Support the Arctic Sea Ice Forum and Blog

Author Topic: 2018 sea ice area and extent data  (Read 522336 times)

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1050 on: July 15, 2018, 11:33:26 AM »
Hudson Bay melt - I am even more confused

University of Bremen 14 July Image says the ice is still gone, as it was on the 13 July.

But here is an SMOS from the University of Bremen posted by Hyperion on the SMOS thread for the 13 July saying the ice was not gone.

But when I go to https://seaice.uni-bremen.de/databrowser/#p=smos I see: "No data during Summer".

What is going on? My brain hurts.
 
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

Viggy

  • New ice
  • Posts: 60
    • View Profile
  • Liked: 22
  • Likes Given: 19
Re: 2018 sea ice area and extent data
« Reply #1051 on: July 15, 2018, 11:34:08 AM »
Canadian Ice Service's daily maps definitely shows all that Hudson Bay sea ice still there.

https://www.canada.ca/en/environment-climate-change/services/ice-forecasts-observations/latest-conditions.html


gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1052 on: July 15, 2018, 11:37:58 AM »
Canadian Ice Service's daily maps definitely shows all that Hudson Bay sea ice still there.


Now that I can believe. Good job, Viggy - data rules, OK.
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

Viggy

  • New ice
  • Posts: 60
    • View Profile
  • Liked: 22
  • Likes Given: 19
Re: 2018 sea ice area and extent data
« Reply #1053 on: July 15, 2018, 11:47:14 AM »
Thanks Geron!

They also have a great 10 day ice concentration map that I like to keep tabs on.

https://www.canada.ca/en/environment-climate-change/services/ice-forecasts-observations/latest-conditions/concentration-map-last-10-days.html

Click image to play

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1054 on: July 15, 2018, 11:59:59 AM »
They also have a great 10 day ice concentration map that I like to keep tabs on.
Thanks, Viggy. I've bookmarked it.
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

Richard Rathbone

  • Grease ice
  • Posts: 533
    • View Profile
  • Liked: 39
  • Likes Given: 7
Re: 2018 sea ice area and extent data
« Reply #1055 on: July 15, 2018, 01:40:38 PM »
The Hudson ice is still there on MASIE too, which is the NSIDC best measure of where the ice is now.

Satellite microwave measurements of ice concentration have quite large uncertainties at the lower end, which is why the 15% cutoff has to be that high to avoid open ocean being measured as ice from time to time. The tradeoff for not seeing open water as extent is that  25% ice gets measured as open water from time to time and if you suspect that is happening, MASIE is a place to check.

https://nsidc.org/data/masie


Ned W

  • Guest
Re: 2018 sea ice area and extent data
« Reply #1056 on: July 15, 2018, 01:41:45 PM »
Thanks Geron!

They also have a great 10 day ice concentration map that I like to keep tabs on.

https://www.canada.ca/en/environment-climate-change/services/ice-forecasts-observations/latest-conditions/concentration-map-last-10-days.html

Here is the Canada ice concentration chart for 14 July.  Red areas are over 90% concentration (and elsewhere are described as "thick first-year ice").  The chart is based on SAR imagery from Radarsat-2 and Sentinel-1A/B:



And here's the same thing, but with the AMSR2 ice concentration data from Uni Bremen overlaid.  Purple patches are any grid cells where AMSR2 shows at least 10% ice concentration:



In other words, there is a huge area of the Bay that is currently 90-100% concentration, but is showing up as < 10% concentration in the AMSR2 sea ice maps. 

This very extensive ice can be clearly seen in the Sentinel-1A SAR imagery that I posted yesterday.  Here's an enlargement showing the ice edge in west-central Hudson Bay:



Again, the purple patches are the only locations Uni Bremen's AMSR2 map is reporting any ice concentration above 10%.  Here's the same area, with the Canada ice charts overlaid:



Unsurprisingly, the ice chart shows the ice edge exactly where it can be seen on the Sentinel-1A cross-polarized imagery.  The orange area ("F") is over 70% concentration, and the red areas (i.e., most of this enlargement) are over 90% concentration. 

So, again, AMSR2 is failing to see most of the ice in Hudson Bay right now.

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1057 on: July 15, 2018, 02:43:31 PM »
Fascinating discussion about Hudson Bay over the last couple of days. Smashing images and a super demo of how there are various ways an oddity (this time ANSR2 data) can be tested from different sources. Which makes me more confident about this posting.

ps: If you are just waking up in your part of the world, have a good look at the last twenty posts or so on this thread. I am tempted to save them all..

NSIDC Total Area as at 14 July (5 day trailing average = 5,974,268 km2
This is up a mere 8 k up to 61 k above the 2010-2017 average


Total Area loss 96K, Central Seas 41 k, Periphery loss 29 k, Other Seas loss 26 k  
Analysis of individual seas.

