Support the Arctic Sea Ice Forum and Blog

Author Topic: PIG has calved  (Read 322143 times)

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #450 on: November 21, 2015, 05:27:19 PM »
Aha, we're making progress if the surface is concave upwards. Note the arches switch their curvature from positive to negative, passing through a zone where they are straight.

...

Something here does not quite make sense as the very fastest zones are shown next to the very slowest, that shear would isolate a visible ice stream. We do see a lagging zone in the arches but over on the other side.

While I am sure that we are in for even better visual treats from A-Team, nukefix, Wipneus and many others, A-Team's initial animations make several point relatively clear including:

(1) To state a few obvious points: (a) the direction of ice flow downstream due to gravity is roughly normal to the face of the transverse crevasses [where the principal tensile stresses cause the transverse crevasses]; and (b) the inflow of lateral ice streams, marginal shear, and bends in the flow directions all modify the stress fields in the main ice stream as it moves downstream, first through an area with a very flat surface slope and then into a zone with a steeper slope [see the image in #438] as the ice accelerates and thins.

(2) The ice "arches" between the transverse crevasses switch curvature from positive to negative [passing thru a zone where they are straight, as A-Team notes] because inflow of lateral ice from the North and South initially support the positive arch shape, but as the ice stream velocities near the middle of the field accelerate [relative to the margins], the arches first flatten and then develop a negative curvature. There could be a minor amount of compressive creep, and plastic deformation, in the ice within the arches; however, I suspect that the different speeds of the ice flow between the margins and the middle of the field is the main reason for the indicated behavior.

(3) The lagging zone to the South side of the arches (noted by A-Team) is most likely due to the fact that the catch basin to the South is much larger than to the North, and thus the larger inflow of ice flux slows the downstream ice velocities to the South in this area.

(4) Between nukefix's image & Wipneus' image we see new transverse crevasses forming upstream of the old crevasse (even while the old crevasses are slowly moving downstream) as the flow area is enlarging.  Furthermore, due to the relatively slow ice velocities upstream (compared to downstream) the width of the transverse crevasses remain narrow, because the flat sloped area (see image in #438) prevents the downstream ice arches from accelerating away from the upstream arches.

(5) The animation(s) also show marginal shear crevasses forming on the margins and at bends in the ice streams.
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #451 on: November 21, 2015, 05:40:19 PM »
To elaborate on some of my comments about the influence of the influx of lateral ice streams for the PIG; I provide the following linked reference and associated images discussing impacts on ice flow patterns and of compressive creep with time (note that the length of the arches cited by A-Team are longer in the zone of positive curvature and shortest when flat, so that as the arches are too thick and well confined to buckle, the length of the arches reduce by a combination of creep and vertical convection through the thickness of the ice streams (see the first image).  Also, I note that as Hughes (2012) focuses on the Byrd Glacier, that his notes about a subglacial lake are not relevant, but I do note that the vertical convective motion causes internal friction that causes ice melt to rain down through the PIG to contribute to basal melt water that flows down the bed stream and out into Pine Island Bay, thus lubricating flow and contributing to future accelerations of the ice flux.

Terence J. Hughes, (2012), " Thermal convection in ice sheets: New data, new tests", Natural Science, Vol.4 No.7, Article ID:20743,10 pages DOI:10.4236/ns.2012.47056

http://file.scirp.org/Html/1-8301678_20743.htm
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #452 on: November 21, 2015, 08:09:59 PM »
As this thread is named "PIG has Calved", I offer the following discuss about recent and future calving from the PIIS:

(1) I have expressed my opinion (which some argue with) in Replies # 218, 220, 225, 261 and 336 that recent and near-term (2016 to 2040) major calving for the PIIS is/will-be dominated by splitting tension stresses associated with compressive fields induced by lateral ice inflow from the Southwest, SW, Tributary Glacier forming a major rift in the ice shelf.  As a side-note I speculate that the next major rift will form a failure mechanism by about July 2016 with a major calving following sometime between July 2016 and November 2016, depending on boundary conditions and on whether fast sea ice is present.

(2) In Reply #449 I expressed my opinion that the current Super El Nino is promoting the advection of larger than typical volumes of warm Circumpolar Deep Water, CDW, beneath the PIIS; which in my opinion is currently causing the grounding line to retreat upstream relatively rapidly across a relatively flat portion of the PIG's bed.  Here I note that I expect a positive PDO to result in a period from 2015 to about 2035 with an above average number/strength of El Ninos that will keep the grounding line retreating upstream across this approximately 40 to 50km long flat portion of bed, where I expect the grounding line to become temporarily pinned by 2040 by the bottom ridge at the "stable position" indicated in the first image from Gladestone et al (2012).

(3) As the grounding line retreats upstream through about 2035-2040 I expect both basal melting (with irregular bottom grooves) and accelerations of local ice velocities to further thin the PIIS so that by 2035 to 2040 I expect the calving face to migrate upstream from the SW Tributary interface due to melt pond failures (see Reply #382, and DeConto, Pollard and Alley's hydrofacturing work).

(4) From 2040 to about 2060, I expect the PIIS calving face to progressively move upstream to the "stable position" shown in the first image.  Furthermore, I expect this calving behavior to be controlled by mechanisms discussed by Jeremy Bassis, of the University of Michigan, in the linked WAIS Workshop PPT and the second, third and fourth attached images; together with accelerating input from melt pond failures of the pre-formed transverse crevasses near the PIIS calving face as I expect the atmospheric CO2eq concentration to reach twice the pre-industrial levels in the 2038 to 2045 timeframe.

https://www.waisworkshop.org/sites/waisworkshop.org/files/files/agendas/2013/presentations/session6/Bassis.pdf

See also:
http://www.nature.com/ngeo/journal/v6/n10/full/ngeo1887.html
&
Heeszel, D. S., H. A. Fricker, J. N. Bassis, S. O'Neel, and F. Walter (2014), Seismicity within a propagating ice shelf rift: The relationship between icequake locations and ice shelf structure, J. Geophys. Res. Earth Surf., 119, 731–744, doi:10.1002/2013JF002849.

http://onlinelibrary.wiley.com/doi/10.1002/2013JF002849/abstract

(6) By 2060 I expect the PIIS to have disappeared and I expect that cliff failures and hydrofracting could thereafter (to at least 2100) cause the calving face to retreat relatively rapidly upstream depending on whether atmospheric conditions sustain periodic surface ice melting during the austral summers near the PIG calving face.
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #453 on: November 23, 2015, 10:52:25 AM »
The attached image from Mouginot et al. (2014) makes it clear that during the faux hiatus period (including the strong 2010/11 La Nina) the ice velocities of the PIIS slowed, on average, while the upstream areas of the PIG accelerated. This indicates that ice flux from the PIG is far from equilibrium; particularly as the current Super El Nino is likely currently accelerating the advection of warm CDW to the PIG grounding line and basal areas of the PIIS.

