Support the Arctic Sea Ice Forum and Blog

Author Topic: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland  (Read 1200048 times)

Shared Humanity

  • Guest
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1650 on: July 01, 2016, 04:39:50 PM »
Wow! A big one, and on all fronts simultaneously.

That the calving occurred across the entire front suggests there is still significant buttressing across a remarkably long calving front! What is the length of this front?
« Last Edit: July 01, 2016, 05:09:52 PM by Shared Humanity »

Shared Humanity

  • Guest
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1651 on: July 01, 2016, 04:45:16 PM »
OK, my attention is now drawn to another striking feature of this image. What is that very large dark spot on the upper right? Can that be a melt pond? It is huge! Is that land appearing through the thinning ice sheet? Is it a spot of ketchup that has been carelessly dripped from a NASA scientist's hotdog?
« Last Edit: July 01, 2016, 05:42:34 PM by Shared Humanity »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1652 on: July 01, 2016, 05:03:03 PM »
Quote
Converting a band to 16bit
Goes the other way. The satellite sensor is digitized directly to 16-bit. Once that is rounded off to 8-bit, there is no going back (no recovery of 1original precision).
Quote
That should not be the case, all bands of an image, and have the output to be a single product.
SNAP was written ab initio to process Sentinel data sensibly. If it doesn't, be good to contact the help desk. Again, the imagery will ultimately need processing in photoshop-type software and end up as 8-bit png, the best format the forum supports.

Cloudy again on 30-JUN-16 LC80812332016182LGN00. S1A orbit is off to the west as usual. It makes more sense to shuffle file formats on the server side, as at http://www.polarview.aq/arctic

Quote
What is the length of the calving front?
The actively calving region is about 5 km across, slightly more following the arc. (Count pixels, multiply by 5 in the 5m resolution imagery above.) The current position is not anomalous and provides no information about whether this year will see a record repeat. Looking at berg cluster spacing, this event may have come in two waves with steady minor calving in between.
Quote
another striking feature of this image. What is that very large dark spot on the upper?
Mmmm, need to provide an arrow or hand-drawn loop.

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1653 on: July 01, 2016, 05:45:24 PM »
OK, my attention is now drawn to another striking feature of this image. What is that very large dark spot on the upper right? Can that be a melt pond? It is huge! Is that land appearing through the thinning ice sheet? Is it a spot of ketchup that has been carelessly dripped from a NASA scientist's hotdog?

I was wondering about that too...

Shared Humanity

  • Guest
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1654 on: July 01, 2016, 06:19:56 PM »
OK, my attention is now drawn to another striking feature of this image. What is that very large dark spot on the upper right? Can that be a melt pond? It is huge! Is that land appearing through the thinning ice sheet? Is it a spot of ketchup that has been carelessly dripped from a NASA scientist's hotdog?

I was wondering about that too...

If it is ketchup, I hope they reprimanded the guy.  ;D
« Last Edit: July 02, 2016, 12:27:16 AM by Shared Humanity »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1655 on: July 01, 2016, 06:22:45 PM »
Besides Sentinel and Landsat there is also Aster. Aster seems to have a cycle of 16 days (like the Landsat's) Dates on which Aster is in a favorable position are 2016-06-05 (+ multiples of 16 days) and 2016-05-31 (+ n*16).
Here is a large, click required, animation from 5 and 21 June to judge the quality.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1656 on: July 01, 2016, 10:34:11 PM »
Nice enough image clarity for 8-bit at 15 meters. It could be useful in other regions too such as Beaufort Gyre or Nares Strait.

What link are you using to download Aster?

In EarthExplorer, upon choosing "NASA LPDAAC collections", I  am getting either .bs in .hdf format from "Standard Product' or pretty-in-pink from "Visible Product'). It is not clear what Aster bands correspond to red, blue, green though GRB gives a somewhat more natural bluish image. It says Band3n was acquired but not Band3B

Quote
April 1, 2016 NASA, Japan Make ASTER Earth Data Available At No Cost. ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) imagery is available since 2000 from the NASA Terra satellite. This sensor is a joint activity of NASA and the Japanese Ministry of Economy, Trade and Industry (METI). ASTER supplies high resolution visible and infrared imagery, plus thermal imagery and stereographic capabilities. The collection of ASTER available through GLCF is designed to compliment overall project goals of distributing a global, multi-temporal, multi-spectral and multi-resolution range of imagery appropriate for land cover analysis.

http://glcf.umd.edu/data/aster/

Beginning today, all Earth imagery from a prolific Japanese remote sensing instrument operating aboard NASA’s Terra spacecraft since late 1999 is now available to users everywhere at no cost.

