Support the Arctic Sea Ice Forum and Blog

Author Topic: Arctic Image of the Day  (Read 894357 times)

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #400 on: January 04, 2015, 04:09:03 PM »
Quote
the quick-look images are false colored: Horizontal for Red, Vertical for Green and and average for Blue, or at least so they say.

It seems that some aspect of the ice flow is polarized (bright yellow). I took apart the original image above into its RGB components. The red is fairly featureless (pure white in the flow); the green captures the anomaly as a brighter radar reflection (rougher, more crevasses, flow lines?); the blue originally appeared completely black and utterly featureless.

However after pounding on it with gimp's "Equalize' [histogram], features (bottom panel) popped out again. However the central black of your flow is still a pure black. Returning the modified threesome of grayscales to RGB was thus not so informative (not shown).

Setting the red channel to 50% transparency over 100% green gives the average (not shown). This average has no apparent relationship to the blue channel. Replacing the entirety of the blue channel with pure black or shades of gray shows the six dark grays in the original blue is not doing some subtle wizardry -- it is almost indistinguishable from reconstitutions with pure darker grays.

To summarize, I'm not real sure what they did here or why.
« Last Edit: January 04, 2015, 04:51:50 PM by A-Team »

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Image of the Day
« Reply #401 on: January 05, 2015, 05:47:14 AM »

A quick survey of the stratigraphy and ocean bottom topography strongly suggest to me a large-scale land slip would be improbable to impossible.

Thank you for detailed reply to my silly speculation.

It was a reasonable question, and you are welcome.  I just happened to be in the rare position of actually having something resembling expertise that could be applied in a reasonable answer.

;)
This space for Rent.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #402 on: January 05, 2015, 03:56:25 PM »
Quote
the quick-look images are false colored: Horizontal for Red, Vertical for Green and and average for Blue, or at least so they say.

...

To summarize, I'm not real sure what they did here or why.

I had some doubts that the documented algorithm was used.

Quote
Quicklook Generation #

Quicklooks are lower resolution images of the product data used to preview the data. Quicklooks are generated by power detecting, averaging and decimating in both azimuth and range directions by a configurable amount. IW and EW SLC product quicklooks are first de-burst and merged.
Single polarisation products are represented with a grey scale image. Dual polarisation products are represented by a single composite color image in RGB with the red channel (R) representing the first polarisation, the green channel (G) representing the second polarisation and the blue channel (B) representing an average of the absolute values of the two polarisations.
Quicklook images are scaled to 8 bit and saved in PNG file format.

Yellow is Blue+Red, so I has a look in gimp myself: the Blue channels is near zero ( zero's and 1's) everywhere. So I think that the Red and Green channels are represented by the H and V polarizations and Blue is either miscalculated or intentionally turned down.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #403 on: January 05, 2015, 11:02:42 PM »
Quote
Quicklooks are generated by power detecting, averaging and decimating in both azimuth and range directions by a configurable amount. IW and EW SLC product quicklooks are first de-burst and merged.  the blue channel (B) representing an average of the absolute values of the two polarisations.
Better that they provide a real explanation to real users -- someone who knows what all this insider jargon means would not be reading the help page in the first place: "power detecting, averaging and decimating in both azimuth and range directions by a configurable amount. de-burst and merged"

Quote
Blue is either miscalculated or intentionally turned down.
I don't see how adding two reasonably ranging positive numbers from R and G could produce 99.6% of the pixels having a grayscale less than 15. They must have crammed values down to the low end with an extreme use of the 'curves' tool in gimp or counterpart. The histogram below is for S1A_EW_GRDM_1SDH_20150103T060735_20150103T060809_004003_004D21_249B.quicklook.png

But if this was just for the pleasing red and yellow optics, they could simply have taken blue identically equal to 1 on the grayscale and gotten something visually indistinguishable, at least on this image. Been there, done that.

But the 'average of the absolute values of the two polarisations' could be a valid re-normalization effort, the purpose being to tone down excesses in the red-green channels. Maybe in some images this takes out gratuitous non-uniformity and mitigates noisy pixels.

Quote
I had some doubts that the documented algorithm was used [for blue channel]
Maybe they will give you a free ride on the space shuttle if you alert them to this problem?