Pacific Side
- The Okhotsk Sea area is 5 k,
- The Bering Sea area is 3 k,
- Chukchi Sea loss 10k,
- Beaufort Sea loss 2 k,
Small Area loss on the day.
Atlantic Side
- Baffin Sea loss 19 k,
- Greenland Sea loss 7 k,
- Barents Sea loss 2 k - Barents Sea area is 12 k, i.e. melting finished for all practical purposes.
- The Kara Sea area loss 24 k.
- The Laptev Sea area loss 14 k.
CAB
- The Central Arctic Sea gain 30 k,
- The Canadian Archipelago loss  6 k,
- East Siberian Sea loss 14 k .
Other seas
- St Lawrence area at 1 k,
- Hudson Bay area loss 26 k. Not 142k.

Daily area loss 14th July at 2010's average.

Extent loss continues to catch up fast to area loss, as shown very well on the attached graph. Will area losses also accelerate ? (JAXA data - where are you ?)
« Last Edit: July 15, 2018, 03:38:57 PM by gerontocrat »
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

A-Team

  • Nilas ice
  • Posts: 2452
    • View Profile
  • Liked: 304
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1058 on: July 15, 2018, 04:05:44 PM »
The data situation has gotten worse, from loss of some swaths over Hudson Bay to loss of all coverage. Ascat is not showing flash melt though it is not suited for this task. (Black swath gaps are removed from Ascat B of July 14th, allowing Ascat A to supplement its view of HB).
« Last Edit: July 15, 2018, 04:44:23 PM by A-Team »

Steven

  • Grease ice
  • Posts: 555
    • View Profile
  • Liked: 133
  • Likes Given: 16
Re: 2018 sea ice area and extent data
« Reply #1059 on: July 15, 2018, 04:20:45 PM »
NSIDC Arctic sea ice area anomaly has been plateauing during the past week:





Looking at the daily regional NSIDC data on Wipneus' website, most of the area loss over the past week was in Hudson (148k), Kara (129k), Baffin (115k) and East Siberian Sea (60k) :

https://sites.google.com/site/arctischepinguin/home/sea-ice-extent-area/grf/nsidc-nt-regional-area-overview.png

Ned W

  • Guest
Re: 2018 sea ice area and extent data
« Reply #1060 on: July 15, 2018, 07:43:05 PM »
All of the passive microwave-based extent metrics (NSIDC, UH, Bremen, ...) are (temporarily) exaggerating the loss of ice in Hudson Bay by roughly 150,000 km2.  Today's ScanSAR imagery from Sentinel-1 continues to show vast expanses of ice that are being missed by the passive microwave systems.  The manually interpreted products (MASIE, Canada ice charts) do show the ice correctly, so there is now suddenly a big gap between the two types of data.

As an example, here's MASIE vs Wipneus's AMSR-2 data for Hudson Bay, over the past two weeks:



What matters is not so much the vertical offset as the difference between the two:



The MASIE/AMSR-2 difference jumped upward by 150,000 km2 over the past two days.  The Canadian ice charts (and visual inspection of today's Sentinel imagery) confirm that the "jump" (or "ice going poof") was an artifact.  The ice is still there (and it's actually a pretty continuous pack over most of the east-central portion of the Bay).

-----------------------------------

Edited to add:

One more chart.  Here's the MASIE vs Wipneus AMSR-2 extent difference, normalized to the average of the differences for each basin during the month of June.  The Hudson Bay artifact is obvious (purple line that jumps up on 13-14 July). 



« Last Edit: July 15, 2018, 08:08:58 PM by Ned W »

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1061 on: July 15, 2018, 08:20:04 PM »
In case JAXA is in real trouble I have been getting the NSIDC Regional Extent Graphs in order (5 day trailing average again).

Just as well, as I found two errors that understated the sum of all the regional tables. The first graph shows a corrected "area minus extent" calculation - the same story except the 2018 lag between area and extent loss is greater. But it also shows that extent is now catching up a bit, as is obvious from the first table below.
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

A-Team

  • Nilas ice
  • Posts: 2452
    • View Profile
  • Liked: 304
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1062 on: July 15, 2018, 09:52:07 PM »
Quote
All of the passive microwave-based extent metrics (NSIDC, UH, Bremen, ...) are (temporarily) exaggerating the loss of ice in Hudson Bay
Maybe give the instruments a chance to get their data down to the receiving stations, this is Day 3 of malfunction on low latitude swaths?

Meanwhile, Modis/Terra just got a decent shot in the visible of Hudson Bay ice or whatever you want to call it. (Cloud cover is not too big an issue in the scene center.)
« Last Edit: July 15, 2018, 10:02:35 PM by A-Team »

Ned W

  • Guest
Re: 2018 sea ice area and extent data
« Reply #1063 on: July 15, 2018, 10:14:48 PM »
Nice pictures of the ice, A-Team.  But I'm not following your explanation that there's a problem with the ground stations for AMSR-2.  I mean, people who use AMSR-2 data seem to be finding it (for example:  http://images.remss.com/amsr/amsr2_data_daily.html ).