Caption: "Pine Island Glacier has a 30-kilometer-wide grounding line fed by nine glaciers. Thwaites Glacier has a 120-kilometer-long grounding line. To the west, a 60-kilometer-wide fast-moving portion of the Thwaites Glacier forms an ice tongue. To the east, a slower-moving portion of the glacier flows into an ice shelf buttressed by ice rumples. Bedrock mapping suggests that this buttressing wall is more easily breached than previously thought. These maps show flow-speed changes in Pine Island (a) and Thwaites (b) Glaciers. Red indicates greater increases in flow speed. The green lines indicate the position of the flow-speed contours four the years 2006-2013. Image courtesy (Mouginot et al. 2014)"

Mouginot, J., E. Rignot, and B. Scheuchl. 2014. Sustained increase in ice discharge from the Amundsen Sea Embayment, West Antarctica, from 1973 to 2013. Geophysical Research Letters 41: 1576-1584, DOI: 10.1002/2013GL059069

http://onlinelibrary.wiley.com/doi/10.1002/2013GL059069/abstract

Abstract: "We combine measurements of ice velocity from Landsat feature tracking and satellite radar interferometry, and ice thickness from existing compilations to document 41 years of mass flux from the Amundsen Sea Embayment (ASE) of West Antarctica. The total ice discharge has increased by 77% since 1973. Half of the increase occurred between 2003 and 2009. Grounding-line ice speeds of Pine Island Glacier stabilized between 2009 and 2013, following a decade of rapid acceleration, but that acceleration reached far inland and occurred at a rate faster than predicted by advective processes. Flow speeds across Thwaites Glacier increased rapidly after 2006, following a decade of near-stability, leading to a 33% increase in flux between 2006 and 2013. Haynes, Smith, Pope, and Kohler Glaciers all accelerated during the entire study period. The sustained increase in ice discharge is a possible indicator of the development of a marine ice sheet instability in this part of Antarctica."
« Last Edit: November 23, 2015, 11:11:20 AM by AbruptSLR »
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #454 on: November 23, 2015, 01:07:58 PM »
Some excellent analysis in the last posts, indeed we are looking at surface expression of the evolving upper trunk stress field.

The lower end can do whatever it wants but nothing happens to sea level from melting of a floating shelf or from grounding zone retreat per se. Models don't raise sea level. The trunk has to get in motion; otherwise the ice volume is just not there.

While we await the arrival of the future at Pine Island, the situation may be more simply and sensitively monitored by trending motion of trunk surface features: the opera hasn't started until the large lady sings.

A few more thoughts on our off-thread excursion: animations are not merely pleasing visuals but rather a study of the  time evolution of a dynamic system. The satellite record is precious so we seek to exhaust it informatically. Having expanded the past to the present in a power series, the near future will follow the fit.

Because this has to be done for every glacier in the cryosphere, we are looking for a large scale automated pre-compute. If a point-and-click algorithm exists (eg animated velocity interferograms for rolling pairs of Sentinel 1A's), then running it does not constitute original research and scientists doing so need redeployment. Factorization is a huge force-multiplier.

Pine Island's trunk seemed like a very favorable prototyping situation because it's so evenly gridded with surface markers. However I encountered a large number of pesky issues in making multi-year Landsat animations.

I'll skip over the ones where it says north is up when south pole is obviously up, where the central meridian (in Antarctica's polar stereographic coordinates PS) is provided in the scene portal but dropped out of the download, and where scenes of the same path,row come in software-disrupting petty dimensional variations such as:

path,row 229,114   path,row 232,113
17541 × 17641       17161 × 17241
17561 × 17601       17161 × 17261
17561 × 17621       17181 × 17261

Inland Antarctica is quite cloudy. Landsat-associated cloud metadata don't speak specifically to a region of interest (which is not fully known at the outset). EarthExplorer provides a very peculiar quality file (eg LC82291142015320LGN00_QB) that could serve as a cloud cover mask. It's not in the download packet though. The idea here is to intersect masks (ROI with QB blue) at the time of search to trigger a bulk download of the useful files instead of walking manually through terabytes of clouds.

I've got a question in to their help desk about how these are supposed to be co-registered (and why the heck doesn't USGS offer this in the first place). They will probably just pass me off on gdal but the problem is that the trunk of Pine Island doesn't have good ground control points -- everything is moving or if fixed, covered with drifting snow. How would I know that gdal is working?

The top of images get squeezed in PS and so the slightest error in registration becomes fully comparable to the actual motion between scenes. If a rotation is needed to bring everything to a common meridian, say 096ºW, that rotation would have to be about the south pole, which is some 1600 km beyond scene boundaries. Gimp is capable of an out-of-scene center of rotation but this might be a bit much.

Factorization ... the idea has not penetrated very far into geo.

In the first image, I sought to fix the A Hogg velocity map so that it might be compared to the clean offering of the Rignot group. However the color key was chosen poorly and this potentially simple task (10 seconds of gimp layer math) becomes unfeasible. Software with giant pull-down menus of pseudo-scientific color keys has been a huge source of problems in glaciology.

The next two images show, in Sentinel and Landsat respectively, crisscrossing of lineation fields in different years and regions. So the first round evidently has been over-written by later developments. I don't have the earliest pattern establishment dated. The Landsat also shows the interplay between the eolian snow drifts (?) and the lineations.

I'll post more of a time series after getting the registration issues resolved. These Landsats are fairly clear and span 715 days, much longer than we have from Sentinel. Some of the snow drifts are persistent but others come and go.

Looking into the 'display issue' vs the vast scale of Antarctica, a 27" retinal display iMac ($1650 new, box opened) can show an animation 28 times the maximal size allowed at this forum but even that -- 71 km by 40 km or 2840 km2 at 15 m Landsat resolution -- is not quite enough for full width of the PI trunk and close-in arms of tributaries.

 2013 312 231 114  LC82311142013312LGN00_B8.TIF
 2013 335 232 113  LC82321132013335LGN00_B8.TIF
 2013 344 231 113  LC82311132013344LGN00_B8.TIF

 2014 013 229 114  LC82291142014013LGN00_B8.TIF
 2014 258 232 113  LC82321132014258LGN00_B8.TIF
 2014 349 229 114  LC82291142014349LGN00_B8.TIF
 2015 064 229 114  LC82291142015064LGN00_B8.TIF

 2015 261 232 113  LC82321132015261LGN00_B8.TIF
 2015 293 232 113  LC82321132015293LGN00_B8.TIF
 2015 304 229 114  LC82291142015304LGN00_B8.TIF
 2015 320 229 114  LC82291142015320LGN00_B8.TIF

nukefix

  • Frazil ice
  • Posts: 461
    • View Profile
  • Liked: 9
  • Likes Given: 1
Re: PIG has calved
« Reply #455 on: November 24, 2015, 10:47:25 AM »
Does anyone have a ballpark estimate on the speed of the movements of the lineations? I need some estimate in order to pick a nice triplet...

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #456 on: November 24, 2015, 02:26:53 PM »
Nicely spaced triple would be great.

I've noticed that ESA and others keep putting out these 12-day repeat images that simply don't work because there has not been enough movement except at the snouts of the very fastest glaciers. One size doesn't fit all -- probably have to montage the outcomes from repeat intervals optimized to different velocities.

It is really quite hard to read off velocities on these illustrations because of  (1) sparsely labelled logarithmic color keys which (2) weren't embedded in the map and so didn't undergo the same color transformation causing the correlation between key and map colors to be lost plus (3) the velocity layer was collapsed onto an underlying satellite grayscale whereas the key was not, followed by separate jpg compressions of key and map.

Unlike in Cartography 101, if you click on a particular color in the key with a tool that can find that same color everywhere else on the map, it won't work regardless of how you set the color radius. The 1st image in #422 shows the velocity contours (isotachs) but their labels or spacing is not known (other than it is logarithmic).

#414-#422 have our best estimates of velocities (and also rates of vertical thinning). In the central region of the trunk, just up from the funnel throat, #418 for 2011 1-2 km per year (5 m per day) in the fastest region but fallowing off to 1 m per day on the side of an (arched) transect orthogonal to flow lines.

So at 15 m resolution @ 12 days, that would 1-4 pixels of movement, probably not enough relative to other issues.

The first image below shows the H grayscale of a HSV decomposition of the log velocity RGB provided by Mouginot (repeats 2nd image in #416)

The second image shows the log isotachs (32 grades) as contoured online at G'MIC with 0.3 degrees of smoothing. The third exponentiates the original H over at ImageJ and contours as above followed by histogram normalization.

This seems to work ok -- the scale ranges from 0 to 3 km per year so the isotachs are separated by 34 m per year -- but you can see the curvy mess this creates in the color key which is supposed to be clean vertical bars of width 8.