The public will have unlimited access to the complete 16-plus-year database for Japan’s  Spaceborne Thermal Emission and Reflection Radiometer (ASTER) instrument. ASTER’s database currently consists of more than 2.95 million individual scenes.

ASTER data are now available via electronic download from NASA’s Land Processes Distributed Active Archive Center (LP DAAC) at the U.S. Geological Survey’s (USGS) Earth Resources Observation and Science Center and from AIST. To access the data, visit:

http://asterweb.jpl.nasa.gov/data.asp
or
https://lpdaac.usgs.gov/dataset_discovery/aster
or
https://gbank.gsj.jp/madas/
or
http://earthexplorer.usgs.gov Select "NASA LPDAAC collections" under "Data Sets".

Quote
What is that huge spot?
It appears just to be an ordinary melt pond, likely one of the regulars that re-appears each year. It fits within a 50x45 pixel box so is 750x675 meters in size.
« Last Edit: July 01, 2016, 10:46:07 PM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1657 on: July 02, 2016, 08:01:35 AM »
Nice enough image clarity for 8-bit at 15 meters. It could be useful in other regions too such as Beaufort Gyre or Nares Strait.

What link are you using to download Aster?

EarthExplorer

Quote
In EarthExplorer, upon choosing "NASA LPDAAC collections", I  am getting either .bs in .hdf format from "Standard Product' or pretty-in-pink from "Visible Product'). It is not clear what Aster bands correspond to red, blue, green though GRB gives a somewhat more natural bluish image. It says Band3n was acquired but not Band3B

I am using the hdf file. Since the image data is not tagged as image data, just stored as flat binary blobs, none of my hdf aware tools did want to extract images.
In python it was simple. The images above are bands 3N,2,1 as RGB.
Band 3B cannot be found anywhere, I think it is just not included. Pity.
(N=normal, B=backward)

My impression is that the browse images have the red and green bands exchanged

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1658 on: July 02, 2016, 07:58:07 PM »
Quote
browse images have red and green bands exchanged
It is their job to get the RGB bands in standard order, not the job of thousands of downloaders to split the color and reshuffle the bands.

Quote
none of my hdf aware tools did want to extract images.
Ah ha. I looked at the imagery packet too, could not find anything in an image format, opened it with Panoply hdf tool but it could not produce anything that looked like a serious image.

They never did get the Sentinel 2A server serving either, just the checkbox search option and walked away. I wrote them a polite question 3 months ago, received a queue ticket but never heard back from them. No one is monitoring.

EarthExplorer ... they seem really under-staffed, as in no one. So many of these image servers seem a one-off grant concept with no budget for follow-thru. Even when it is just a couple of lines of code, they won't do it without that next big grant. But maintenance is not as exciting as a breakthru concept for funders.

Since Aster has different orbital timing than Landsat and Sentinel, it might be worthwhile to look at the 15 Aug 15 mega-event to see if it got in a shot at a supplemental time.
« Last Edit: July 02, 2016, 11:50:00 PM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1659 on: July 03, 2016, 08:44:41 AM »
2016/7/1 ( R068) is not worth the trouble. Dark melt ponds "shine" through the clouds.

Iceismylife

  • Frazil ice
  • Posts: 281
    • View Profile
  • Liked: 1
  • Likes Given: 1
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1660 on: July 03, 2016, 11:32:01 PM »
That is a way to see if the clouds are thick enough to cool.  If you can see the melt ponds then they are warming clouds.  If you can't then they are thick enough to cool.  Give or take a bit.  If the bottom of the clouds are bright then they are warming if they are dark then they are cooling.  But how to see that top down on a white background?  Melt ponds.

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1661 on: July 05, 2016, 03:44:04 AM »
I keep trying to figure out how to download WEB22 images via ESA SNAP, but I still don't have a clue how to work the thing, so I will just point out that the melt ponds mostly up and drained today, and it appears that something has gone on north and to the west of the calving front of the south branch, as in it looks like there was a lot of movement, such that even that dirty line coming out of the north branch was diverted. 