Apologies if there is something obvious that I'm missing. I had been assuming the Hudson Bay issue was just a combination of dense cloud cover, melt ponding, and other "environmental" factors that make the passive microwave algorithms unreliable.

JayW

  • Grease ice
  • Posts: 506
    • View Profile
  • Liked: 131
  • Likes Given: 182
"To defy the laws of tradition, is a crusade only of the brave" - Les Claypool

Tor Bejnar

  • Young ice
  • Posts: 3112
    • View Profile
  • Liked: 405
  • Likes Given: 200
Re: 2018 sea ice area and extent data
« Reply #1065 on: July 15, 2018, 10:52:51 PM »
Today's PolarView image of central Hudson Bay.  (I think of the same area in A-Team's images.)  There are details that I cannot read, but it's almost a 'picture of the day'.
Edit: the first image shows the 'yellow box' for location.  Each succeeding image is of the same area (including the protruding bit of ice) only enlarged more.  The last image is at "100%"  (and I thought to capture the caption that said so).
« Last Edit: July 16, 2018, 02:01:19 AM by Tor Bejnar »
Arctic ice is healthy for children and other living things.

A-Team

  • Nilas ice
  • Posts: 2452
    • View Profile
  • Liked: 304
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1066 on: July 16, 2018, 01:30:31 AM »
Quote
been assuming the Hudson Bay issue was just a combination of dense cloud cover, melt ponding, and other "environmental" factors that make the passive microwave algorithms unreliable.

Well those too are probably contributing, awkward ice this time of summer. But look at the UH AMSR2 product I posted up-forum. Solid unnatural block of white in HB and to north, a strictly local effect. Never seen anything like it in thousands before (though 1-2 day glitches are not unusual, as are polygonal swath gaps). The 12th looks ok but with odd faint slanting striations (scan lines?). The 15th should be up shortly if ever. Jaxa is up and running but way overdue, not serving anything past the swath-damaged 13th. So not their server or electrical system glitches.

ftp://ftp-projects.cen.uni-hamburg.de/seaice/AMSR2/3.125km/Arc_20180714_res3.125_LARGE.png
ftp://ftp-projects.cen.uni-hamburg.de/seaice/AMSR2/3.125km/Arc_20180713_res3.125_LARGE.png
ftp://ftp-projects.cen.uni-hamburg.de/seaice/AMSR2/3.125km/Arc_20180712_res3.125_LARGE.png

Nice, Tor. The middle frame is definitely a radar view of the previous visible wavelength ice. The Sentinels have great time stamps, Terra's can be dug out too but not at WorldView per se.

[Edit: just in, July 15th 3.125 km UH AMSR2. Still seeing data corruption at lower latitude swaths (gray), HB as well as Bering Sea. The bad swath just west of Baffin Bay is not quite identical to the bad swath in Jaxa RGB for the 13th. The abrupt white in the southern HB is not right but some of the middle may be ok.]
« Last Edit: July 16, 2018, 04:21:56 AM by A-Team »

Juan C. García

  • Nilas ice
  • Posts: 1532
    • View Profile
  • Liked: 546
  • Likes Given: 517
Re: 2018 sea ice area and extent data
« Reply #1067 on: July 16, 2018, 05:10:59 AM »
Well, there was really a sensor problem in AMSR-2...
[Images from Uni Bremen AMSR-2]

And it seems that we will have to wait another day to have ADS-NIPR (JAXA) data.
Japan has a national holiday on Monday. Let's just hope it's not a long weekend outage.
« Last Edit: July 16, 2018, 05:50:21 AM by Juan C. García »
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4009
    • View Profile
    • Arctische Pinguin
  • Liked: 644
  • Likes Given: 0
Re: 2018 sea ice area and extent data
« Reply #1068 on: July 16, 2018, 08:24:09 AM »

Well those too are probably contributing, awkward ice this time of summer. But look at the UH AMSR2 product I posted up-forum. Solid unnatural block of white in HB and to north, a strictly local effect. Never seen anything like it in thousands before

I find them quite common at low latitudes, St. Lawrence region if you want to look for them, other years in Hudson. It looks like a loss of precision, concentration values binned at exactly 100/n % (n=1-7) in UH's product, larger than usual pixel noise in that from Bremen. Also no doubt that the low number of swaths (due to the low latitude) is cause.

Quote
(though 1-2 day glitches are not unusual, as are polygonal swath gaps). The 12th looks ok but with odd faint slanting striations (scan lines?). The 15th should be up shortly if ever. Jaxa is up and running but way overdue, not serving anything past the swath-damaged 13th.

