Support the Arctic Sea Ice Forum and Blog

Author Topic: S-1 Greenland campaign starting 16.1.2015  (Read 14082 times)

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
S-1 Greenland campaign starting 16.1.2015
« on: January 15, 2015, 12:43:50 PM »
ESA:s S-1 acquisition campaign over Greenland is starting tomorrow. The plan is to cover the whole Greenland in IW-mode from both ascending and descending orbits in this 12-day cycle, and to repeat these acquisitions for the two following cycles. If everything goes well we will have ascending and descending image triplets covering the whole Greenland.

All data will be processed into SLC and GRDH and will be placed available on scihub. As scihub is a rolling archive, please grab the data before it will be lost..

https://sentinel.esa.int/web/sentinel/missions/sentinel-1/observation-scenario

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #1 on: January 15, 2015, 01:57:18 PM »
Great news, nukefix, I think that will be exciting to follow particularly this winter.

PS: Any hint on where one can find MODIS/sentinel type images for Western Norway? I have a favourite glacier there which always seems to be under a cloud on Google Earth.
[]

Jim Hunt

  • First-year ice
  • Posts: 6268
  • Don't Vote NatC or PopCon, Save Lives!
    • View Profile
    • The Arctic sea ice Great White Con
  • Liked: 893
  • Likes Given: 87
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #2 on: January 15, 2015, 02:15:15 PM »
PS: Any hint on where one can find MODIS/sentinel type images for Western Norway? I have a favourite glacier there which always seems to be under a cloud on Google Earth.

You could try the Landsat 8 archive at http://earthexplorer.usgs.gov/. You can specify the acceptable level of cloud cover!

N.B. I just tried that experiment myself, and received the following message:
Quote
On December 19, 2014, the Landsat 8 Thermal Infrared Sensor (TIRS) was reconfigured due to the detection of anomalous current levels associated with the scene select mirror (SSM) encoder electronics. The SSM electronics have been turned off with the instrument pointed at nadir.  Over the upcoming weeks, the calibration/validation team will continue to investigate whether updated calibration parameters can be derived.
 
Landsat 8 Level 1 data products continue to be generated and contain valid Operational Land Imager (OLI) data; however the TIRS band data will consist of zero-fill and will be unusable at this time.
"The most revolutionary thing one can do always is to proclaim loudly what is happening" - Rosa Luxemburg

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #3 on: January 15, 2015, 03:10:20 PM »
Landsat-8 is just now seeing enough light to make picture-taking worthwhile in southern Greenland. By trial and error, I determined that 12 Jan is the first date and the latitude shown (62.8º) is the coverage farthest north. I recall that in 2014 they first got up to Jakobshavn on the 9th of February.

It will be very cool to have a pair of matching low sun angle shots a full year apart in terms of being able to measure icesheet velocity in the slower regions west of the summit ridge. More features are auto-detectable because of enhanced shadowing. Farther down the ice stream, the movement may be too extreme to make a year-on-year contour map of surface displacement.

So mark your calendars, practice your auto-correlation software; with cloud-free conditions we can beat the journal types out the door with some original research.

LC80090112014040LGN00
Coordinates: 69.60632,-49.42945
Acquisition Date: 09-FEB-14
Path: 10
Row: 11
Start 15:07:22
Sun Elevation    8.13135887
Sun Azimuth    172.51552474

I added some directly determined velocities and elevations at Summit, Greenland. These were done about 15 years ago with fiberglass stake movement over the course of a year. Note how the ice is moving off the summit in all directions but it would be problematic to extend the 13 data points to an actual velocity field.

The image is about 120 km top to bottom with a 100 m elevation change. It might be fairly hard even today to measure 1-2 m/yr velocities and even harder to detect any velocity difference.

It's shocking to see the primitive graphics software of that era and the necessity to present them as expensive photographic plates. (I added the color here in gimp with a radial fill constrained to the irregular boundary.)
« Last Edit: January 15, 2015, 06:04:59 PM by A-Team »

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #4 on: January 15, 2015, 03:12:59 PM »
Great news, nukefix, I think that will be exciting to follow particularly this winter.

PS: Any hint on where one can find MODIS/sentinel type images for Western Norway? I have a favourite glacier there which always seems to be under a cloud on Google Earth.
As Norway is part of the regular S-1 acquisition-plan it is acquired frequently and you can download the scenes from:

https://scihub.esa.int

Under Advanced Search you should select IW for Sensor Mode (higher resolution than EW) and GRD for Product Type (as it sounds like you will not be doing interferometry).