Our region of interest is slightly above the first tributary trunk coming in from the right, as illustrated by the dotted in the 4th image. This whole process of making the 4 images took me OVER 10 MINUTES (grr!?$#@!) so we cannot expect busy glaciologists to produce reader-usable products for journal articles, regardless of unlimited space available in article supplemental.
« Last Edit: November 24, 2015, 03:38:56 PM by A-Team »

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #457 on: November 24, 2015, 03:05:45 PM »
Here are the 5 cloud-free Landsats of path,row 229,114. They are not really aligned as I'm still awaiting word from Landsat (after holidays, or will they forget altogether?) as to what they did vis-a-vis polar stereographic. To get movement accurately especially with mixed path,rows  to remove projection error.

However the images, 15 m resolution of the upper right corner of the box in the previous post, do give some idea of the stability of the lineations and streaks (not so much). I'll add the other 6 dates in a bit (these require individualized contrast reprocessing) to give a 715 day span of monitoring and see if there is not a common fixed ground control point.

Note blog width of 700 pixels only covers 10.5 km at Landsat 15 m resolution whereas the feature here is ~70 km wide and perhaps 40 km in height so there has to be some selection on what sub-scenes are of interest, or else stand back and look at lesser resolution.

The time interval between cloud-free Landsats is very erratic and no images can be taken during the dark months which together would make for rather jerky animations. Thus Sentinel has some real advantages (though it does not do as well at imaging the streaks).

2013 312 231 114  LC82311142013312LGN00_B8.TIF
2013 335 232 113  LC82321132013335LGN00_B8.TIF
2013 344 231 113  LC82311132013344LGN00_B8.TIF
2014 013 229 114  LC82291142014013LGN00_B8.TIF
2014 258 232 113  LC82321132014258LGN00_B8.TIF
2014 349 229 114  LC82291142014349LGN00_B8.TIF
2015 064 229 114  LC82291142015064LGN00_B8.TIF

2015 261 232 113  LC82321132015261LGN00_B8.TIF
2015 293 232 113  LC82321132015293LGN00_B8.TIF

2015 304 229 114  LC82291142015304LGN00_B8.TIF
2015 320 229 114  LC82291142015320LGN00_B8.TIF
« Last Edit: November 24, 2015, 03:34:49 PM by A-Team »

nukefix

  • Frazil ice
  • Posts: 461
    • View Profile
  • Liked: 9
  • Likes Given: 1
Re: PIG has calved
« Reply #458 on: November 24, 2015, 04:57:29 PM »
HAIL THE POWER OF RADAR! Notice the fresh new cracks that are only visible in the B-channel looking deep blue while new cracks that no not appear in the first R-channel are cyan. Pretty stuff!

24 days between acquisitions shows the movement of the cracks but only from a close-up. Movement is very slow at the tips of the cracks and faster in the middle of the ice-stream.

This is UTM 10m pixel size (could not make Polar Stereo to work in SNAP). Sentinel-1 IW RGB of the following dates:

R= 11.9.2015
G= 5.10.2015
B= 29.10.2015
« Last Edit: November 24, 2015, 05:56:42 PM by nukefix »

solartim27

  • Grease ice
  • Posts: 548
    • View Profile
  • Liked: 17
  • Likes Given: 19
Re: PIG has calved
« Reply #459 on: November 24, 2015, 07:34:17 PM »
The pinned berg is showing a good bit of rotation on todays world view.

Meanwhile, in case you've missed the Thwaites thread, there's going to be some more calving there soon, and the big berg off of there is moving out.  Here are two of the same gifs from Oct 24 to Nov 11, one should animate, and one is saved full size (2.7 Mb) if you want to zoom in closer.
FNORD

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #460 on: November 24, 2015, 09:47:48 PM »
Quote
could not make Polar Stereo to work in SNAP
Awesome. What UTM zone, bro?

Be a good idea to write ESA help desk about their SNAP tool not working in PS Antarctica.

Note Landsat is posting as polar stereographic if the scene's center latitude is less than -63ºS, whereas the recommended use is UTM north of -80º and PS only from there to the pole, whereas the center of Pine Island trunk is something like -77.3º, -97.5º. So if only Landsat could fall in line, we have all the coastal cryosphere in a UTM.

Google Earth UtM is saying zone 14, C 536453 m, E 1641157 m (relative to what?). We should supply EPSG codes as that is really what most software is looking for in terms of specs.

We have come a long way since Sputnik-1? That was 4 Oct 1957, over 58 years ago. These projections and more were already in use by the pharaohs.

[Edit: 3rd image is upper corner of nukefix's 10 m 3 day color interferogram displayed at 1000 ms with last frame duplicated. 4th image just animates the BGR as grayscales at 320 ms. It is amazing how l.i.t.t.l.e changes in these images over these 48 days.]
« Last Edit: November 24, 2015, 11:09:24 PM by A-Team »

nukefix

  • Frazil ice
  • Posts: 461
    • View Profile
  • Liked: 9
  • Likes Given: 1
Re: PIG has calved
« Reply #461 on: November 25, 2015, 09:31:08 AM »
It's UTM zone 14 (auto determination).

Interesting that the new cracks can also be created in a fan-like shape. The big one appearing in the animation is about 500m wide after fanning out for 2.4km.

Here's a nice feature from the shear-zone...crevasses are formed, bent and finally they disappear, at least from sight..



« Last Edit: November 25, 2015, 10:09:43 AM by nukefix »

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #462 on: November 25, 2015, 02:37:19 PM »
Quote
It's UTM zone 14 (auto determination).
Even though in the overall Sentinel image (third image in #458) the ice shelf, calving front and lower trunk are in UTM zone 13?

There is an astonishing amount of information in these images about differential motion of various parts of the trunk (and the stress field implied by that motion). No question in my mind, Sentinel greatly outperforms published velocity products in terms of detail and accuracy.

Looking at the motion in the hair-dresser region (first image #458) by shifting the earlier date to the left until the difference (grayscale subtract) with the latest image was minimized, it appears that 6 pixels (60 m) is the best fit over the 48 day interval (1.25 m per day or 456 m per year) for the central and lower part of the image. This is slower than we were thinking but this region is on the periphery of the trunk where the velocity is falling off to zero.

The new 'crevasses' opened to a width of 430 m yet this had no impact whatsoever on the overall dimensionality despite the 660 m stretch in the direction of motion, suggesting that these new lineations were merely newly exposed (in the sense of becoming reflective to Sentinel radar) rather than newly created, melt season starting and all that.

Best cloud-free Landsat-8 matches to Sentinel-1A dates above are quite favorable for first and last:
                 
LC82321132015261LGN00 18 Sep 15  11 Sep 15
LC82321132015293LGN00 20 Oct 15  05 Oct 15
LC82291142015304LGN00 31 Oct 15  29 Oct 15
LC82291142015320LGN00 16 Nov 15


Note there's no need to restrict to 3 Sentinel dates because of only 3 color channels. From a chronologically order stack of Sentinels, any size, it's easy to make a rolling RGB image from each consecutive triplet. There is another option (less clutter) of just using pairs for RG and setting B to a favorable uniform gray. However for animation purposes it may be better to leave them as grayscales.
« Last Edit: November 25, 2015, 03:11:44 PM by A-Team »

nukefix

  • Frazil ice
  • Posts: 461
    • View Profile
  • Liked: 9
  • Likes Given: 1
Re: PIG has calved
« Reply #463 on: November 25, 2015, 03:38:18 PM »
Looking at the motion in the hair-dresser region (first image #458) by shifting the earlier date to the left until the difference (grayscale subtract) with the latest image was minimized, it appears that 6 pixels (60 m) is the best fit over the 48 day interval (1.25 m per day or 456 m per year) for the central and lower part of the image. This is slower than we were thinking but this region is on the periphery of the trunk where the velocity is falling off to zero.
There no discernible movement at all at the tips of the "hairs". The image is resampled at 10m so even a 5m movement should show up as tinted edges of the hairs. It's all just white in the RGB.