That would be ADS, what ever their relationship with Jaxa is, I have not noticed anything missing from Jaxa ( as available on server addresses *.jaxa.jp). ADS (ads.nipr.ac.jp) is the more unreliable.

Quote

[Edit: just in, July 15th 3.125 km UH AMSR2. Still seeing data corruption at lower latitude swaths (gray), HB as well as Bering Sea.

That would be because the 15th was not complete (not to mention the delay between the sattelite pass and the time it was received on Uni Hamburgs's server). UH seems to update a particular day three times, the first one has always missing swaths the second is mostly complete. Today the second update (updated about 4.30 GMT) is complete.

And the un-flash did happen, about +100k in Uni Hamburgs product.

oren

  • Young ice
  • Posts: 4536
    • View Profile
  • Liked: 912
  • Likes Given: 1307
Re: 2018 sea ice area and extent data
« Reply #1069 on: July 16, 2018, 08:45:15 AM »
Thanks Wipneus for clarifying everything. I hope UH correct the data retroactively to set the record straight.

Alexander555

  • Grease ice
  • Posts: 822
    • View Profile
  • Liked: 42
  • Likes Given: 34
Re: 2018 sea ice area and extent data
« Reply #1070 on: July 16, 2018, 12:53:36 PM »
What was the NSIDC minimum area for 2017 ?

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4009
    • View Profile
    • Arctische Pinguin
  • Liked: 644
  • Likes Given: 0
Re: 2018 sea ice area and extent data
« Reply #1071 on: July 16, 2018, 01:39:08 PM »
What was the NSIDC minimum area for 2017 ?

minimum NSIDC extent/area with date below.

Note that 2018 area is still last in ranking, but  only 200k above 1980 annual minimum, that could change any day now.

NH Extent
1 2012-09-16 3.339905
2 2016-09-07 4.144993
3 2007-09-14 4.146931
4 2011-09-08 4.332572
5 2015-09-08 4.387092
6 2008-09-18 4.548265
7 2010-09-19 4.589683
8 2017-09-12 4.611109
9 2014-09-16 4.988244
10 2013-09-13 5.039617
11 2009-09-12 5.046571
12 2005-09-20 5.314284
13 2002-09-18 5.625038
14 1999-09-12 5.676431
15 2006-09-14 5.746404
16 2004-09-18 5.770495
17 2000-09-11 5.942586
18 2003-09-17 5.969052
19 1990-09-21 6.010841
20 1995-09-04 6.012446
21 1993-09-13 6.161196
22 1991-09-16 6.258512
23 1998-09-17 6.293315
24 1984-09-16 6.396239
25 1985-09-09 6.486437
26 2001-09-19 6.567265
27 1997-09-03 6.601921
28 1989-09-22 6.888294
29 1987-09-02 6.889959
30 1979-09-21 6.895215
31 1981-09-10 6.901518
32 1994-09-05 6.933656
33 1988-09-11 7.048098
34 1986-09-06 7.121551
35 1996-09-10 7.146860
36 1992-09-07 7.158961
37 1982-09-13 7.159853
38 1983-09-08 7.204490
39 1980-09-05 7.533277
40 2018-07-14 8.315115
NH Area
1 2012-09-12 2.228429
2 2016-09-06 2.463209
3 2011-09-05 2.916451
4 2017-09-11 2.939311
5 2007-09-07 2.954993
6 2010-09-07 3.071674
7 2008-09-08 3.079552
8 2015-09-07 3.139137
9 2009-09-08 3.547109
10 2014-09-09 3.576643
11 2013-09-08 3.607877
12 2006-09-22 4.082958
13 2002-09-05 4.098445
14 2005-09-03 4.134433
15 2003-09-05 4.155104
16 2000-09-03 4.235279
17 1999-09-16 4.318875
18 1998-09-11 4.346996
19 2004-09-07 4.375240
20 1995-09-29 4.410301
21 1991-09-18 4.463582
22 1993-09-06 4.488183
23 2001-09-16 4.592484
24 1990-09-21 4.661325
25 1984-09-13 4.691219
26 1994-09-05 4.854227
27 1989-09-11 4.872464
28 1997-09-17 5.008589
29 1985-09-10 5.043057
30 1981-09-07 5.068809
31 1992-08-31 5.084517
32 1982-09-12 5.112870
33 1988-09-03 5.209891
34 1996-09-08 5.297000
35 1987-09-02 5.339545
36 1983-09-02 5.357635
37 1979-09-19 5.413227
38 1986-09-11 5.495114
39 1980-09-05 5.588796
40 2018-07-14 5.767418

Ned W

  • Guest
Re: 2018 sea ice area and extent data
« Reply #1072 on: July 16, 2018, 02:12:50 PM »
OK, here's a comparison of today's Sentinel-1 imagery (HV polarization) with yesterday's and today's Bremen AMSR-2 ice maps, for central Hudson Bay. 