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #5 on: January 15, 2015, 06:16:54 PM »
Thanks, nukefix. I feel really stupid now. After my first half hour of registration and login I still haven't seen one single satellite image of Western Norway. Do you have a direct link like the humanly readable DMI site? http://ocean.dmi.dk/arctic/lincoln.uk.php

Maybe if I put in a full day's work of reading user guides and data descriptions I will get there myself, but this appears to be a very self–centered site that requires registration without actually showing the registered users any real data.

Thanks for the tip to go IW & GRD, though. Unfortunately it doesn't get me any closer to the images.
« Last Edit: January 15, 2015, 07:18:26 PM by viddaloo »
[]

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #6 on: January 15, 2015, 06:50:37 PM »
Quote
Maybe if I put in a full day's work of reading user guides and data descriptions I will get there myself, but this appears to be a very self–centered site

I had exactly that same frustrating experience: page after page after page of gibberish that bragged about some old launch, Europe's aspirations in space, program goals and applause for visionary EU leaders and agency partners but told me nothing. No examples, no explanation, no interpretive, no help. Drilling down endlessly through pages that said 'data' but didn't have any to finally get to somewhere that did.

They have a lot to learn from the Denmark site design in terms of giving users what they want, fast and fresh.

After your registration clears, find the search button along the top. This will take you to a map page. Click on the second globe with the miniature orange rectangle. Struggle with the too-small mercator map to get something in Arctic at a good resolution.

Then draw your rectangle as small as you can possibly make it, or you'll get swamped. Add the advanced terms like nukefix said. Now press the 'search' button to the lower right of the map.  Then prepare to download the whole 835 MB file. Do not drill down through their xml directory structure hoping to find a preview file, there's no way to restrict your download to the part that's showing. Each download folder will be filled with repetitive useless metadata.  Copy out your image and then delete their whole folder.
« Last Edit: January 15, 2015, 09:45:56 PM by A-Team »

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #7 on: January 15, 2015, 06:56:45 PM »
I try to Google Image Search the site for 'satellite images' and quite typically found a lot of pictures of the satellites themselves (being self–centered), while what I really would like was to be up there looking down on the Earth surface. I'm giving up now and don't really feel like visiting an ESA site ever again. IMO my tax money could have been used more wisely.

[]

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #8 on: January 15, 2015, 07:31:45 PM »
Quote
Maybe if I put in a full day's work of reading user guides and data descriptions I will get there myself, but this appears to be a very self–centered site

I had exactly that same frustrating experience: page after page after page of gibberish that bragged about some old launch, Europe's aspirations in space, program goals and applause for visionary EU leaders and agency partners but told me nothing. No examples, no explanation, no interpretive, no help. Drilling down endlessly through pages that said 'data' but didn't have any to finally get to somewhere that did.

They have a lot to learn from the Denmark site design in terms of giving users what they want, fast and fresh.

After your registration clears, find the search button along the top. This will take you to a map page. Click on the second globe with the miniature orange rectangle. Struggle with the too-small mercator map to get something in Arctic at a good resolution.

Then draw your rectangle as small as you can possibly make it. Add the advanced terms. Now press the 'search' button to the lower right of the map.  Then prepare to download the whole 835 MB file. Do not drill down through their xml directory structure hoping to find a preview file, there's no way to restrict your download to the part that's showing. Each download folder will be filled with repetitive useless metadata.  Copy out your image and then delete their whole folder.

I did the whole 800 MB + download, could only find a crappy preview image, so I gave up >:(
Have a ice day!

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #9 on: January 15, 2015, 10:23:28 PM »
Quote
I did the whole 800 MB + download, could only find a crappy preview image, so I gave up

Wise beyond your years, Espen.

I keep downloading over and again, thinking this week I will figure out something to do with those two 434 MB blurry monsters. I did manage to get them both open eventually in ImageJ, combine them into a stack so that they could be cropped identically, tried a few operations like contrast and GB, exported the stack as a gif for gimp and so on.

Now I have two blurry mini's in gimp that don't improve on that crappy little preview image you mentioned. And in terms of remote sensing re-processing, this isn't my first rodeo.