(And, btw, what is that dirty line, anyway?  It is so persistent, one could even speculate that it was an oil slick.  How can it be coming out black (or dark) like that for so long?  Maybe y'all have already discussed this somewhere, but I missed it.)

iwantatr8

  • New ice
  • Posts: 79
    • View Profile
  • Liked: 21
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1662 on: July 05, 2016, 11:10:14 AM »

(And, btw, what is that dirty line, anyway?  It is so persistent, one could even speculate that it was an oil slick.  How can it be coming out black (or dark) like that for so long?  Maybe y'all have already discussed this somewhere, but I missed it.)
Posts around #285 cover that area in detail, continuous movement but no real change in the area since 2014.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1663 on: July 06, 2016, 12:08:26 AM »
Tenney, it is easier for S2A July images just to use the Amazon site, http://sentinel-s2-l1c.s3-website.eu-central-1.amazonaws.com/#tiles/22/W/EB/2016/7/.

There is a new 'WEB22' up there today for July 5th. It is wonderfully clear but a bit east of the calving front and bays, still of interest for those making animations of melt pond and crevasse draining and filling. This is the direct link to the band 4 (red) jpeg2000 file.

http://sentinel-s2-l1c.s3.amazonaws.com/tiles/22/W/EB/2016/7/5/0/B04.jp2

The stain on the north bank, as noted above, is glacial rock till scraped off by movement over fixed rock and after calving, visible as a track on downstream surface melange. It can become deflected by faster moving melange, eg from the south branch. It appears whitish in the July 4th S1A survey image below.
« Last Edit: July 06, 2016, 01:03:56 AM by A-Team »

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1664 on: July 06, 2016, 03:25:27 AM »
Extreme neophyte that I am, I don't know how to open a .jp2 file.  Any hints would be appreciated!

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1665 on: July 06, 2016, 04:12:54 AM »
Extreme neophyte that I am, I don't know how to open a .jp2 file.  Any hints would be appreciated!
There's a thread here
https://forum.arctic-sea-ice.net/index.php/topic,1348.msg59097.html#msg59097

I tried a while ago, and had given up.  I have not yet investigated the other software mentioned.
FNORD

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1666 on: July 06, 2016, 07:58:56 AM »
Quote
I don't know how to open a .jp2 file.... I tried a while ago, and had given up.

On a mac, just double-clicking on the download will open it in Preview which is a weak freebie that comes with the operating system. From there you can adjust contrast fairly well, then take a screenshot (to greatly reduce file size, command-option-shift-4, and paste into your favorite photoshop-like software.

The problem comes if you want color. You can repeat the process on the visible bands to get RGB but it won't be possible to replicate the crop boundaries in Preview (as it doesn't support layer stacking). So then you would need to import into software like gimp that does and manually align the three channels and crop them simultaneously. Once the three grayscales are in register, any graphics software will display them as one natural color image.

It remains very perplexing why ESA chose an obscure format like jp2 (jeg2000) since it never gain any traction in the 15 years after it was proposed. At first I imagined that it offered some savings in file size (ie storage costs) for lossless compression.

The graphic below from polarview -- which unlike EarthExplorer offers you an opt-out -- suggest this is not the case -- there's no savings (though it needs a comparison with the standard format for 16-bit, namely tif).

My current theory is (1) they themselves use linux software that supports 16-bit jp2, (2) this being ESA's first venture into free imagery open to the public, they have no clue what public desktops can do, not knowing anyone themselves who uses AppleL or Microsoft products, (3) they made a dumb mistake but cannot admit it now because it would mean admitting to a dumb mistake plus creating a huge pipeline backlog, so (4) they'll continue with jp2 under the theory 'I got mine, fy.' That's the great thing about owning the satellite, you don't have to explain or justify anything, just do as you please.

As mentioned, wipneus is currently providing a 16-bit jp2-to-tif conversion service for Jakobshavn band 4 only. Try that site to see if the file sizes overwhelm your system. Most people can squeak by with the 8-bit files at the W/EB site or PolarView (which are not restricted to Jakobshavn).

I came across another unbelievable do-as-you-please change at DMI - an unsupervised graphic artist there decided to delete the Jakobshavn calving front from all the radar imagery! In favor of a white space, a big DMI promotional logo, and a repetitive locator map. Nothing can be done, even Danish registrants here never could get the time of day out of DMI.

i've worked on projects where enabling the biggest possible user community actually was a top priority. It all starts with working directly with typical users. If you let the programmers shield themselves behind multiple layers of help desk hires, feedback never happens.

After seeing the same thing on so many satellite server sites, I've come to realize there was never any sincere interest in opening access up to the unwashed public. So they just go through the motions, analogous to other social programs for non-whites.

https://drive.google.com/folderview?id=0B8NUHa4P2gmlfnhfcUwtRHVzZmRnWnZhQ0UwSkI3ZW12R0dYQ0R4UUR5eElPeEJSYzBsRFk&usp=sharing#list
« Last Edit: July 06, 2016, 08:26:22 AM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1667 on: July 06, 2016, 09:24:46 AM »
The 2016/7/5 ( R125) is in as mentioned above. I have uploaded S2A_R125_V20160705T150020.22WEB.B04.tiff.