Surely they intended square root of sum of the squares rather than sum of absolute values. Then to calculate the overall mean of the whole image and its std deviation, to divide for the multiples of it. These would reside in some small range mostly at the low end, which could be stretched to [0,255] to assign to each pixel in the blue channel.

Finalize the file system and downloads by Q2 2015?!? Holy toledo, that should have been finalized -- using synthetic data -- on the day of the launch. Parallel processing, not serial. They knew exactly what the data would look like the day the design was finalized.
« Last Edit: January 06, 2015, 12:27:44 AM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #404 on: January 07, 2015, 10:01:49 AM »

Surely they intended square root of sum of the squares rather than sum of absolute values.


Not if the value represent power (instead of amplitude). I bet they are using power because of the "power detecting" phrase.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #405 on: January 07, 2015, 12:03:45 PM »
Quote
the blue channel representing an average of the absolute values of the two polarizations.

Power would already be positive, no need for absolute values. Maybe they mean 'average of the absolute value squared 1/2((pol1 x pol1*)+(pol2 x pol2*)) of the two polarizations.' I'm still going with ((pol1 x pol1*)+(pol2 x pol2*))exp1/2, not 1/2[((pol1 x pol1*)exp1/2+(pol2 x pol2*)exp1/2].

In any case, they have not provided a satisfactory explanation of the algorithm nor do we know what was actually applied. It is almost always better to describe a process through an concrete example.

My experience is that when the writer cannot explain clearly what was done, the writer does not really know what was done. Here I picture a self-important engineer giving a harried explanation as he walks out the door to a lowly contract tech writer.

Looking again at the 4th panel above in Reply #400, the blue channel constant color islands do make some sense in the central region though two ragged vertical bars do not have an obvious interpretation. The actual blue channel used has no visual consequences at all so it must serve some 'politically correct' purpose in the scientific imaging context.

In other words, the blue channel is just coming along for the ride but might really be used for something later. It reminds me of Howat using grayscale channels on a DEM to carry the error and date (source) of each pixel in an image, not for making an RGB but to provide (alpha channel) masking.

Peter Ellis

  • Grease ice
  • Posts: 619
    • View Profile
  • Liked: 33
  • Likes Given: 14
Re: Arctic Image of the Day
« Reply #406 on: January 07, 2015, 02:16:50 PM »
If you're measuring polarisation in two orthogonal directions then shouldn't the average be identical for every pixel?  That's kind of how polarisation works (if I remember my physics from 20 years ago - I'm really a biologist!)

In that case I imagine the blue channel is simply there as a QC channel to spot and exclude contaminated data, and it's deliberately kept to low values so as not to pollute the visual image with QC data.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #407 on: January 08, 2015, 02:30:45 PM »
In the histogram panel above, the pixel values in the blue channel are crammed in to [0,16) which to my mind is not coincidentally the square root of the full range of possibilities, 256. (Note ImageJ2 offers sq roots of images.)

I think there is merit in your noting A and B are orthogonal for cross terms in the dot product (A,B)^2. The blue channel might be QC, a carried-along storage channel, a weighted masking device, a correction for very low power pixels, or an advanced statistical tool needed for rational scientific adjustment beyond our mere aesthetic ogling.

If we find all the quicklook blue channels are confined to (0,16), that would be a strong clue. I looked just now at a Nares region quicklook posted by Wipneus, http://forum.arctic-sea-ice.net/index.php/topic,176.msg42492.html#msg42492

The blue channel range there is [0,9] so still below 16. There are 876 pixels right at 0 for 0.2% of the 415,100 = 700 x 593 image whereas the peak at 2 has 171,912 pixels for 44.1%. The effect of adding a bit of blue to a predominantly yellow = red + green is to push the pixel over towards the diagonal of the color cube (gray).

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #408 on: January 08, 2015, 03:25:27 PM »
Making things more confusing I found that the images presented in the graphical search application, and also called "quicklook" are different in this respect. I already knew that they where jpeg instead of png, but the blue channel is fully developed.
Filenames are identical except the documented "quicklook" ends in ".quicklook.png" and these images have "-ql.jpg" endings.