Is this 500 sq pixel preview the best they can make from a 400,000,000 pixel scene? A great swath of visitors would be satisfied with a decent red/yellow quicklook, especially if they could really explain the blue channel and why they bothered with it. How hard is it really to carve out the quicklook from the mega-download? How hard is to make a basic interferometric product?

I think they need feedback ASAP from a mixed users committee, not just insiders and specialists. This satellite will have a very finite lifetime and they have wasted a good chunk of it already. They have taken far too narrow a view of the potential scientific user community (the taxpaying public be damned of course).

Rolling buffer for Greenland campaign? You are kidding me? -- don't throw away data like this hoping a few individuals will maybe notice, catch it and share.  ESA, if you are too cheap to buy storage for a billion euro satellite project, open a paypal account where we can send twenty euros to have Amazon store it for a couple years.
« Last Edit: January 15, 2015, 10:30:39 PM by A-Team »

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #10 on: January 15, 2015, 10:35:35 PM »
By the way I am getting my 1,5m x 2m (Zachariae) and 1,3m x 1,3m (Petermann) Landsat images digitally printed on canvas on Tuesday courtesy of Wipneus and Landsat.  ;)

It is almost like being a teenager again, back then it was Led Zeppelin, The Who, Pink Floyd, Emerson Lake and Palmer and few others hanging on wall, now it is glaciers, becoming more eccentric by the age I suspect?
« Last Edit: January 15, 2015, 10:47:20 PM by Espen »
Have a ice day!

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #11 on: January 16, 2015, 02:55:43 AM »
Quote
1,5m x 2m (Zachariae) and 1,3m x 1,3m (Petermann) Landsat images digitally printed on canvas
Hey this is supposed to be a classy forum: giclée (faux french). Those sizes ... was there enough dpi to have good resolution? Could look awesome on the wall. Those quicklooks are quite dramatic too.

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #12 on: January 16, 2015, 06:35:31 AM »
Yes the resolution was ok, they could if my wall was big enough by twice the size, they will be in real colors.
Have a ice day!

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #13 on: January 16, 2015, 11:35:51 AM »

I keep downloading over and again, thinking this week I will figure out something to do with those two 434 MB blurry monsters. I did manage to get them both open eventually in ImageJ
It would be a lot easier to do opening, cropping, co-registration and other low-level operations in the Sentinel-1 toolbox and then export to the image post-processing software of your choice.

https://sentinel.esa.int/web/sentinel/toolboxes/sentinel-1

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #14 on: January 16, 2015, 11:43:43 AM »
I did the whole 800 MB + download, could only find a crappy preview image, so I gave up >:(
You should try the Sentinel-1 toolbox, it has a learning-curve but as it's designed to handle SAR data it is well suited to the task.

https://sentinel.esa.int/web/sentinel/toolboxes/sentinel-1

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #15 on: January 16, 2015, 12:38:52 PM »
The procedure for opening the S-1 zipped products in the S-1 Toolbox is the following:

1. Open S-1 toolbox
2. In the file-menu, select Import Raster Data/SAR Data/Sentinel-1 and import the product
3. The product appears in the Products View window. Click open the bands in the tree to display amplitude and intensity bands

In order to subset one should first display a band, which can be done by double-clicking on a band. Note that the image can be flipped due to the peculiarities of SAR imaging (if you want to fix this you should do SAR Processing/Geometric/Ellipsoid Correction/Average-Height Range-Doppler, but it's advisable that you subset first due to the huge size of the imagery. Amplitude looks better than intensity, or you might want to experiment with intensity in decibel-scale which you can create in the Product View by right clicking on an intensity-band and selecting Linear to/from dB.

Once you are seeing the band you want to use for subsetting/cropping you should zoom to the view you want, right-click on the image itself and select Spatial Subset From View, or if you want just a simple image-file instead of a satellite-product you can also select Export View as Image.

If you keep working with the subset in the S-1 toolbox you can do many things like experimenting with RGB-views, multilooking in order to reduce both resolution and speckle-noise, correct the scene geometrically into the map-coordinates of your choice. If you have a DEM you can orthorectify the scene and get rid of foreshortening that makes mountains look weird in SAR imaging. These images are not photographs like optical data so the learning-curve is steeper, but also rewarding :)

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #16 on: January 16, 2015, 12:43:37 PM »
I think they need feedback ASAP from a mixed users committee, not just insiders and specialists. This satellite will have a very finite lifetime and they have wasted a good chunk of it already. They have taken far too narrow a view of the potential scientific user community (the taxpaying public be damned of course).