Quote
The new 'crevasses' opened to a width of 430 m yet this had no impact whatsoever on the overall dimensionality despite the 660 m stretch in the direction of motion, suggesting that these new lineations were merely newly exposed (in the sense of becoming reflective to Sentinel radar) rather than newly created, melt season starting and all that.
I have to disagree with that, there's no sign of features getting buried/exposed on the whole 170km * 250km coregistered frame that I have (dry snow does not impede radar-waves much). A simple explanation is that these are really "hairline" cracks and possibly much smaller than the resolution-cell. Since the radar is penetrating rather deep into the ice even a deep vertical 30cm-wide crack should be clearly visible on the SAR image due to the created ice-air interface. Also due to deep penetration the image of the crack can be distorted in the SAR image leading to misinterpretation.

The optical images should be useful in inferring what is the real surface-width of the crack.

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #464 on: November 25, 2015, 11:09:32 PM »
October 31 southern hemisphere corresponds to 52 days before austral midsummer solstice or April 30th in the northern hemisphere; the latitude of Pine Island calving front is not too different from Jakobshavn which would be in early melt; the iSTAR traverse of Antarctica www.istar.ac.u experienced milder weather than I ever saw as a paperboy during Chicago winters; the dielectric constant of wet snow or liquid water makes it impenetrable to Sentinel radar.

The first image below just is just catch-up on location maps of the 10 m forum scenes relative to the overall Pine Island trunk. The second shows the movement of the shear zone over the same dates as above which is that of a rigid body over the 48 day span of the animation. The third shows 10 Nov 15 posted as is by Wipneus which is very very similar in its details to the 31 Oct 15 Sentinel as posted in UTM14S by nukefix.

There are a great many curious features in this large complex glacier. The fourth image just shows one of this at the bottom; this area has not yet been posted at 10 m.
« Last Edit: December 03, 2015, 03:42:22 PM by A-Team »

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #465 on: November 25, 2015, 11:42:05 PM »
Here is Landsat bumped to 10 m resolution showing same flow entrainment region of #461 (29th Oct extracted) but in PS rather than Mercator projection. On a clear day (not common for the PI trunk), Landsat is quite impressive. I'll add the other Landsat counterparts in a bit. Both images need a click to see at full resolution.
« Last Edit: November 26, 2015, 01:04:56 PM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 3996
    • View Profile
    • Arctische Pinguin
  • Liked: 619
  • Likes Given: 0
Re: PIG has calved
« Reply #466 on: November 29, 2015, 04:21:25 PM »
For the PIG, the 26th of November was a fantastic day. Fine weather after weeks of thick clouds and not one but two Landsat 8 passes! The first one is during the local Austral (near) summer night and the second as usual at the local morning. That is 6:50 and 14:58 UTC.

 I made an animation of the two images (45m/pix).

The nightly image has a solar elevation of only 5.7o. Shadows are long so features that otherwise (the day image with a sun elevation of 29 degrees) are difficult to see.
The new crack can now be estimated to be about 130m wide, wider  than last years crack at this time. In front of the crack two more shallow dents are visible, also been seen on Sentinel images. They lack the sharp features of the crack so they may be different.
A minor calving seems to be happening just at the bottom of the images, not visible in the day image or in any recent Sentinel image.

(must click to start).

oren

  • Young ice
  • Posts: 4478
    • View Profile
  • Liked: 872
  • Likes Given: 1287
Re: PIG has calved
« Reply #467 on: November 29, 2015, 08:12:54 PM »
Fantastic images indeed.
It's amazing how much different the image looks when the sun angle is low. So many things jump out, it's almost 3-D.

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #468 on: November 29, 2015, 09:43:38 PM »
That would be 5.7º sun elevation on LC81551312015330LGN01. That image is a beauty. Below, some benefit to maybe working the contrast over with 'Integral Image Filters' in Image J --> normalize local contrast, block size 45, 1.75 std devs followed by 'Feature J" first derivative in y direction. Have to wait and see if this really has found a new crack in the making. The main crack appears a bit longer than it does in Wip's natural color pan-enhanced.

The other image of the same date LC80021132015330LGN02 only covers lower Pine Island. It is a bit of an oddity in terms of sun azimuth. Between the two of them, there is a lot of information about local topography in the shadows. However this would already be known better from lidar and stereo photogrammetry.

Sun Elevation    29.0º  vs 5.7º
Sun Azimuth      61.4º  vs 172.6º

The second image below  (from first scene, bumped to 10 m) shows a shear zone about mid-trunk where the faster moving ice stream is warping slower adjacent flow, to the point of lineation discontinuity. We've looked at this region before with both radar and panchromatic in #465 but this is the clearest view to date.

The third image shows a very pervasive and peculiar feature that seems to represent wind-driven snow. These originate in all cases from a small hole or defect on one of the lineations and extend for ~200 m to the north (bottom of image) as an elongated snowdrift. (Dry snow deposits and sun shadows don't show on radar.) PI has tens of thousands of these. By all accounts, it is very windy there; katabatic winds would blow consistently down the trunk to the coast.

Note these snow streaks are all aligned south to north, irrespective of flowline which is oblique to this axis in the example scene shown. Nonetheless, the streaks always originate as a dark spot associated with a lineation.

This Landsat in fact is not so favorable for imaging the upper trunk -- it appears that either new snow or wind-blown redistributed snow has obscured detail on the mid and upper trunk. However in the fourth image, you can see there is a whole lot going on with this glacier that needs interpretation: zones of differential speeds, over-writing of past stress fields, marginal interactions, tributary contributions getting sorted out, and so forth.

Some of this history may reveal tele-effects of recent changes down at the disintegrating ice shelf that are propagating up-trunk; others are just the consequence of buried landforms, gravitational gradients, and tributary speeds, volume and geometrical relations.
« Last Edit: November 30, 2015, 03:30:56 PM by A-Team »

tombond

  • New ice
  • Posts: 93
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: PIG has calved
« Reply #469 on: November 29, 2015, 10:53:39 PM »
Wipneus and A-Team, thank you so much for the fantastic images they are much appreciated!

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #470 on: November 30, 2015, 03:24:20 PM »
Pine Island ... hmm, no island of that name in Amundsen Bay, no trees in sight -- the only vascular plants in Antarctica are Daeschampsia antarctica (hairgrass), from a genus of facultative wetland grasses, and Colobanthus quitensis (Antarctic pearlwort) in the Caryophyllaceae.

No one has the slightest idea why these and only these two species have colonized Antarctica. They do not have any specific adaptations to the Antarctic climate at either the morphological or genomic level. There's a nice review of the whole situation at http://www.scirp.org/journal/PaperDownload.aspx?paperID=7595 (

Poa annua, an annual grass, has made some recent inroads due to increasing tourism. Some tour companies have made laudable efforts to require boot washing in chlorine bleach; however shoes are not a common route of seed transportation compared to cuffs and pockets; dilute bleach would have no effect on seed viability (it is commonly used as a seed rinse prior to storage).

It turns out Pine Island Glacier is named after a US Navy sea plane tender ship, the Pine Island, that supported Antarctic mapping during Operation Highjump in 1946. The ship was decommissioned for the last time in 1972, sold to Zidell scrapyards in Portland and the metal recyled into a barge.

The ship was NOT named after Pine Island (off Ft Myers, west coast of Florida) but rather Pine Island Sound in which Pine Island resides. (The only other 3 ships in this class were named for Currituck Sound, Norton Sound and Salisbury Sound.) The correct name however is USS Pine Island, not USS Pine Island Sound.