Example, the Severnaya Zemlya" islands yesterday:


A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #409 on: January 08, 2015, 05:01:22 PM »
Quote
Making things more confusingimages presented in the graphical search application also called "quicklook" are different. I already knew that they where jpeg instead of png, but the blue channel is fully developed.Filenames are identical except "quicklook" ends in ".quicklook.png" and these images have "-ql.jpg" endings.
Please, please can someone email the ESA help desk to put us out of our misery? Or before they did themselves into a deeper archiving hole

One difference, as you can see below, is they apparently take the square root of the blue channel in what we have been looking at so far. (You can see the same vertical artifact bars in what I posted earlier for Svalbard.) But I am quite concerned about their poor file naming system and above all the possible use of lossy jpeg. These quicklooks are fairly small files that can be losslessly compressed with png.

In the second image below, I replaced the blue channel grayscale with its square root and reconstituted as RGB, with more or less the expected outcome, probably off by a scalar.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #410 on: January 09, 2015, 12:04:58 PM »
Quick-look panorama of the West Greenland coast.

(click to enlarge, but probably won't fit on your single monitor)

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #411 on: January 09, 2015, 02:11:03 PM »
Gorgeous. Worth noting that top is east and that Jakobshavn and Disko Bay are dead center. Blue channel is, as before, <16. I'm going to try displaying in vertically as width is just 526 pixels... no, blog will does not display that tall even at that width even compressed to a 440 k jpeg. The overlay below has north on top.
« Last Edit: January 09, 2015, 02:26:52 PM by A-Team »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #412 on: January 09, 2015, 03:04:49 PM »
Below I  'sentinelized' the 07-08 Jan scenes of Nares Strait. That is, the three gray scales were taken as red the 8th, green the 7th and blue just straight black, then composed into RGB color. It kinda shows the new developments in red and yesterday in green. Yellow is where not much changed.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #413 on: January 10, 2015, 03:38:27 PM »
The Sentinel radar uses C-band microwave reflection that is supposed to be rather insensitive to clouds and water vapor. Nevertheless a small cyclone is clearly visible. Perhaps what we are seeing are waves? To the left Greenland, Iceland is just south of this image and the island is Jan Mayen.
 

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #414 on: January 10, 2015, 05:09:08 PM »
It is worth marking up some of this Sentinel imagery as we gradually figure out what the satellite is seeing (and why).

Here, I am thinking strong wind from the east, a blocking of clouds and a calming effect on waters in the lee of Jan Mayen. With clouds, it seems like there should be more of a difference due to island elevation, least at the saddle, but not so much for calming of water. Not seeing von Kármán vortex eddies, which would be in clouds. They seem to be having technical difficulties in the green channel...

https://en.wikipedia.org/wiki/K%C3%A1rm%C3%A1n_vortex_street

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #415 on: January 11, 2015, 03:06:14 PM »
Part of the explanation seems to be the much lower HV signal of open sea compared with land/ice.
From the data files, the average pixel value of the HH image is 504 (16-bit values here), for the HH image it is 91 or about 18%. For comparison an image of NE Greenland, with land ice, rocks and sea ice the average HH pixel value was 1067 and for HV a value of 551, about 50%.

(Channel values are signal amplitudes rather than intensities.)

The attached image of Jan Mayen is created from the HH values with some linear stretching applied.
Compared with an optical Landsat image a typical distortion in SAR images called foreshortening can be seen. The slope of the Beerenberg facing the satellite (which is on the right of this image) is compressed while the slope on the other side has been stretched. The NE Beerenberg slope is more than twice as bright as anything I see in the NE Greenland image.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #416 on: January 12, 2015, 04:04:57 PM »
I cannot help thinking that ESA thinks the Zachariae Isstrom and Nioghalvfjerdsfjorden must be one of the most uninteresting places in Greenland. In the last 7 days they managed to produce 74 Sentinel 1A images of the East Coast of Greenland, none covering the two giant glaciers.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #417 on: January 12, 2015, 04:49:55 PM »
It is horrifying to think they might just tossing the raw data for Zachariae. I realize that these satellites produce too many terabytes to store everything and a lot of what they do store will never be requested by a user. Still, maybe a polite letter (citing # of goog scholar Zach studies)?

