in the process on 2011, extent will equally pass 2011 with 10 x 25K (?) at least.
    While we still see 7,1 MKm2 on the extent charts, the drop in the Chukchi/ESAS/Basin is bound to be staggering in the next two weeks. I expect some of the largest daily drops on extent ever for the date.
    The path to 5 MKm2 around 10 August and record minimum a month later is wide open.

    Kris

    Werther wrote

    High concentration slush freckled with small, thicker floes

    Presumely next stept will be a large ice floe island spanning the East Siberian and Chucksi seas. Tends to be "an habit" ever since 2009.

    Kris

    If you navigate from

    http://psc.apl.washington.edu/northpole/NPEO2012/10.jpg over NPEO2012/11.jpg, NPEO2012/12.jpg ... etc... to http://psc.apl.washington.edu/northpole/NPEO2012/18.jpg you can see the disappering act of the "ice island" just in front of the fore left surveyor's rod, as well as some other "islands" too.
    And as a sweet some rain too at NPEO2012/18.jpg

    Bfraser

    New update on Artic Sea Ice News:

    http://nsidc.org/arcticseaicenews/

    Matthew Opitz

    I want to ask everyone: even without the right winds for compacting ice, what are the chances that all of the ice except for the stronghold area outlined in black in the following image ends up melting in place by the end of the melting season?

    http://i42.photobucket.com/albums/e326/Zeiter/arctic_SSMIS_nic2.png

    Also, if this came to pass, would this not amount to a new record for extent and area?

    Account Deleted

    I think that if anything you are too optimistic . I anticipate a greater loss on the Beaufort sea edge and much more on the Laptev sea edge of the central Arctic. I see a distinct possibility of open water to beyond the 85th parallel in the latter area.

    I also think that the North West passage is now fully broken up judging by the glimpses seen peeping through the clouds on MODIS.

    It is my firm opinion that unless there is a dramatic weather event we will see a new record minimum this year. Followed by a new minimum again in about 5 or 6 years with a nominal (extent/area but not volume) "recovery" in between. The 5/6 years oscillation with a downward trend in the minimum is obvious to my eyes.

    The comments to this entry are closed.