According to wikipedia, a sound (Sund in Scandavian) is a large sea or ocean inlet larger than a bay, deeper than a bight (a bay that can be left in a single sailing tack), and wider and different in origin from a fjord; or a narrow sea or ocean channel between two bodies of land. There is little consistency in the use of 'sound' in English-language place names. Below is the namesake sound for which Pine Island Glacier is named. It is very shallow throughout, 1-2 m in depth at mean high tide. I last visited the sound in 2011.
« Last Edit: November 30, 2015, 03:44:48 PM by A-Team »

sidd

  • Young ice
  • Posts: 4946
    • View Profile
  • Liked: 349
  • Likes Given: 0
Re: PIG has calved
« Reply #471 on: December 01, 2015, 06:24:08 AM »
This reply is tardy, mea culpa. In post 408, the question was raised as to the concavity/convexity of surface profile. This tickled my neuron, but i had not the time to track down the reference until now.

doi:10.1002/jqs.2683

Fogwill etal. use PISM, which i am partial to, and UVic for ocean, and address the contribution of EAIS to Eemian sea level high stand, and putative weaknesses of EAIS basins today. But the particular passage which strummed my neuron was:

"We propose that the sensitivity of these sectors of the ice sheet relates to two major factors: firstly the coincidence of EAIS basins with concave ice-sheet surface profiles; secondly, the basins’ connectivity to the ocean. The basins’ bed topography and ice-sheet geometry control the mass flux from the ice sheet; those which have a weak or sliding bed have a faster flow regime, resulting in a concave surface profile (Cuffey and Paterson, 2010). In contrast, basins where ice is flowing slowly with little or no basal sliding exhibit a parabolic, or convex, surface profile and much lower ice fluxes. Our results imply that basins with concave surface profiles are particularly susceptible to ocean forcing, bringing about greater rates of surface lowering than in other areas ..."

Perhaps we should talk some more about Fogwill(2014) in an EAIS thread.

sidd

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #472 on: December 01, 2015, 01:29:11 PM »
There's a lot of data out there on surface elevation of Pine Island and how it has changed over the years. The first two images illustrate the situation  a few years back at a mid-trunk transect. The bottom image shows the surface and bedrock along a 2014 longitudinal profile through the lowest point of the transect. There are convex, concave and indeterminate regions. PISM calls for some ice piracy from Filchner–Ronne and Ross.

Quote
Spatio-temporal analysis of surface elevation changes in Pine Island Glacier, Antarctica, from ICESat GLAS data and ERS-1 radar altimeter data
U Herzfeld B Farley
Annals of Glaciology 55(66) Nov 2013
Characterized by fast movement, low surface slope and grounding below sea level, Pine Island Glacier (PIG) plays an important role in the stability of the West Antarctic ice sheet. In previous work, we reported that the spatial distribution of 1995–2003 surface lowering in PIG suggests an attribution of changes to an internally forced process in the glacier. Other work associates changes in PIG entirely with processes in its ice shelf. Here time series of maps of surface elevation change in PIG and its ice shelf are derived from geostatistical analysis of ICESat GLAS and ERS-1 radar altimeter data. Based on spatio-temporal analysis of 1995–2007 elevation change, we discuss indications of processes that initiate from changes in the ice shelf versus processes that start internally in the glacier. Thinning rates continued to increase after 2003, regionally to >15 m a–1. The initiation of acceleration occurred in the interior of the ice stream, while in later years largest elevation loss was driven by changes in the ice shelf and upward propagation. By 2006, the region of thinning had expanded up-glacier beyond the initial areas of surface lowering to 100 km above the hinge line. More than one process causes dynamically complex changes in PIG

Quote
Collapse of the West Antarctic Ice Sheet after local destabilization of the Amundsen Basin
J Feldmann, A Levermann
http://www.pnas.org/content/112/46/14191 ($10 paywall Nov 2015)

Recently satellite observations and high-resolution simulations have suggested the initiation of an ice-sheet instability in the Amundsen Sea sector of West Antarctica, caused by the last decades’ enhanced basal ice-shelf melting. Localized destabilization will yield a full discharge of marine ice from West Antarctica, associated with a global sea-level rise of more than 3 m unless the ice loss is limited by ice dynamics and topographic features. Here we show that in the Parallel Ice Sheet Model PISM at 5-km horizontal resolution, a local destabilization causes a complete disintegration of the marine ice in West Antarctica. Thereafter, the marine ice-sheet instability fully unfolds and is not halted by topographic features. In fact, the ice loss in Amundsen Sea sector shifts the catchment's ice divide toward the Filchner–Ronne and Ross ice shelves, which initiates grounding-line retreat there. Antarctica will irrevocably contribute at least 3 m to global sea-level rise…

Quote
IceBridge Surveys More of West Antarctica 2014 and 2015: [no link to actual data]
The next morning, Nov. 6, mission planners returned to the Punta Arenas weather office to find that things still looked good in the Pine Island Glacier region. With a good forecast in hand, the team took off for a survey to collect data on tributaries feeding into the main trunk of Pine Island Glacier. This repeat of a survey last flown in 2010 was designed to measure changes to the ice surface beyond IceBridge’s other Pine Island Glacier missions.

On Nov. 7, favorable weather conditions were still holding in West Antarctica, so the IceBridge team headed out for a survey of the Thwaites, Smith and Kohler glaciers. This flight repeated parts of a survey flown in 2012 and primarily measured ice elevation in this rapidly-changing part of West Antarctica. In addition, researchers were able to collect high altitude sea ice elevation data with the onboard laser altimeters on the way to and from the survey area.
« Last Edit: December 01, 2015, 01:41:45 PM by A-Team »

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #473 on: December 01, 2015, 01:52:39 PM »
Here is what we are aiming for with Pine Island -- a modern representation of surface and bedrock. Unfortunately this is Store Glacier in west-central Greenland, from Project Safire http://www.spri.cam.ac.uk/research/projects/safire/  Animations like this can also be done as .mov with a slider but are best viewed as a 3D graphic with handles (as in ImageJ) that can be redrawn to any perspective, with the top surface animated to show thinning, velocity and change in velocity. This would take ten minutes to do for Pine Island were it not for the obtusity with which the data is stored.
« Last Edit: December 01, 2015, 02:08:11 PM by A-Team »

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #474 on: December 01, 2015, 09:35:22 PM »
I located a nice GeoTiff of Antarctic surface elevation at 1 km horizontal resolution from V Helm et al www.the-cryosphere.net/8/1539/2014/ that they stored at the Pangaea repository http://doi.pangaea.de/10.1594/PANGAEA.831392

This is the right way to do it, all too rare. GeoTiff has plenty of precision and its header carries along all the metadata anyone would need. You can see at a glance what the data looks like, stack as a multi-tif, process as needed and crop out the region of interest. ImageJ has a clever interactive 3D surface plot tool from which this rough animation of Pine Island surface elevation could be made.

Google topic, Find article, locate its data link --> do something, upload the product to forum: 10 minutes, not 10 hours of thrashing around in netCDF or 10 days waiting for an author to maybe respond to an email request. Now if only I could find something similar for ice thickness or bedrock and velocities ...

« Last Edit: December 01, 2015, 09:46:22 PM by A-Team »

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #475 on: December 01, 2015, 09:48:15 PM »
ImageJ has a clever interactive 3D surface plot tool from which this rough animation of Pine Island surface elevation could be made.

Very nice 3D image.  I will be interesting to see if/when cliff failures occur if/when the PIIS calves away, leaving a bare & relatively steep cliff face.
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #476 on: December 02, 2015, 03:11:13 AM »
Good point. It gets steep there in a hurry, slope attached (derived from V Helm DEM of 2013 via D Tschumperlé's G'MIC). I wonder what the history of the steep section is, presumably ice from the five tributaries held back at the funnel making it thicker there but thin at the shelf.