Rolling buffer for Greenland campaign? You are kidding me? -- don't throw away data like this hoping a few individuals will maybe notice, catch it and share.  ESA, if you are too cheap to buy storage for a billion euro satellite project, open a paypal account where we can send twenty euros to have Amazon store it for a couple years.

So true, A–Team, and this is so important (and at the same time revealing).

If your satellite costs a billion euros, at least take the time to register for a free Google Drive account to store the daily previews on. Forever.
[]

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #17 on: January 16, 2015, 04:46:07 PM »
That sounds simple, but unfortunately it is not the way institutions in general will work or execute ::)
Have a ice day!

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #18 on: January 16, 2015, 08:11:31 PM »
Quote
not the way institutions in general will work or execute

I would guess that there are certain Big Customers out there, like the Principle Investigators by whom and for whom this thing was built, whose voices are heard. And then there Little Customers whose grievances are forwarded to a boiler room in India that primarily handles cruise ship reservations. If there is ever a job ticket generated from that, it goes (and stays) at the very bottom of the work queue.

We were hoping this thing could see through the dark and through clouds, generating regular time series all year long. It can but has just as many misses as Landsat for whatever technical reasons. I was hoping to see what the calving front at Jakobshavn but Sentinel is just not providing anything near enough optical resolution.

Some days I wonder if the train left the station before Sentinel could be launched. This is really something to have 0.5 m WorldView, as processed to 2 m stereo by Howat, in the public domain. And the 16 bit Landsat8 was also a game-changer, see T Scambos AGU2014 for that.


viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #19 on: January 17, 2015, 12:46:38 AM »
In the future, I'm hoping all NASA and ESA programs will lose 50% of their funding if they haven't got a plan or strategy for simple taxpayer use of their billion euro satellite imaging projects.

The only language they'll understand, I'm afraid.
[]

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #20 on: January 17, 2015, 07:55:17 AM »
I'd give them some benefit of doubt, it's still the very early days and the system is being ramped up to the ~1TB per day product generation rate. Sentinel-2 should also prove interesting as it should provide better imagery than Landsat.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #21 on: January 17, 2015, 03:02:42 PM »
Quote
I'd give them some benefit of doubt, it's still the very early days and the system is being ramped
Agree. And let's see what experts in this type of imagery can do with it in Greenland.

Can't save everything but I'd still be willing to toss a few coins in their tin cup for storage, here's a year's worth of storing everything for $6k.

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #22 on: January 18, 2015, 01:46:16 PM »

Is this 500 sq pixel preview the best they can make from a 400,000,000 pixel scene? A great swath of visitors would be satisfied with a decent red/yellow quicklook, especially if they could really explain the blue channel and why they bothered with it. How hard is it really to carve out the quicklook from the mega-download? How hard is to make a basic interferometric product?
What kind of quicklook would you be happy with? If the width of the original scene is 25000 pixels going down to a 500 pixel quicklook is compression by a factor of 50 and ending up with a pixel-size of 500m. Would compression by a factor of 25 with 250m pixels be enough?

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #23 on: January 18, 2015, 02:01:49 PM »
Quote
Would compression by a factor of 25 with 250m pixels be enough?
For making a 'movie' of winter movement of the Jakobshavn glacier, I am recalling the calving front as being ~6 km wide. So that would be 24 pixels, borderline ok. Twice that would be quite satisfactory.

Since JI runs east-west, it could be displayed with a 90º rotation to let us stay within the 700 pixel width blog constraint at good resolution with the height having more flexibility. Alternatively, a 1400-1900 pixel width (which would display at 700) but still be viewable on a 21" monitor without scrolling.

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #24 on: January 18, 2015, 02:19:51 PM »
250m pixel size grows the QL by a factor of four and 125m pixel-size by a factor of 16. I have some contacts in ESA and can inquire whether this poses a problem for the ground segment...

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #25 on: March 20, 2015, 05:41:23 PM »
A new version v1.1 of the S-1 toolbox is now available. This version also supports S-1 interferometry.

https://sentinel.esa.int/web/sentinel/toolboxes/sentinel-1

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: S-1 Greenland campaign starting 16.1.2015
« Reply #26 on: May 22, 2015, 03:05:24 PM »
A new Greenland campaign has started in Cycle 49, i.e. yesterday. The Greenland margin will be covered in IW-mode (~10m pixel size) in HH-polarization.