Purple blobs = AMSR-2 ice map for 14 July
Other colors (red, yellow, green, blue) = AMSR-2 ice map for 15 July



« Last Edit: July 16, 2018, 02:37:46 PM by Ned W »

Archimid

  • Nilas ice
  • Posts: 2408
    • View Profile
  • Liked: 307
  • Likes Given: 184
Re: 2018 sea ice area and extent data
« Reply #1073 on: July 16, 2018, 04:57:08 PM »
Quote
It looks like a loss of precision, concentration values binned at exactly 100/n % (n=1-7) in UH's product, larger than usual pixel noise in that from Bremen. Also no doubt that the low number of swaths (due to the low latitude) is cause.

This makes me wonder, is the loss of precision predictive of future losses?
I am an energy reservoir seemingly intent on lowering entropy for self preservation.

Steven

  • Grease ice
  • Posts: 555
    • View Profile
  • Liked: 133
  • Likes Given: 16
Re: 2018 sea ice area and extent data
« Reply #1074 on: July 16, 2018, 07:31:49 PM »
Hudson Bay ice reappeared in the NSIDC images too.  The NSIDC sea ice extent for Hudson Bay increased by 169 thousand km2 over the past 2 days  (54 and 115 thousand km2, respectively). 

For sea ice area in Hudson Bay, the effect was much smaller: it increased by only 43 thousand km2 over the past 2 days  (22 and 21 thousand km2, respectively).



gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1075 on: July 16, 2018, 10:35:54 PM »
Caveat = Assumes NSIDC data is more or less reliable again.

NSIDC Total Area as at 15 July (5 day trailing average =  5,905,472 km2
This is up 36 k to 97 k above the 2010-2017 average


Total Area loss 69K, Central Seas 31 k, Periphery loss 21 k, Other Seas loss 16 k  
Analysis of individual seas.

Pacific Side
- The Okhotsk Sea area is 4 k,
- The Bering Sea area is 2 k,
- Chukchi Sea loss 6k,
- Beaufort Sea loss 6 k,
Small Area loss on the day.
Atlantic Side
- Baffin Sea loss 13 k,
- Greenland Sea loss 5 k,
- Barents Sea loss 3 k - Barents Sea area is 9 k, i.e. melting season over.
- The Kara Sea area loss 22 k.
- The Laptev Sea area loss 9 k.
CAB
- The Central Arctic Sea gain 46 k,
- The Canadian Archipelago loss  6 k,
- East Siberian Sea loss 28 k .
Other seas
- St Lawrence area at 1 k,
- Hudson Bay area loss 16 k.

Daily area loss 15th July 35k below 2010's average. Kara Sea graph attached as light relief from all these numbers.

(JAXA data - where are you ?)
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

Neven

  • Administrator
  • First-year ice
  • *****
  • Posts: 7209
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 732
  • Likes Given: 477
Re: 2018 sea ice area and extent data
« Reply #1076 on: July 16, 2018, 11:27:54 PM »
Area and extent both going up, and so no real change in compactness for the fifth day in a row:
Il faut comparer, comparer, comparer, et cultiver notre jardin

A-Team

  • Nilas ice
  • Posts: 2452
    • View Profile
  • Liked: 304
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1077 on: July 17, 2018, 03:37:38 AM »
Quote
ADS (ads.nipr.ac.jp) is more unreliable. relationshiip?
It is two academic Japanese labs that add value to *.jaxa.jp. I much prefer it for the incredible portal design which practically puts us out of business. They've bounced back a few minutes ago, the only data gap on PR89 is for the 13th and it is repairable for time series purposes.

RGB 363618 is fully loaded and does a nice job delineating rapidly shrinking HB ice (passing weather is yellowish-white)' the transition to gray mush is on June 11th.

July 15th ice is looking notably dodgier than the 14th. The four more derived products are in limbo. July 16th ice is nearing total collapse on a beautiful cloud-free worldview.

https://tinyurl.com/y88fc28u
https://ads.nipr.ac.jp/vishop/#/monitor

Quote
low latitudes, Hudson. oss of precision, concentration values binned at exactly 100/n % (n=1-7) in UH's product, larger than usual pixel noise in that from Bremen. Also no doubt that the low number of swaths (due to the low latitude) is cause.
The encircled area below has a full gamut palette yet cannot make transitions a few tens of km to the south.

Quote
St. Lawrence, Hudson Bay
Not in the Arctic; even upper HB is not quite to the Arctic Circle. An expanded remit might take in Buffalo (and western) NY because of its incredible lake effect snow pack which averages 2.4 m, vastly more than most permafrost lands or the Arctic Ocean
« Last Edit: July 17, 2018, 04:00:00 AM by A-Team »

Juan C. García

  • Nilas ice
  • Posts: 1532
    • View Profile
  • Liked: 546
  • Likes Given: 517
Re: 2018 sea ice area and extent data
« Reply #1078 on: July 17, 2018, 04:43:59 AM »
[ADS-NIPR-JAXA] ASI Extent.