(link https://drive.google.com/open?id=0B8NUHa4P2gmlZTd3NVJLNUJ2Z2s)


Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1668 on: July 06, 2016, 09:50:55 AM »

It remains very perplexing why ESA chose an obscure format like jp2 (jeg2000) since it never gain any traction in the 15 years after it was proposed. At first I imagined that it offered some savings in file size (ie storage costs) for lossless compression.

The graphic below from polarview -- which unlike EarthExplorer offers you an opt-out -- suggest this is not the case -- there's no savings (though it needs a comparison with the standard format for 16-bit, namely tif).


It saves storage costs and data communication time and costs.

Here is a random sample:
/opt2/sentinel2/zach/S2A_R125_V20160705T145922/27XVH$ ls -lgG  B04*
-rw-rw-r-- 1 129871440 jul  5 20:52 B04.jp2
-rw-rw-r-- 1 199572094 jul  6 08:51 B04.png
-rw-rw-r-- 1 226144250 jul  6 08:47 B04.tiff


All are 16 bits single band files.

File size compared to PNG: 65%
File size compared to TIFF:  57%

Some people would consider such savings huge.

BTW, the only valid comparison realistically  is with TIFF as they both can contain GEO metadata.

I doubt that linux is any reason for preferring jpeg2000, Imagemagick, work horse for multiple programs, does not work out of the box.

I am afraid that the real reason that jpeg2000 did not catch on, is fear of hidden patents. Although all parties involved in the design have donated their known patents and vowed never to enforce any not known yet, there is nothing preventing a patent harvester to try to profit after the standard is well established. The GIF format was a prime example.
If so, IMHO ESA deserves some praise for choosing jp2 and risking bringing such problems above the water.

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1669 on: July 06, 2016, 11:19:11 AM »

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1670 on: July 06, 2016, 10:05:26 PM »
On a mac, just double-clicking on the download will open it in Preview
I recall trying this before.  I tried again from the polarview portal.  It just gives me a black screen, perhaps my system is to old, or not enough memory.  I think any further discussion should go over to the developers thread.
FNORD

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1671 on: July 07, 2016, 05:11:01 AM »
I've got a Dell laptop with a 4K screen, 16 Gb and a Core i7 processor.  Seems adequate, no?

I've wanted to get a Mac for years but still can't afford one.

I might try to finance a Mac around Black Friday in November, so I guess I just have to wait to see these images.




A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1672 on: July 07, 2016, 05:57:14 PM »
Quote
File size compared to TIFF: 57% Some people would consider such savings huge.
I would guess inconsequential at a commercial cloud site, especially in the overall ESA budgetary context. it would not even be 0.1 staff FTE over lifetime of the satellite. Clue: nobody else is using jp2. Makes sense to offload both storage and serving.

Quote
bravo if taking on IP issues
Mission creep. It does not make sense for ESA to take on distractive unknowns. Hmmm, 15 years have gone by, maybe everything has expired. I don't recall that unisys made a penny off gif (LZW compression actually), just patent attorneys. Png was invented as the response.

Quote
linux is no reason for preferring jpeg2000, Imagemagick does not work out of the box there.
Question then is, what is the most common imaging software used at large satellite centers? I am guessing not something that the public can use (or has ever heard of).

Quote
streaming format
More rearview mirror. Wouldn't need in Latvia, the #1 country now in terms of best EU internet, sure you need help if you live in the US.

Quote
Move to a dev forum
Good idea. Be great if we could get everyone empowered but a lot of obstacles to that.

Another clear day, another near miss. This Landsat did get a piece of the calving front though. The floating lab bergs are midway down the channel now, they have not rolled, split or lost features to melt.
« Last Edit: July 07, 2016, 06:04:16 PM by A-Team »

Tealight

  • Frazil ice
  • Posts: 490
    • View Profile
    • CryosphereComputing
  • Liked: 176
  • Likes Given: 17
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1673 on: July 08, 2016, 03:11:42 AM »
I would guess inconsequential at a commercial cloud site, especially in the overall ESA budgetary context. it would not even be 0.1 staff FTE over lifetime of the satellite. Clue: nobody else is using jp2. Makes sense to offload both storage and serving.