A lot of effort is expended on obtaining accurate elevation maps, not to get height about sea level, but to get at slopes, in particular to get the gradient or direction of maximum slope. Glaciers flow because of a gravitational force gradient; an ice cube sitting on a table melts but does not flow.

On a digitized DEM (a scalar field or 0-tensor representable by a grayscale), that will involve subtracting pixels in all directions from a given pixel so the error bars need to be small for this to work. The resulting derivative is a vector field (or 1-tensor) requiring two grayscales for its representation, one for magnitude and the other for direction. The Glen flow law governing glacial rheology uses invariants of 2-tensors derived from ultimately from the gradient vector field.

The three images below illustrate the issues. The first is the experimental surface elevation data cropped to PI out of a much larger map (6669 × 6669 pixels) of all Antarctica compiled by V Helm et al as finalized on 01 Sep 2013 and archived as a 170 MB 32-bit geotif at Pangeae (along with the associated error map).

The second image has calculated the gradient, determined its magnitude at each point, re-scaled those values to a 0-255 range of grays that can be displayed on a computer monitor, drawn contours connecting slopes with the same value, and displayed direction via brightness along these contours, using one of the 305 online image processing tools at https://gmicol.greyc.fr/.

The third image uses the color wheel to tint the contours of the second image to indicate direction. This direction corresponds to flow of rainwater over the surface, often but not necessarily the direction of flow of glacier ice. That flow will be orthogonal to the contours, hence their utility.

For purposes of glaciological computations within a GIS stack, two grayscales suffice: one for the magnitude of the slope and another for compass direction (no derived contouring). These grayscales would retain the bit depth appropriate to the precision of the original experimental data, as degraded by error and processing.

For example if the elevations ranged from sea level to 512 m and were known to the nearest meter, 8-bit images would be inappropriate as their 256 levels could only distinguish 2 m intervals. Here 9-bit would work satisfactorily though these would take up 16-bit given computer memory architecture.

On the other hand, 32-bit tifs are totally excessive as they can distinguish 232 = 4 294 967 296, over 4 billion values which in our example works out to 0.19 microns in a situation where random experimental error might be a meter or more. While there's no gain here, there's no harm done either with 32-bit. Other data or derived products might require it and bit depth can always be dropped at the end when all calculations have finished..
« Last Edit: December 02, 2015, 01:16:27 PM by A-Team »

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #477 on: December 02, 2015, 01:53:48 PM »
The original article shows not only surface elevations for the Antarctic continent but also slopes in degrees (Fig.6 of www.the-cryosphere.net/8/1539/2014/ ) at an author-supplied resolution of 1917 x 2179 pxls. (Images in pdfs should not be copied by screenshot but rather imported by software like ImageJ or Gimp that can extract the full native resolution supplied to the publisher.)

The legend on that figure is very instructive: Antarctica is so very flat that the greatest slope encountered (at this resolution) is 1º. For comparison, the steepest road in the world (Baldwin Street in Dunedin, NZ) is 19º. It is common to have truck escape ramps on long 5º-7º slopes However a 1º slope is quite significant for a railroad bed and early trains were restricted to half that (both for power and braking).

Simply showing the magnitude of the slope in arbitrary colors is not useful. The animation below shows futile attempts to extract the underlying slope grayscale from the original coloration by decomposition into RGB, HSV, or Lab channels. None of these work, with the possible accidentl exception of the b channel in Lab (2nd frame). Alternatively, it is easy to make or find attractive palettes from which the result data is easily recovered.

You won't see the Rignot group using a nonsense palette. For example, the Pine Island velocity map above uses a simple tint in HSV in the hue channel to carry the data.
« Last Edit: December 02, 2015, 09:13:16 PM by A-Team »

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #478 on: December 02, 2015, 09:18:57 PM »
The attached Terra image of the PIIS for Dec 2 2015, shows that a minor calving event occurred on the Southwest face of the PIIS.

Edit: Possibly associated with the build-up & release-of compressive stresses in this area.

Edit2: When I say the Southwest face of the PIIS; I mean the area immediately to the Northeast of the ice shelf for the Southwest Tributary Glacier.
« Last Edit: December 02, 2015, 09:48:00 PM by AbruptSLR »
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

sidd

  • Young ice
  • Posts: 4946
    • View Profile
  • Liked: 349
  • Likes Given: 0
Re: PIG has calved
« Reply #479 on: December 03, 2015, 02:31:39 AM »
for someone who whines about basal hydrology all the time, i haven't brought up Livingstone(2013) in a while ... that might be relevant in the Greenland threads as well ...

doi:10.5194/tc-7-1721-2013
open access, so nice

fig 1 for antarctica

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #480 on: December 03, 2015, 10:50:38 AM »
Pulling out the Pine Island portion of the predictions of SJ Livingstone 2013 for basal hydrology, the first frame shows a classical drainage system that corresponds fairly well with the center lines of surface tributaries. However the large lake at the upper confluence has no surface manifestation in either Landsat or Sentinel.

The lake is fairly stable to monte carlo re-sampling of the bedrock DEM with respect to its error map, second frame. The bed of Pine Island is cold according to 2010 modeling by Pattyn (still paywalled), rather than polythermal, third frame, meaning the bed is not at the pressure melting point, the streams and lake don't actually form, and rapid flow here is not primarily deformational.

Overall, the predicted basal hydrology of PI is not remarkable relative to Thwaites, Vostok, and other features of the continent. However there is little here that is experimentally testable any time soon. It may be better to stick with surface observables as change can be readily monitored over the 12 day Sentinel return window by folks at home.

There is quite a bit of new information about Pine Island emerging at conferences such as the Nov 13 ISTAR, one component of which traversed some of the tributaries with seismic, radar, and 50 m deep firn cores. It's not clear how many of these will put sufficient detail online to accompany the abstracts.

The Pine Island twitter site is the best place to look for new releases though many of the tweets are cryptic and their links dead ends. https://twitter.com/AntarcticPIG

Quote
Late glacial sub glacial lake sediments recovered and sampled in Pine Island Bay
XII International Symposium on Antarctic Earth Sciences, Goa, India, 13 July 2015 - 17 July 2015 .
G Kuhn et al

Subglacial meltwater facilitates rapid ice flow beneath concurrent ice sheets, and there is widespread evidence for a dynamic subglacial water system beneath the Antarctic Ice Sheet. It steers and affects the pattern of ice flow and is a direct result from boundary processes acting at the base of the ice sheet, i.e. pressure induced basal melting. Consequently, the occurrence of subglacial meltwater plays an important role in bedrock erosion, subsequent resedimentation, and in shaping the topography of icesheet beds.

Here we present new geological and geochemical data from sediments recovered on the West Antarctic continental shelf in Pine Island Bay that we interpret as reliable indicators for deposition in a palaeosubglacial lake beneath the formerly expanded West Antarctic Ice Sheet, presumably during or following the Last Glacial Maximum. Characteristic changes of sedimentary facies and geochemical profiles within these cores taken on RV Polarstern expeditions support the presence of an active subglacial lake system during the late stages of the last glacial period.

These findings have important implications for palaeo icesheet dynamics, suggesting there was considerable water available to lubricate the bedrockice interface and deposit water saturated subglacial sediments (soft tills).... Our findings may also have implications for ice sheet models, which have to consider the predominantly nonlinear effects related to subglacial hydrology.
« Last Edit: December 03, 2015, 04:16:34 PM by A-Team »

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #481 on: December 03, 2015, 06:25:02 PM »
for someone who whines about basal hydrology all the time, i haven't brought up Livingstone(2013) in a while ... that might be relevant in the Greenland threads as well ...

doi:10.5194/tc-7-1721-2013
open access, so nice

fig 1 for antarctica

For those new to this Antarctic folder, more information on subglacial meltwater drainage systems can be found at the following link:

http://forum.arctic-sea-ice.net/index.php/topic,404.0.html
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #482 on: December 03, 2015, 08:32:07 PM »
The image below shows the 20 and 200 m/year velocity contours for Pine Island but there is a lot more for us in this Antaractic-wide article. We may be able to improve it locally and at Thwaites using incoming Sentinel 1A's if we can replicate Ng's procedures.