ADS-NIPR already published the data until July 15th, 2018.
Waiting an hour until we have -hopefully- the July 16th data.
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

Juan C. García

  • Nilas ice
  • Posts: 1532
    • View Profile
  • Liked: 546
  • Likes Given: 517
Re: 2018 sea ice area and extent data
« Reply #1079 on: July 17, 2018, 06:11:34 AM »
ADS-NIPR is already half an hour late.  :-\

I can be wrong, but I see that there has been a melting on the Russian side of the Arctic (Kara, Laptev and East Siberian Sea). And even a little at Beaufort.
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

Juan C. García

  • Nilas ice
  • Posts: 1532
    • View Profile
  • Liked: 546
  • Likes Given: 517
Re: 2018 sea ice area and extent data
« Reply #1080 on: July 17, 2018, 06:52:55 AM »
[ADS-NIPR-JAXA] ASI Extent.

July 16th, 2018: 8,136,394 km2, a drop of -43,283 km2.
2018 is the eleventh lowest on record.
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

Ned W

  • Guest
Re: 2018 sea ice area and extent data
« Reply #1081 on: July 17, 2018, 10:59:51 AM »
This week's data glitch has kind of messed up the JAXA numbers.  Within a span of three days we now have:

* JAXA's biggest (and only second-ever) July extent "increase" (+5390 km2, on 14-15 July).  The only other July increase was in 2013 (+4102 km2 on 29-30 July).

* JAXA's second-biggest July extent "decrease" (-177574 km2, on 12-13 July).  The only bigger July decrease was in 2012 (-228644 km2, on 23-24 July).

On the assumption that we're now more or less back to normal operations, today's predict-o-matic projection for the minimum is 4.74 million km2. 

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1082 on: July 17, 2018, 01:16:23 PM »
JAXA ARCTIC EXTENT - 8,136,394 km2(July 16, 2018)


Once again something to post. Just to add to Juan's post:-
- Extent is 497 k (6.1%) above the 2010's average extent on this date,
- Extent loss to date is 830k km2 (12.9%) below the 2008-2017 average, with 66% of the average melting season done.

Resulting minimum from average remaining melt = 4.75 million km2, (excluding 2012 from the average 4.85 million km2). Range of results from last ten years remaining melt is 3.85 to 5.36 million km2.

I am getting that 2017 feeling more and more - extent losses never enough to catch up on slow melt to date. There is, on average, just one third of further extent loss to go. In one weeks time daily extent loss, on average, will be already in significant decline. A September minimum of 4.75 to 5.25 million km2 looks very possible.

ps:- After all the kerfuffle - predictions of icemageddon et al, the data is just about exactly back to where it was four days ago, except with four days less of melting to go.

EDIT - Whoops re missing graph
« Last Edit: July 17, 2018, 02:32:32 PM by gerontocrat »
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

A-Team

  • Nilas ice
  • Posts: 2452
    • View Profile
  • Liked: 304
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1083 on: July 17, 2018, 03:42:34 PM »
All the ducks are back in a row ... but it's not entirely clear what the best way to score Hudson Bay is, even if we had someone out there in a rowboat.

Note the PR89 (radar polarization ration at 89 GHz) is perhaps becoming useful with this late-stage ice when it's not under too much of a weather overlay; the resolution is not great but it is fairly direct like visible, not so much algorithmically processed.
« Last Edit: July 17, 2018, 03:55:39 PM by A-Team »

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1084 on: July 17, 2018, 03:55:26 PM »
From my post yesterday:-
Quote
Caveat = Assumes NSIDC data is more or less reliable again.
Ho hum.
NSIDC was testing 9 to 12 July. That data has now disappeared from their spreadsheets - it has not done my tables and graphs a lot of good. I will not correct by interpolation. Presumable NSIDC will have to decide what to do. If, sometime later, they put in data that will automatically sort out my data.

We wait and see, methinks. So here are the tables, warts and all. AND IGNORE ALL VALUES IN PREVIOUS POSTS FOR JULY 10 TO JULY 12?

NSIDC Total Area as at 16 July (5 day trailing average =  5,826,332 km2
This is 124 k above the 2010-2017 average


Total Area loss 117K, Central Seas 60 k, Periphery loss 33 k, Other Seas loss 24 k  
Analysis of individual seas.