A-Team, what is wrong with you lately? Why do just dismiss everybodies contribution without proper thought? I'm sure you can figure out with a rough calculation that 0.1 staff is more off than claiming the arctic will be completly ice free this year.


I might try to finance a Mac around Black Friday in November, so I guess I just have to wait to see these images.
Nooooooooo! If you don't have any other program then you should use ESAs SNAP application.

I could create a User guide on the Developers Corner or if Wipneus updates the existing Sentinel 2 topic all important information would be in the first post and not scattered among different topics and posts.




TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1674 on: July 08, 2016, 05:09:47 AM »
I did download SNAP and it seems to work, insofar as it let's download .jp2 images that I cannot open.  I have heard there are apps for that, but really haven't wanted to experiment by downloading any of them because I don't want to download possible threats to my laptop.

I don't need to manipulate the photos, and I don't care if they are really large files.  I just want to look at the details.

Oh, I guess I should say that I really hate Win10's photo viewer, so am very open to replacing it with something better. 

My needs are simple.  I used to do everything I wanted with the Office Suite accessory Picture Manager.  'course Microsnot thinks better, and got rid of it.

sidd

  • First-year ice
  • Posts: 6774
    • View Profile
  • Liked: 1047
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1675 on: July 08, 2016, 05:19:42 AM »
ImageMagicK for windows

http://imagemagick.sourceforge.net/http/www/windows.html

does jp2, and conversions into a million formats

i7 with 16 gig is adequate

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1676 on: July 08, 2016, 06:05:40 AM »
Thank you so much, sidd!

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1677 on: July 08, 2016, 06:08:01 AM »
Uh, do I need the download with binaries or without?  What do they do?

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1678 on: July 08, 2016, 09:56:56 AM »
I did download SNAP and it seems to work, insofar as it let's download .jp2 images that I cannot open.
I am puzzled, SNAP is designed to open S-2 imagery, so in what sense you "cannot open" them? Note that you should open the .zip and not the .jp2 file inside SNAP (satellite imagery comes with scores of metadata that will be lost that way).

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1679 on: July 08, 2016, 04:51:05 PM »
FNORD

sidd

  • First-year ice
  • Posts: 6774
    • View Profile
  • Liked: 1047
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1680 on: July 08, 2016, 09:32:50 PM »
http://imagemagick.sourceforge.net/http/www/windows.html

ImageMagicK: use the binaries and  16 bit depth version

"Retrieve Install Package

Retrieve the ImageMagick Windows binary install package from a one of the ftp sites listed at the ImageMagick Download Sites page. The file will be named similar to ImageMagick-i686-pc-windows.exe (QuantumDepth=16 version) or ImageMagick-i686-Q8-windows.exe (QuantumDepth=8 version),and should be found in the 'binaries' subdirectory of the ftp site. Note that the QuantumDepth=8 version consumes much less memory and CPU than the QuantumDepth=16 version, but provides less color resolution."

The instructions on that page are correct. Ignore the later sections on "Advanced Installation" for now. Consider installing Ghostscript as well.

http://www.ghostscript.com/download/gsdnld.html

Use the 64 bit windows version exe file, non commercial license will do for research use.

sidd

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1681 on: July 08, 2016, 10:13:25 PM »
Here is an interesting expert discussion of the relative merits of Landsat-8 and S1A applicable to detecting melt lakes on ice sheets and ice shelves or melt ponds on sea ice. We need to move on from hunch-based interpretations.

Quote
Sentinel-1A level-1 high resolution GRD data have 20X22m resolution while  Landsat-8 bands have 30m resolution prior to pan-sharpening. For a 100m x 100m area, there would be approximately 25 pixels in the Sentinel image and 9 pixels in the Landsat image. There's the mixed pixel problem -- those on the edge cover some lake and some land.

The lake area was determined using NDWI index (not RGB). This  transformsthe data from multivariate to univariate, throwing data away. The NDWI was developed pre-Landsat 8 and does not utilize the blue edge band which is very relevant to water/wetland gradients.

The Sentinel-1A sensor is an active radar system carrying a C-band synthetic aperture radar array whereas Landsat 8 is a passive spectral system with 16-bit radiometric resolution across 0.43 - 2.29 microns. The characteristics of the sensors dictate the feature resolution of the objects being detected.

Because the Landsat 8 sensor is carrying a blue edge band (0.43 - 0.45 um), it is likely able to discriminate water/wetland features more effectively than the backscatter signal from Sentinel-1A. This would account for the differences in measured feature size, regardless of the on the ground cell resolution differences. If you want to use radar SAR data for water/wetland classification it would be better to use L-band data from ALOS PALSAR. This would provide much better discrimination than Sentinel-1A C-band data.

