Support the Arctic Sea Ice Forum and Blog

Author Topic: Interpretation of satellite images  (Read 4481 times)

Andreas T

  • Nilas ice
  • Posts: 1149
    • View Profile
  • Liked: 18
  • Likes Given: 4
Interpretation of satellite images
« on: May 21, 2017, 08:40:47 PM »
As suggested by shared Humanity:
I've never posted these kinds of detailed images of the ice and, quite frankly, don't know how to do this but I think this community should try to do this across the entire basin this year. Perhaps each of us could take a sea or area of our choice and post these with analysis for the community.

Can somebody here tell me how to do this? Please keep in mind that I am old and technically challenged. Imagine having to teach your grandmother over the phone and we will be fine.

There will be some fascinating threads to follow this melt season but this is one of the most important. Lets use this thread, as a group, to increase our level of understanding, both for the community and visitors. If we do this every melt season from now on, we will be conducting research, I believe, that is not currently being done, a detailed, live analysis of the behavior of weak ice under various forces, sun, storms, waves and wind etc.

(Neven) Could you post detailed instructions here to let various regulars pick this up and run with it? Assume we are all grandmothers even though some of us are computer wiz's.
« Last Edit: May 21, 2017, 08:46:41 PM by Andreas T »

Andreas T

  • Nilas ice
  • Posts: 1149
    • View Profile
  • Liked: 18
  • Likes Given: 4
Re: Interpretation of satellite images
« Reply #1 on: May 21, 2017, 09:03:28 PM »
The most easily accessible images are on NASA's EOSDIS worldview
https://go.nasa.gov/2qIKL3m
there are three satellites which cover the whole arctic once a day at slightly different times: Suomi, Aqua, Terra
To annotate these images I save them, usually by taking a screenshot
 (possibly a key on your keyboard "prt sc")
and open them in a free program called GIMP. If I use "open as layers" to  open a set of images from different dates I can easily switch and compare features. worldview too is great for doing that for subsequent days but not so good if I want to compare random dates further apart

Shared Humanity

  • Guest
Re: Interpretation of satellite images
« Reply #2 on: May 21, 2017, 10:43:39 PM »
Thank you Andreas. I'm going to dig in to figure this technique out.

johnm33

  • Guest
Re: Interpretation of satellite images
« Reply #3 on: May 21, 2017, 10:54:41 PM »
SH Open worldveiw AndreasT layers are a good choice, top right there's a camera, click, pick your shot, download will open into a new tab to copy into gimp or imageJ, alternatively copy the tab address and paste that into 'insert image' second icon second row [mona lisa], click it and img /img appears, but both in brackets, check with preveiw. Another alternative, on worldveiw, is to click the chain/links, copy the shortened link and paste that using 'insert image', if you want to switch between layers click the icon [eyeball] on the left of layers/base layers, that should get you started.

Neven

  • Administrator
  • First-year ice
  • Posts: 9470
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 1333
  • Likes Given: 617
Re: Interpretation of satellite images
« Reply #4 on: May 21, 2017, 11:22:52 PM »
What really counts is a comparison with previous years, which is why I created the Sea ice concentration maps page on the ASIG.

I used to use the LANCE-MODIS Arctic Mosaic for comparisons, ie download images from different years (or several days in a row for a comparison), crop them and add a date etc in Photoshop (I use templates/actions for that now) and then use GIF animation software.

I'm going to have to develop a method to do the same with WorldView when I resume blogging, as I'm not sure if the Arctic Mosaic will be around much longer.

I can't volunteer to organize this as a group effort, as I'm on a 'sabbatical', but I may in the future. Now, if could program and develop websites, I'd make a simple one where you can compare one particular region for different years. Maybe in a next life.  ;)
The enemy is within
Don't confuse me with him

E. Smith

Reallybigbunny

  • New ice
  • Posts: 82
    • View Profile
  • Liked: 8
  • Likes Given: 57
Re: Interpretation of satellite images
« Reply #5 on: May 21, 2017, 11:32:34 PM »
I am happy to report on particular coordinates on a regular basis using worldview.  8)

Andreas T

  • Nilas ice
  • Posts: 1149
    • View Profile
  • Liked: 18
  • Likes Given: 4
Re: Interpretation of satellite images
« Reply #6 on: July 12, 2017, 12:05:07 AM »
The higher resolution sentinel images are more promising for the purpose of finding out more about ice properties I think, but they are not available north of 83deg latitude. But there are intersting areas south of that.
I have looked at Parry channel in the North West Passage where Obuoy14 is sending images. Hoping to compare last years break up with the expected one this year, I have found this sequence of 6. to 19. July 2016.
http://apps.sentinel-hub.com/sentinel-playground/?lat=73.94071394020898&lng=-104.622802734375&zoom=7&preset=1_NATURAL_COL0R&layers=B04,B03,B02&maxcc=45&gain=0.6&gamma=2.1&time=2015-01-01|2016-07-12&cloudCorrection=none&atmFilter=&showDates=false&evalscript=
What I find interesting is that the initial cracks run across the older floes (remnants of the previous melting season) Only when the tension in the ice has been relieved by the first cracks, does the thinner (darker in the images) ice break between the thicker floes preferentially.
animation may need click to run
« Last Edit: July 12, 2017, 09:12:47 AM by Andreas T »

J Cartmill

  • New ice
  • Posts: 84
    • View Profile
  • Liked: 30
  • Likes Given: 16
Re: Interpretation of satellite images
« Reply #7 on: July 12, 2017, 09:24:03 PM »
Noticed today that EOSDIS Worldview images now go all the way back to 2009.


Andreas T

  • Nilas ice
  • Posts: 1149
    • View Profile
  • Liked: 18
  • Likes Given: 4
Re: Interpretation of satellite images
« Reply #8 on: July 14, 2017, 07:37:39 PM »
I have been looking for areas of interest which can be investigated more closely with the high resolution Sentinel 2 images. One place which I have looked at earlier is north of Svalbard where ice has been compacting against the island for much of the last 4 months. A recent cloudfree image, worldview at max resolution of the 2. July shows some cracks with the typical hexagon pattern which  has been  pointed out as a sign of strong ice in the central arctic.
https://go.nasa.gov/2uhTnAs
looking at this with Sentinel I found that some of that ice has moved into that triangle of compacter ice quite recently (after 25. June).  I have marked a group of floes which remained recognizable over those dates.
http://apps.sentinel-hub.com/sentinel-playground/?lat=80.51918613079921&lng=18.090362548828125&zoom=9&preset=1_NATURAL_COL0R&layers=B04,B03,B02&maxcc=96&gain=0.5&gamma=1.1&time=2015-01-01|2017-06-25&cloudCorrection=none&atmFilter=&showDates=false&evalscript=
This demonstrates what the level of detail in Sentinel 2 can add. This area also has frequent Sentinel1 images