Pacific Side
- The Okhotsk Sea area is 4 k,
- The Bering Sea area is 2 k,
- Chukchi Sea loss 11k,
- Beaufort Sea loss 4 k,
Small Area loss on the day.
Atlantic Side
- Baffin Sea loss 23 k,
- Greenland Sea loss  k,
- Barents Sea loss 4 k - Barents Sea area is 8 k, i.e. melting season over.
- The Kara Sea area loss 23 k.
- The Laptev Sea area loss 12 k.
CAB
- The Central Arctic Sea gain 38 k,
- The Canadian Archipelago loss  5 k,
- East Siberian Sea loss 32 k .
Other seas
- St Lawrence area at 1 k,
- Hudson Bay area loss 24 k.

Area minus Extent Graph attached - as it says on the platforms on London Underground stations - " (don't) Mind the Gap! "
« Last Edit: July 17, 2018, 04:41:07 PM by gerontocrat »
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

colchonero

  • New ice
  • Posts: 88
    • View Profile
  • Liked: 13
  • Likes Given: 2
Re: 2018 sea ice area and extent data
« Reply #1085 on: July 17, 2018, 05:03:09 PM »
Could someone post regional extent graphs, please? We haven't had them in a long time.

Juan C. García

  • Nilas ice
  • Posts: 1532
    • View Profile
  • Liked: 546
  • Likes Given: 517
Re: 2018 sea ice area and extent data
« Reply #1086 on: July 17, 2018, 05:50:25 PM »
Could someone post regional extent graphs, please? We haven't had them in a long time.

You can always see them on Neven's Blog:

https://sites.google.com/site/arcticseaicegraphs/regional
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

colchonero

  • New ice
  • Posts: 88
    • View Profile
  • Liked: 13
  • Likes Given: 2
Re: 2018 sea ice area and extent data
« Reply #1087 on: July 17, 2018, 07:55:58 PM »
Thanks, that's exactly what I was looking for.

Juan C. García

  • Nilas ice
  • Posts: 1532
    • View Profile
  • Liked: 546
  • Likes Given: 517
Re: 2018 sea ice area and extent data
« Reply #1088 on: July 18, 2018, 06:50:28 AM »
Information of ADS-NIPR-JAXA ASI Extent is 1+ hours late.  ???
I will wait half an hour more, to see if they update it.

https://ads.nipr.ac.jp/vishop/#/extent

Edit:
Now it is 2+ hours late.
Please feel free to post the updated information, if it appears.
« Last Edit: July 18, 2018, 07:41:13 AM by Juan C. García »
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

Ned W

  • Guest
Re: 2018 sea ice area and extent data
« Reply #1089 on: July 18, 2018, 11:44:02 AM »
Now it is 2+ hours late.
Please feel free to post the updated information, if it appears.

JAXA extent for 17 July = 8042914 km2.
This is 11th lowest in the JAXA record.
This is a drop of 93480 km2 from the previous date.

Here are the differences from previous years:

YearExtentDelta
20117151306-891608
20127374047-668867
20167454956-587958
20177479327-563587
20077510248-532666
20137533852-509062
20147577065-465849
20107917418-125496
20157920694-122220
20068022931-19983
201880429140
20098236301193387
20088359989317075
20058361816318902
20028856961814047
20038861718818804
200490634871020573

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1090 on: July 18, 2018, 12:14:28 PM »
Has normal data collection by JAXA (and NSIDC) been resumed? I assume yes.

JAXA ARCTIC EXTENT -8,042,914 km2(July 17, 2018

Once again something to post. Just to add to NedW's post:-
- Extent is 492 k (6.1%) above the 2010's average extent on this date,
- Extent loss to date is 820k km2 (12.3%) below the 2008-2017 average, with 67% of the average melting season done.

Resulting minimum from average remaining melt = 4.75 million km2, (excluding 2012 from the average 4.85 million km2). Range of results from last ten years remaining melt is 3.85 to 5.35 million km2.

It is that 2017 feeling  - extent losses even when above average never enough to catch up on the slow melt to date. There is, on average, one third of further extent loss to go. In one weeks time daily extent loss, on average, will be already in significant decline. A September minimum of 4.75 to 5.25 million km2 looks the most likely outcome.

For those who like numbers, the last table has all the data from which the "predictions" table (which are not predictions) is made.
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1091 on: July 18, 2018, 08:56:07 PM »
Ho hum.

NSIDC has put in the data that automatically sorted out my data. Pity that all the posts since 9th July were pretty much a waste of time.
LESSON TO MYSELF:- When NSIDC is in testing mode, go on holiday until at least 2 days after testing over.


NSIDC Total Area as at 16 July (5 day trailing average =  5,749,799    km2
This is 148 k above the 2010-2017 average


Total Area loss 77 K, Central Seas 53k, Periphery loss 18 k, Other Seas loss 5 k  
Analysis of individual seas.