You really need to consider a classification algorithm. It is impossible to visually elucidate information contained in both of these data. There could very well be information contained in the data that is providing feature discrimination that is not readily apparent in a visual assessment.

The difference in the lake measurement results could be associated with user bias and not information content. This is particularity relevant in a multivariate context where there may be very relevant information, contained across multiple bands that are not being used in and RGB composite, comprising the image backdrop being used in heads-up digitizing. You are effectively throwing away data.

In the radar data there is likely a pixel gradient that is not apparent in your visual interpretation. The image can also be seriously compromised based on the type of stretch used in display.

It is not all about resolution, you also have to factor in separability and specific feature detection capability of a given sensor
There's more back and forth on this and related issues at:
http://gis.stackexchange.com/questions/187645/sentinel-1a-and-lansat-8-resolution-difference

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1682 on: July 08, 2016, 10:28:02 PM »
Sorry for getting back on topic, but there looks like another good size calving on the south branch.  It's a little blurry, but it looks the same in Aqua and Terra, so it's not clouds, and there is a lot of motion of the melange in the channel.
FNORD

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1683 on: July 09, 2016, 01:38:26 AM »
Quote
another good size calving on the south branch? It's a little blurry, but it looks the same in Aqua and Terra, so it's not clouds, and there is a lot of motion of the melange in the channel.
Good practice to use both Terra and Aqua (for time lag) and melange areal displacement (as proxy for volume calved)!

Kinda fun to watch July 6th and 8th at high frame rate. Note the small stationary region, lower left. This is a long-known pinning point near the surface.The 8th has a full view of the calving front.

The second animation shows bergs stuck on the terminal moraine sill but a passageway towards the north where they can get through to Disko Bay. Note the large berg adjacent to the 'ice camp' has crashed.

LC80 84 232 2016 188 LGN00.jpg
LC80 82 233 2016 190 LGN00.jpg
« Last Edit: July 09, 2016, 02:16:50 AM by A-Team »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1684 on: July 09, 2016, 05:36:19 AM »
The balance of calving-driven retreat and gravity-driven extension has tilted towards net forward positioning of the front in the 16 days since the previous image with the same path,row. There's no telling how this season will end up.

We now have three in this series that could be used for RGB false color, velocity determination, or even seasonal acceleration. The temptation though is to use S2A because of its better resolution, as the shadowing of these low sun angle Landsat images brings in too much variation optimal for feature recognition.

LC80822332016190LGN00_B8.TIF
LC80822332016174LGN00_B8.TIF
LC80822332016157LGN00_B8.TIF
« Last Edit: July 09, 2016, 05:53:06 AM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1685 on: July 09, 2016, 12:16:22 PM »
Sentinel 2A 2016/7/8 ( R025) is in. Uploaded as S2A_R025_V20160708T151305.22WEB.B04.tiff

Metadata says it has only 3% cloud cover. Yet there are a more cloud-like dark blotches than that.
One of them (to the east, 20km below the south branch) has even some rainbow color visible (or is it?).

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1686 on: July 09, 2016, 04:46:35 PM »
Quote
Rainbow
Oh that is just from the scales on the back of that big salmon. This S2A is quite an amazing instrument, I'm sceptical Landsat could capture this.

For TenneyN, the other bands are at the link below. If you double-click on jp2 downloads, there may be an app on your computer defaulted to open them (Preview in the mac OS). Then copy them over to preferred image processor.

http://sentinel-s2-l1c.s3-website.eu-central-1.amazonaws.com/#tiles/22/W/EB/2016/7/8/0/

The cloud cover should be more apparent in longer wavelength bands, as with 3-6-7 Modis below of the same date but maybe different time. ESA may not include thin clouds in statistics if ground features can be seen underneath. Like cloud shadows on the ground, thin uniform ones can be removed using tools like CLAHE in ImageJ. This R025 is a very sharp image around and to the north of  the calving front.
« Last Edit: July 09, 2016, 05:43:29 PM by A-Team »

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1687 on: July 09, 2016, 05:31:22 PM »
Here is a sentinel 1 gif from Jul 4 to Jul 9 showing the latest calving, not the best images, or matchup between them, but you can get an idea of what happened.  Lots of action on the north side of the south branch.
http://www.polarview.aq/images/105_S1jpgfull/S1A_EW_GRDM_1SDH_20160709T100733_9191_N_1.final.jpg

S1A_EW_GRDM_1SDH_20160704T095945_8965_N_1.final

FNORD

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1688 on: July 09, 2016, 06:44:39 PM »
Note the images are in different geometries (not being from an orbital repeat) and so don't align well.  The distortion varies across the scene. The only desktop way we know of fixing these is to go through SNAP, bringing them in UTM22 mercator using nukefix's operations graph which has the advantage of allowing Landsats (except for 8,11's) to be tossed into the mix.