Reprojecting Sentinel 1A to 10 m UTM, two pixels of movement in scenes a year apart amounts to the 20 m/yr contour. That's within range, depending on the presence of scene markers co-moving in the ice without distortion. With this contour overlaid on the Sentinels above, we could check now for availability of markers.

The lineations don't extend this far up the tributaries but other wave-like patterns do and may be more favorable. In comparing sensitivities, recall interferometry looks at the horizontal component of line-of-sight displacement of ice to satellite over two dates whereas nadir views look at a projection of (sloped) ground displacement, and GPS (or flag) movement refers to net surface displacements in absolute coordinates, whereas tape measurements of flag position before and after refer to a lumpy, curved and sloped earth surface, with meters per year on the WGS84 ellipsoid being the site of comparison.

Quote
Spatial complexity of ice flow across the Antarctic Ice Sheet
Felix S. L. Ng   
Nature Geoscience 8  847–850 2015
doi:10.1038/ngeo2532 Published online 16 Sep 2015
 
Fast-flowing ice streams carry ice from the interior of the Antarctic Ice Sheet towards the coast. Understanding how ice-stream tributaries operate and how networks of them evolve is essential for developing reliable models of the ice sheet’s response to climate change.

A particular challenge is to unravel the spatial complexity of flow within and across tributary networks. Here I define a measure of planimetric flow convergence, which can be calculated from satellite measurements of the ice sheet’s surface velocity, to explore this complexity. The convergence map of Antarctica clarifies how tributaries draw ice from its interior. The map also reveals curvilinear zones of convergence along lateral shear margins of streaming, and abundant ripples associated with nonlinear ice rheology and changes in bed topography and friction.

Convergence on ice-stream tributaries and their feeding zones is uneven and interspersed with divergence. For individual drainage basins, as well as the ice sheet as a whole, fast flow cannot converge or diverge as much as slow flow. I therefore deduce that flow in the ice-stream networks is subject to mechanical regulation that limits flow-orthonormal strain rates. These findings provide targets for ice-sheet simulations and motivate more research into the origin and dynamics of tributarization.
« Last Edit: December 04, 2015, 04:34:56 PM by A-Team »

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #483 on: December 04, 2015, 06:05:18 PM »
I'm looking now inside the FSL Ng paper (unusual for having no co-authors), several posts on this are in the works. It covers all of Antarctica but with special blow-ups of Pine Island/Thwaites, Bindschadler/MacAyeal  Whillan/Mellor, Lambert, Larsen C (unlabelled Fig1e) and a few others but not including Toten.

The caption of Fig.1 got my attention as it states "Color scale applies to all panels [and two full-resolution Pangaea tifs] and is optimal for rendering features within streaming texture" (emphasis added).

The maps are quite unattractive visually but that could be forgiven if it was indeed highly effective. But what could optimal even mean -- how is such a thing scored, what is the next closest key scheme, how could it simultaneously be optimal at vastly different display scales and for both convergence and strain rate textures, by what methodology was this discovered, who should be credited, and above all, where else in science should this palette be deployed?

The data being displayed has an effective range of  [-1,1] up to a scalar and low enough precision that 8-bit suffices (rare in glaciology to match bin number to data precision). Although that range could be displayed as an ordinary grayscale, zero has a special meaning that would not stand out as 128 gray. The data ranges here are symmetric but otherwise similar to the needs of elevation maps ranging about sea level as zero.

Ng followed best-practices (again rare in glaciology) embedding the color scale within the figure rather than providing it as a loose supplemental figure. This means the colors used in the maps are in fact identical to those used in the key (these won't match after separate lossy jpeg compression). It also means a color picker can pull out and count exactly those pixels of a given value (or range of values), eg all the extreme strain rate locations in Antarctica.

Opening the full-resolution tifs in ImageJ showed the map to be indexed color. The LUT (lookup table that maps [0,255] --> RGB color space assignment) can be conveniently viewed, edited, swapped out for any other color table, or saved to re-color other people's sub-optimal maps. The LUT shows exactly which map colors were chosen.

Since white and its pale shades are intuitively appropriate to display less significant regions here and the parameters are inherently linear with no way-stations of significant thresholds, it makes sense to run gradients out from white to two fixed end point colors in 128 steps each and butt these back to back, which is what Ng did, using contrasting colors for the two end points (opposite colors across the color wheel). These are called diverging palettes.

So far so good, but what is optimal about this choice given the muddy non-terminal values, with shades of blue being the very worst for human eye color discrimination? In advanced perceptual palettes, gradient steps are not made in RGB or HSV but equidistant in CIELAB, which would be optimal in that experimental sense though that might not carry forward to monitor display devices very well. Here there is no strict regularity in any channel after RGB, HSL, HSV or LAB decomposition.

It's imperative to stay within sRGB, the colors that can be displayed by a monitor. Reverse image search does not find a match with a previously known palette. Matlab offers many palettes, some of them like jet and parula are fairly widely used.

It's probable some software somewhere claimed the palette here was optimal under some technical criterion for which there is no consensus. The empirical approach has proven far more effective: simply scroll rapidly through a large family of palettes to find the effective ones for a given data set, not seek to deduce an optimal palette from first principles.

http://photorealizer.blogspot.com/2012/04/colorimetry.html
https://mycarta.wordpress.com/2012/02/09/visualization-tips-for-geoscientists-matlab/
http://www.codeproject.com/Articles/243610/The-Known-Colors-Palette-Tool-Revised
http://rsb.info.nih.gov/ij/plugins/lut-editor.html

Here we also need to look at the streaming textural results themselves and what their inherent display issues are relative to human pattern recognition abilities. For example, the figure caption states "dashed ellipses mark convergence ripples associated with bumps/steps in BEDMAP2 topography; these often pair together to show divergence followed by convergence along flow."

A good color palette might suggest this hypothesis but is no substitute for a automatic classification regime. That would require a correlation scheme between the topography bump layer and the flow convergence layer.

It is critical not to alter palettes from the original in the process of moving them from the official archived tif and analyzing them. Here the palette, a peculiar 1600 by a reasonable 200 pixels, can be exchanged between ImageJ and Gimp, each of which has its strengths, using compressed but lossless png format. Since the palette is really only 1 row of pixels extended identically vertically, its compressed size is a mere 2 kb.

The width of an indexed color palette should always be a multiple of 256 (so that each bin has equal width). Here it is not: it should have been 6*256=1536 pixels in width. Something is therefore wrong with palette construction and placement, probably during resizing which introduces interpolation artefacts. (To extract an object like a palette, crop to a generous border, add an alpha channel, select and delete the border, use 'autocrop image' to cleanly remove blank space, obtaining a palette with no border.)

The great thing about indexed color and archiving full-resolution tifs is that the scientific reader can easily change the colors to something else without any degradation of data quality. Indexed color is also best for animations since the gif89a standard allows every frame to have its own LUT.
« Last Edit: December 05, 2015, 03:16:35 PM by A-Team »

sidd

  • Young ice
  • Posts: 4946
    • View Profile
  • Liked: 349
  • Likes Given: 0
Re: PIG has calved
« Reply #484 on: December 04, 2015, 09:21:36 PM »
"...2010 modeling by Pattyn (still paywalled) ..."

available at

homepages.ulb.ac.be/~fpattyn/papers/Pattyn2010_EPSL.pdf

nice paper.