Meanwhile, your west coast quick-tile was not only very instructive for the land and marine terminating glaciers, but also for the interior ice (mostly shown in the north). In fact the pitmarks and ice divides bore an uncanny resemblance to the surface elevation DEM. I would like to see the coast strip widened to get more of the interior. However one just downloaded for JI was completely blown out in the green interior (pure 255),

I think there would be quite a 'market' for a dramatic red/green image of the entire coastline as well as the interior. Seems like I get different responses to coverage whether I draw a large or tiny rectangle at the search interface. I wonder if it has a small max for the number of responses it seeks to fill a query. https://scihub.esa.int/dhus/

Frustrating to see a little thumbnail icon on the search results but have no way of seeing it at reasonable scale, short of a giga-download (which I can only do, from Excede satellite, between 1-5 am in the morning!).

I still don't see how I go about making my own full scale quicklook from the R and G channels, in this case s1a-ew-grd-hh-20141201t095119-20141201t095223-003524-004252-001.tiff and s1a-ew-grd-hh-20141201t095119-20141201t095223-003524-004252-003.tiff in the 'measurement' subfolder.
« Last Edit: January 12, 2015, 07:53:42 PM by A-Team »

Apocalypse4Real

  • Frazil ice
  • Posts: 370
    • View Profile
  • Liked: 3
  • Likes Given: 1
Re: Arctic Image of the Day
« Reply #418 on: January 13, 2015, 04:41:00 AM »
The attached is the Metop 1 AVHRR image of the Nares Strait sea ice collapse over the last few days. I do not recall seeing this in January in the past.

A4R

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Image of the Day
« Reply #419 on: January 13, 2015, 08:27:55 AM »
Nares?  Heck, the Beaufort isn't looking particularly happy, Either...
This space for Rent.

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: Arctic Image of the Day
« Reply #420 on: January 13, 2015, 12:25:07 PM »
Nares?  Heck, the Beaufort isn't looking particularly happy, Either...

An extract from the Canadian Ice Service's latest RADARSAT mosaic:

"The most revolutionary thing one can do always is to proclaim loudly what is happening" - Rosa Luxemburg

Apocalypse4Real

  • Frazil ice
  • Posts: 370
    • View Profile
  • Liked: 3
  • Likes Given: 1
Re: Arctic Image of the Day
« Reply #421 on: January 13, 2015, 02:00:08 PM »
The Beaufort was broken up by Nov 27 2014, and has gone through waves of fracturing since. Here is the 112714 view.

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Image of the Day
« Reply #422 on: January 13, 2015, 05:44:39 PM »
Thank A4R, Jim;

I'd gotten out of the habit of looking at the Canadian Weather Service winter IR and Radar imagery.  I need to start that back up again.  EOSDIS got me spoiled, and I'd stopped browsing once it went dark.

Is it as much that the Beaufort fractured in late November, or that it simply hasn't consolidated since reaching minimum?  There was an immense expanse of open water across the Chukchi and ESS until quite late.  That's reflected in the rather low ice thicknesses across those regions.  Regardless, its not healthy.
This space for Rent.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #423 on: January 23, 2015, 02:08:53 PM »
A bit late (image is from 2014-12-24) but the image is worth looking at.
Composed with 4 quick-look images, in the lower left corner the Mackenzie river estuary. To the right surprisingly clear broken sea ice and in between fast ice, dark because it is flat.

Espen

  • Young ice
  • Posts: 3707
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Arctic Image of the Day
« Reply #424 on: January 24, 2015, 09:01:03 PM »
Here is something fishy, below Ross Ice Shelf (found 850 km beyond the calving front) :
« Last Edit: January 24, 2015, 09:13:41 PM by Espen »
Have a ice day!

Espen

  • Young ice
  • Posts: 3707
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Arctic Image of the Day
« Reply #425 on: February 18, 2015, 07:13:19 PM »
Iceberg shadows:

The sun is still low, but the shadows are great, here is an image from Scoresby Sund, February 18 2015, courtesy of Landsat and Daugaard-Jensen Gletscher (there might be a piece or two from Charcot Gletscher and F. Graae Gletscher):
« Last Edit: February 18, 2015, 07:28:14 PM by Espen »
Have a ice day!