The new series ASTER level 1T release does not say anything in the metadata about its projection. The new Sentinel 1B will be the same as 1A so not raise any further issues.

When there is a major event, we scramble to get all the available imagery co-registered in a common geometry including follow-on so as to improve the time sampling. This year will be a big improvement over Landsat-8 alone (16 day return). We've also chased down the seismic traces which provide continuous coverage and so precise timing of events (though it is often murky as to what exactly is giving rise to the signal). The tidal record is also useful if it picks up calving tsunamis and timing of high tide relative to seismic initiation.

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1689 on: July 09, 2016, 09:15:59 PM »
I was hesitant to post this, but saw what looks like the start of a 2 mile melt pond draining on the Suomi pass in worldview.  I do think it's odd that this pond drains right before a calving, similar to the Zachariae-Isstrom pond draining.  This pond formed in mid June.

Perhaps there is a disturbance under the glacier, such as a surge of meltwater draining at the front, that creates a void for the pond to drain to, but it's about 15 miles away, so it seems unlikely.
FNORD

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1690 on: July 10, 2016, 10:44:59 PM »
Jakobshavn Isbræ calves another big one, it happened between July 8 and July 9 (July 9 was cloudy therefore July 10):
Have a ice day!

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1691 on: July 12, 2016, 08:47:26 AM »
2016/7/11 ( R068) is in, metadata says it: 0.01% cloud cover. S2A_R068_V20160711T151912.22WEB.B04.tiff is uploaded.

In the attachment the mentioned big calving is visible.
« Last Edit: July 12, 2016, 05:35:25 PM by Wipneus »

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1692 on: July 12, 2016, 06:08:33 PM »
How far has the calving front of the southern branch retreated since this time last year?

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1693 on: July 12, 2016, 08:11:56 PM »
Quote
How far has the calving front of the southern branch retreated since this time last year?
Keep your eye on the last rock, left corner. As long as the calving front is near that, we are nowhere near record retreat. Year-on-year comparisons for a given date would seem to have predictive value but they have not.

The position of the calving front can go back and forth numerous times during a single season depending on the balance between forward motion and calving events, see 2015 animations for that year.

We are not sure if the calving front is stuck on a ramp, approaching a ramp, or already beyond. Some calving fronts of glaciers just to the north have not retreated in the last 70 years because of  restorative feedback from bedrock slope.

We don't know yet whether JI will do the same or continue to retreat to the elbow and beyond. It's proven very difficult to get reliable high resolution data of bedrock in the current vicinity of the calving front's grounding zone because the ice is so thick and there's too much calving to allow instrumental approach.

Attn: please please don't re-post published bedrock depth models. We've looked at them many times before and found they simply do not provide the resolution needed at the calving front. The pixel spacings in transects are just too big and their values largely represent guesswork. This glacier is very difficult to study with ice penetrating radar and has too many unknown interior and basal parameters to reliably model right where it is needed the most.

It might be more informative to compare JI velocities with 2015 for the same date range. There though we wouldn't have 10 m S2A from both years at the same spacing. The 5x view below shows an average advance of 45.8 meters per day (nearly a record) over the three days separating wipneus's pairs assuming taken at the same time of day despite not being 10 day repeat orbits.

This Sentinel 2A imagery is absolutely fabulous, there's a lot more we could do with the pair below and others to come.
« Last Edit: July 12, 2016, 08:24:38 PM by A-Team »

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1694 on: July 12, 2016, 10:56:21 PM »
Even if the calving-front does not recede more than during previous summers, we know that overall the ice-sheet is thinning several meters per year at Jakobshaven:

http://m.esa.int/Our_Activities/Observing_the_Earth/CryoSat/CryoSat_reveals_recent_Greenland_ice_loss

..sooner or later this will make a diffrenence..

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1695 on: July 13, 2016, 02:03:24 AM »
Quote
Jakobshavn thinning vertically each year gonna make a difference.
Right. Free full text too, decent as just published. The CryoSat-2 radar altimetry had low spatial  resolution at 5 km and covered 2011-2014 so it won't give us thinning at the JI calving front having just one pixel there. However JI was shown to be the worst loss situation, with Petermann not even making the cut though Zachariae did.