Pacific Side
- The Okhotsk Sea area is 4 k,
- The Bering Sea area is 2 k,
- Chukchi Sea loss 16 k,
- Beaufort Sea loss 12 k,
Atlantic Side
- Baffin Sea loss 14 k,
- Greenland Sea loss 3 k,
- Barents Sea loss 1 k - Barents Sea area is 6 k, i.e. melting season over.
- The Kara Sea area loss 16 k.
- The Laptev Sea area loss 6 k.
CAB
- The Central Arctic Sea gain 21 k,
- The Canadian Archipelago loss  8 k,
- East Siberian Sea loss 15 k .
Other seas
- St Lawrence area at 1 k,
- Hudson Bay area loss 24 k.

Barents Sea Graph attached - simply to show that that sea is not in the game any more (along with St. Lawrence, Okhotsk and Bering Seas).
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

Richard Rathbone

  • Grease ice
  • Posts: 533
    • View Profile
  • Liked: 39
  • Likes Given: 7
Re: 2018 sea ice area and extent data
« Reply #1092 on: July 18, 2018, 11:37:55 PM »
Did NSIDC revise the June data too? That was the testing period that had the most blatantly wrong figures in it.

Ho hum.

NSIDC has put in the data that automatically sorted out my data. Pity that all the posts since 9th July were pretty much a waste of time.
LESSON TO MYSELF:- When NSIDC is in testing mode, go on holiday until at least 2 days after testing over.



Juan C. García

  • Nilas ice
  • Posts: 1532
    • View Profile
  • Liked: 546
  • Likes Given: 517
Re: 2018 sea ice area and extent data
« Reply #1093 on: July 19, 2018, 06:36:39 AM »
Now it is 2+ hours late.
Please feel free to post the updated information, if it appears.

JAXA extent for 17 July ...


Thank you, Ned W  :)

Again, ADS is one hour late (without any post), so I will thank if someone can post the updated data. The link is:

https://ads.nipr.ac.jp/vishop/#/extent
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1094 on: July 19, 2018, 10:32:07 AM »
Now it is 2+ hours late.
Please feel free to post the updated information, if it appears.

JAXA extent for 17 July ...
Looks like no data today.


Thank you, Ned W  :)

Again, ADS is one hour late (without any post), so I will thank if someone can post the updated data. The link is:

https://ads.nipr.ac.jp/vishop/#/extent
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1095 on: July 19, 2018, 02:38:37 PM »
Still not sure about Hudson Bay data. Anyway feeling lazy, so tables and a graph and limited commentary only.

NSIDC Total Area as at 16 July (5 day trailing average =  5,693,169    km2
This has increased to 189 k above the 2010-2017 average

     
Total Area loss 57 K, Central Seas 52k, Periphery loss 11 k, Other Seas GAIN 7 k

Area daily loss 40k below 2010's average.

Greenland Sea graph attached. If Fram Strait Export of Sea Ice remains zero, how low will it go?

Note how over the satellite record the story for this sea (until this year) is a steady significant loss of WINTER sea ice.


"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1096 on: July 19, 2018, 03:01:46 PM »
And for a bit of variety - a look at the 3 million km2 elephant in the room - the Central Arctic Sea.

Extent and area graphs attached.
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

Istari

  • New ice
  • Posts: 8
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: 2018 sea ice area and extent data
« Reply #1097 on: July 19, 2018, 03:12:18 PM »
Probably a stupid question, but how can we keep seeing these numbers:
- The Okhotsk Sea area is 4 k,
- The Bering Sea area is 2 k,
Where is this ice, or is something wrong with the data ?

gerontocrat

  • First-year ice
  • Posts: 6816
    • View Profile
  • Liked: 1711
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1098 on: July 19, 2018, 03:54:16 PM »
Probably a stupid question, but how can we keep seeing these numbers:
- The Okhotsk Sea area is 4 k,
- The Bering Sea area is 2 k,
Where is this ice, or is something wrong with the data ?
It might be a case of limitations to the sensors on board  the satellite(s) and /or the algorithm, and / or little bits of ice stuck up sheltered nooks and crannies along the coast. But of no significance and can be ignored as it happens every year.
"Para a Causa do Povo a Luta Continua!"
"And that's all I'm going to say about that". Forrest Gump
"Damn, I wanted to see what happened next" (Epitaph)

A-Team

  • Nilas ice
  • Posts: 2452
    • View Profile
  • Liked: 304
  • Likes Given: 22
Re: 2018 sea ice area and extent data
« Reply #1099 on: July 19, 2018, 05:17:27 PM »
What may be happening is that even at the posted 3.125 km resolution of UH AMSR2, bays, inlets, lagoons, rugged coastlines, island edges and so forth can have many marginal pixels partly over land, partly over water. This can result in 'ice' even when it is long gone, as the July 7th WorldView and July 17th AMSR2 show for the Bering Sea side of the Bering Strait. However the error is negligible in overall percentage terms for the Arctic Basin and even more so in differencing consecutive days.