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #426 on: February 19, 2015, 09:32:21 AM »
Laptev Feb 13, close up of the Arctic "ice factory".  Processed using ImageMagick and Gimp. Red is HH polarization, Green is HV and Blue is none.

(click for the big picture)
« Last Edit: February 19, 2015, 09:39:57 AM by Wipneus »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Arctic Image of the Day
« Reply #427 on: February 20, 2015, 04:46:14 PM »
There are more things in heaven and earth, oh Modeller,
Than are dreamt of in your philosophy. [Hamlet, scene v, updated]

These are three consecutive north-south slices of ice penetrating radar just below the Petermann Glacier proper and aligned with its flow. The overall size of the Eemian ice upheaval feature exceeds 800 km3.
« Last Edit: February 20, 2015, 04:54:47 PM by A-Team »

pikaia

  • Frazil ice
  • Posts: 395
    • View Profile
  • Liked: 121
  • Likes Given: 39
Re: Arctic Image of the Day
« Reply #428 on: February 22, 2015, 10:32:54 AM »
Island in a Lake on an Island in a Lake on an Island!



http://earthobservatory.nasa.gov/IOTD/view.php?id=85342

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Arctic Image of the Day
« Reply #429 on: February 22, 2015, 06:46:58 PM »
The Sentinel radar uses C-band microwave reflection that is supposed to be rather insensitive to clouds and water vapor. Nevertheless a small cyclone is clearly visible. Perhaps what we are seeing are waves? To the left Greenland, Iceland is just south of this image and the island is Jan Mayen.
The radar is highly sensitive to sea surface roughness so that it probably what you are seeing.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #430 on: February 23, 2015, 06:24:40 AM »
Nice Sunday shot of (very) green Greenland and the new cyclone south of Iceland.

[]

Pmt111500

  • Guest
Re: Arctic Image of the Day
« Reply #431 on: February 23, 2015, 10:07:00 AM »
Island in a Lake on an Island in a Lake on an Island!



http://earthobservatory.nasa.gov/IOTD/view.php?id=85342

That's nice. I found a couple of examples (from Finland) that are islands in a lake in an island in a lake, the largest was be c. 1.75km long but that one takes the cake being one more I guess  :)

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Arctic Image of the Day
« Reply #432 on: March 10, 2015, 03:50:10 PM »

Espen

  • Young ice
  • Posts: 3707
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Arctic Image of the Day
« Reply #433 on: March 22, 2015, 06:04:02 PM »
Sperm cells captured from the 79 Gletscher:
Have a ice day!

Espen

  • Young ice
  • Posts: 3707
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Arctic Image of the Day
« Reply #434 on: March 24, 2015, 08:41:32 PM »
The "tropical" island of Jan Mayen with its lagoons along the coast:
Have a ice day!

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Arctic Image of the Day
« Reply #435 on: March 25, 2015, 06:12:44 PM »

Sonia

  • New ice
  • Posts: 57
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #436 on: March 26, 2015, 12:51:57 AM »
Detail from above link.  I have no idea what would make the loopy line.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #437 on: March 26, 2015, 01:13:45 AM »
Detail from above link.  I have no idea what would make the loopy line.
If I were a space alien tasked with making intriguing patterns in corn fields, I'd certainly consider switching to ice graffiti, if I knew how eagerly sea ice was being watched from a growing number of satellites.
[]

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Image of the Day
« Reply #438 on: March 26, 2015, 06:31:09 AM »
Detail from above link.  I have no idea what would make the loopy line.
If I were a space alien tasked with making intriguing patterns in corn fields, I'd certainly consider switching to ice graffiti, if I knew how eagerly sea ice was being watched from a growing number of satellites.
Never ascribe to mystery what can be explained by a failure of quality control. It looks like a hair or filament of some sort, that got into the mechanism past the inspectors.
This space for Rent.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #439 on: March 26, 2015, 06:52:27 AM »
I very much doubt it's a hair, JD. How do you suggest a hair got into these computer images? If a real snow/ice feature, I think we can all agree that it must have been some continuous movement, probably by wind (I've seen natural snowballs blow this way in changing wind), but the scale here is so enormous that I'm utterly mystified. Any idea how broad this looping band is?
[]

gideonlow

  • New ice
  • Posts: 33
  • Fast Data/Big Data Systems Architect
    • View Profile
    • Pivotal Initiative -- Next Generation Data Technology
  • Liked: 0
  • Likes Given: 1
Re: Arctic Image of the Day
« Reply #440 on: March 26, 2015, 08:43:44 AM »
Attached is the March 25th Aqua-Modis 7-2-1 close-up for the same spot.  Something loopy going-on here?  Occam's Razor would certainly suggest this is caused by something in the image processing chain . . .