Quote
A high-resolution record of Greenland mass balance
Malcolm McMillan et al DOI: 10.1002/2016GL069666 9 July 2016
http://onlinelibrary.wiley.com/doi/10.1002/2016GL069666/full

Figure 3. Greenland dynamic mass balance, 2011–2014. (a) Difference between Cryosat-2 and RACMO2.3 mass balance. Several marine-terminating glaciers exhibiting signs of dynamic ice loss are apparent; Kangerdlugssuaq (Ka), Jacobshavn Isbræ (JI), Upernavik Isstrøm (UI), Steenstrup (SP), and Zachariae Isstrøm (ZI); all of which have undergone velocity change during the preceding decade. For each of these glaciers, the total 4 year mass loss from the rapidly thinning terminus regions, bounded by the gray lines, is shown. The color wheels indicate the partitioning of mass losses within the bounded regions, according to surface mass and dynamic processes, with the radius of each wheel scaled according to the magnitude of the total losses.

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1696 on: July 13, 2016, 10:05:21 AM »
-69 gigatonnes from the bounded area of Jakobshaven in the image in four years? I'd call that substantial mass-loss.

ps. which publicly available DEM of the Jakonshaven area (or entire Greenland) would you recommend? In the SNAP toolbox the automatically available GETASSE30 has nasty-looking artefacts...

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1697 on: July 13, 2016, 11:26:59 PM »
I like the DEM from I Howat at UOhio/UMinn. They are made from stereo pairs of sub-1m horizontal resolution of commercial WorldView visible images.and have about 2m vertical resolution. I'm recalling they come as 16-bit grayscale with a clever secondary layer that tells where each pixel came from.

Quote
69 gigatonnes from the bounded area of Jakobshaven in the image in four years? I'd call that substantial mass-loss.
If that 69 GT spreads out over a million km2, how thick a layer does it give?

1 Gt = 1 km³ raises global sea level 2.78 microns but that is over 361 x 106 km² so 2.78*361 over just a million km2 raises it 0.001 m so 69 km³ of melt raises that 1000x1000 km2 0.07 m over the four years. Right, substantial.

The question I have is what does any of this have to do with the dome of cold fresh water said to be building over the AMOC overturning site offshore to the southeast of Greenland? That's commonly said to be "Greenland melt water" but that level of vagueness covers a whole lot of ground.

For example, it might primarily be coming from melt on the southeast side ie Kangerdlugssuaq via the Irminger Current (or even from Fram sea ice melt) and have no connection at all with Jakobshavn despite its dominant volume loss. JI's melt is swept way to the north before swinging around.

I could see where the freshwater over the AMOC might have an isotopic signature allowing inference of its origin, the source of the snow/ice being differentiated by stable isotopes (H1, H2, O16, O17, O18) and their multitudinous ratios and  H20 combinatorics.

That is, the water in snow falling at Jakobshavn might have originated from and be mass-fractionated differently than water in snow falling in southeast Greenland, the temperature of the source water being evaporated is key.
« Last Edit: July 13, 2016, 11:33:42 PM by A-Team »

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1698 on: July 14, 2016, 05:17:13 AM »
I think I should have phrased my question about the calving front differently.

I'm pretty interested in how much area has been gobbled up along the north wall of the south branch -- JI is far wider, now, isn't it? 

And the same is true of the north branch -- it used to be very noticeably narrower than the south branch. 

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1699 on: July 14, 2016, 10:24:12 AM »
If that 69 GT spreads out over a million km2, how thick a layer does it give?

1 Gt = 1 km³ raises global sea level 2.78 microns but that is over 361 x 106 km² so 2.78*361 over just a million km2 raises it 0.001 m so 69 km³ of melt raises that 1000x1000 km2 0.07 m over the four years. Right, substantial.
It looks more like 250kmx250km = 62500km2 so over that area the surface lowering of the ice-sheet should be ~1.12m in four years. Of course, most of the surface-lowering happens in the ice-stream and relatively close to the calving front.

Quote
The question I have is what does any of this have to do with the dome of cold fresh water said to be building over the AMOC overturning site offshore to the southeast of Greenland? That's commonly said to be "Greenland melt water" but that level of vagueness covers a whole lot of ground.
I wonder where the icebergs that Jakobshaven calves end up..
« Last Edit: July 14, 2016, 11:00:03 AM by nukefix »