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #441 on: March 26, 2015, 09:52:10 AM »
The scribble is visible on Sentinel images from Polar View as well (but also directly from the public archives of ESA). I see the feature moving but not with the ice. So somehow it is an artifact, although I don't see how: hair on the lens is not an option.

(click to download a 3MB animation)

Peter Ellis

  • Grease ice
  • Posts: 619
    • View Profile
  • Liked: 33
  • Likes Given: 14
Re: Arctic Image of the Day
« Reply #442 on: March 26, 2015, 11:28:57 AM »
Don't see why hair on the lens isn't an option. Dirt and contamination is an issue in any mechanism, and this is no exception.  Whatever it is, it's obviously mobile, which explains why it's not been noticed before - presumably it wasn't in the way of the sensor in previous images from the same source.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #443 on: March 26, 2015, 11:59:00 AM »
Don't see why hair on the lens isn't an option. Dirt and contamination is an issue in any mechanism, and this is no exception.  Whatever it is, it's obviously mobile, which explains why it's not been noticed before - presumably it wasn't in the way of the sensor in previous images from the same source.

There is no lens involved. The imaging in SAR is purely in the processing software, based on the timing of the returned signal. Any obstruction of the antenna (beam) would affect all imaged pixels.
(as far as I know of course)

Jim Pettit

  • Nilas ice
  • Posts: 1175
    • View Profile
  • Liked: 41
  • Likes Given: 11
Re: Arctic Image of the Day
« Reply #444 on: March 26, 2015, 01:50:23 PM »
I don't believe the curlicue line is an instrumental artefact of any kind, for portions of it stay in place between images--that is, over precisely the same parts of the ice even as the ice itself moves and shifts. However, the line appears to evolve lengthwise over time, with older parts of the curlicue disappearing even as newer parts are created. My guess, then: an icebreaker path (maybe continuing sea trials for Russia's sideways-travelling Baltika?)
« Last Edit: March 26, 2015, 02:14:14 PM by Jim Pettit »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #445 on: March 26, 2015, 02:18:54 PM »
Wonderful, I think you nailed it Jim.

Neven

  • Administrator
  • First-year ice
  • Posts: 9470
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 1333
  • Likes Given: 617
Re: Arctic Image of the Day
« Reply #446 on: March 26, 2015, 05:04:58 PM »
It's Cyrillic. Rough translation: Masha, will you marry me?  ;) ;D
The enemy is within
Don't confuse me with him

E. Smith

Espen

  • Young ice
  • Posts: 3707
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Arctic Image of the Day
« Reply #447 on: March 26, 2015, 05:37:31 PM »
And now it is spring, the watering hole for all Polar Beers is now opening their service at Kap Jackson / Washington Land:
Have a ice day!

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #448 on: March 26, 2015, 06:13:05 PM »
I don't believe the curlicue line is an instrumental artefact of any kind, for portions of it stay in place between images--that is, over precisely the same parts of the ice even as the ice itself moves and shifts. However, the line appears to evolve lengthwise over time, with older parts of the curlicue disappearing even as newer parts are created.

Good observation, this is what I see, too: A movement from the right edge to the left, and fresher trails on the left.

Quote
My guess, then: an icebreaker path (maybe continuing sea trials for Russia's sideways-travelling Baltika?)

No space–aliens?! Dang!  :D
[]

Nightvid Cole

  • Frazil ice
  • Posts: 438
    • View Profile
  • Liked: 5
  • Likes Given: 0
Re: Arctic Image of the Day
« Reply #449 on: March 29, 2015, 04:00:29 PM »
Island in a Lake on an Island in a Lake on an Island!



http://earthobservatory.nasa.gov/IOTD/view.php?id=85342

69.7923954,-108.2398877 for anyone who wants to look on Google Maps...

It's a sub-sub-island (only two "subs".)