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #485 on: December 05, 2015, 11:08:07 AM »
sidd, thx for the full text link to Pattyn 2010. Fascinating that the first maps of basal conditions of the Antarctic ice sheet were published 52 years ago. The experimental situation on Antarctic geothermal heat has not improved much since and remains much worse than Greenland, which is abysmal but at least has potential (dense radar tracks which can yield temperature profiles: http://onlinelibrary.wiley.com/doi/10.1002/2014JF003418/full).

However any kind of map suffices for playground development (sharpening of techniques to be applied meaningfully at some later date). The map below suggests that Pine Island and Thwaites have quite a warm bed, though correspondence with tributary structure is disturbingly poor.

Quote
Basal conditions (temperature and hydrology) of the Antarctic ice sheet are poorly understood. Nevertheless, basal conditions govern to a large extent the dynamical behavior of ice masses. Underneath present-day ice streams and outlet glaciers of Antarctica and Greenland, water and deformable wet sediments lubricate the base, hence facilitating fast ice flow. The interaction between the ice and its underlying bed, which controls how basal velocity will change as ice sheet stresses evolve, remains a key uncertainty.

Knowledge of basal ice sheet conditions is also essential in the quest for oldest ice, which occurs where basal ice layers are frozen to the bed. Obvious places to look for are the deepest parts of the ice sheet, where ice is thick, and accumulation rates are low. However, a thick ice cover insulates very well and keeps the geothermal heat from escaping to the surface.

Direct measurements of basal temperature are limited to a small number of boreholes. Several sites neared the pressure melting point at depth (eg Vostok, EPICA Dome C, EPICA DML, Dome Fuji). Thence, we have to rely on other techniques, such as ice sheet modeling, to catch a glimpse of the sub-ice environment.

One of the earlier estimates at basal temperature conditions, Zotikov 1963, concluded the central parts of the Antarctic ice sheet are at pressure melting point. The lack of detailed measurements on surface topography and bedrock prevented a more rigorous approach and the basal temperature field remained rather schematic. However, the quality of the basal temperature distribution remains highly dependent on the distribution of geothermal heat flux. [edited]
« Last Edit: December 05, 2015, 11:13:11 AM by A-Team »

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #486 on: December 05, 2015, 05:16:39 PM »
However any kind of map suffices for playground development (sharpening of techniques to be applied meaningfully at some later date). The map below suggests that Pine Island and Thwaites have quite a warm bed, though correspondence with tributary structure is disturbingly poor.

A-Team,

In Reply #26 of the Tectonic thread (see link below), I point-out that while the geothermal pattern of the figure that you provide has value, the units of the figure are so out of date as to be considered wrong by current standards.

http://forum.arctic-sea-ice.net/index.php/topic,393.0.html

My wife needs me to run errands so I do not have time to point-out in which threads there is more current information about the BSB geothermal readings.

Best,
ASLR

Edit: I am temporarily back from my errand(s) & provide the following, somewhat more updated (previously posted in the " Subglacial Lake and Meltwater Drainage Systems" thread) information focused on the Byrd Subglacial Basin, BSB:

The first attached image comes from:
Dustin M. Schroeder, Donald D. Blankenship, Duncan A. Young, and Enrica Quartini, (2014), "Evidence for elevated and spatially variable geothermal flux beneath the West Antarctic Ice Sheet", PNAS, doi: 10.1073/pnas.1405184111

http://www.pnas.org/content/early/2014/06/04/1405184111.abstract

http://www.pnas.org/content/suppl/2014/06/04/1405184111.DCSupplemental

The second attached image comes from:
http://www.scientificamerican.com/article/high-heat-measured-under-antarctica-could-support-substantial-life/

Additionally, Replies #72-73 in the "Surge of WAIS Ice Mass Loss" thread contain discussions from a couple of years ago about the subglacial drainage system in the BSB.
« Last Edit: December 05, 2015, 06:17:14 PM by AbruptSLR »
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #487 on: December 05, 2015, 06:19:18 PM »
Quote
wife needs me to run errands so I do not have time
Ditto. End of year stuff closing down on me as well. We were only interested in that map because of its cited use as a basis of the Ng article. We can maybe use a better map to see if the results there hold up. Same for the updated Bedmap2 that  came out the other day (according to twitter, I don't have link yet to the file).
« Last Edit: December 05, 2015, 08:40:37 PM by A-Team »

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #488 on: December 06, 2015, 02:26:45 PM »
Quote
wife needs me to run errands so I do not have time
Ditto. End of year stuff closing down on me as well. We were only interested in that map because of its cited use as a basis of the Ng article. We can maybe use a better map to see if the results there hold up. Same for the updated Bedmap2 that  came out the other day (according to twitter, I don't have link yet to the file).

I am not sure whether the following link leads to the Bedmap2 data that you are looking for or not:

https://www.bas.ac.uk/project/bedmap-2/#data
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #489 on: December 08, 2015, 03:28:35 PM »
Quote
updated Bedmap2 that  came out the other day (according to twitter)
That's the correct link. I had merely come across some idiot grad student breathlessly twittering on 21 Nov 2015, urging us to drop what we are doing to check out the latest news: the 06 Mar 2013 data release of Bedmap2.
Quote
Updates to Bedmap2 - 23rd April 2013
ASCII grid dimensions: During the export process, 67-cell 'No Data' buffers were inadvertently added around the ASCII surface and bed grids, so these grid dimensions were larger than necessary (6801x6801 pixels). All files were, however, properly geolocated and the file headers correctly described the file dimensions. We have recreated these grids without the buffers so their dimensions are now 6667x6667 pixels.

Ice thickness data format: 4he geodatabase and tiff versions of the ice thickness grid were inadvertently in 32-bit integer format rather than 16 bit. We have corrected this.

Bed elevation inconsistencies: 36 pixels failed a test of consistency for bed+thickness=surface over grounded ice, or bed+thickness<surface for floating ice. We have corrected this by adjusting the bed elevation accordingly in all of the formats.

Update to Bedmap2 – 31st August 2013: Conversion of Bedmap2 height reference to WGS84 ellipsoid
The bedmap2_readme .txt file incorrectly stated that the gl04c_geoid_to_wgs84 grid should be subtracted from the Bedmap2 grids to convert to WGS84. It should be added. The readme files are being changed.
« Last Edit: December 08, 2015, 03:47:58 PM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 3996
    • View Profile
    • Arctische Pinguin
  • Liked: 619
  • Likes Given: 0
Re: PIG has calved
« Reply #490 on: December 09, 2015, 12:49:03 PM »
Access to the Sentinel 1A images has been difficult, if not impossible, for the last 10 days (or since Sentinel 2A image stream has started). It is the reason for the belated availability of the 4 December IW image. It shows the grounded calving is now resisting the push from the advancing glacier and rotates rather than moves. Small calving(s) can be seen in this animation (@80m/pix) with the 22 November image. Note that the "medium" sized iceberg visible in the 22 Nov frame has its origins in the big calving this summer and has been blown back.


A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #491 on: December 09, 2015, 03:00:13 PM »
Quote
Access to the Sentinel 1A images has been difficult, if not impossible
I eventually got the new MS Bing-based interface to open. It is highly defective at the Ellesmere, northern Greenland, Spitzbergen latitudes. I drew a select rectangle but could not find a way to load its corners in the search box.

Astonishing really that an organization this size does not have the slightest idea how to set up a web server or provide a stable search interface for a simple database.

The visitation load being reported is not heavy as these things go. They don't seem to be mirroring which has been widely used since the mid-90's to distribute load. It appears that they are have not separated the visitor interface server from the download server.

Meanwhile, not utilizing cloud storage but rather a 'rolling archive' means a whole lot of imagery has been deleted without anyone having been able to access it. Lots more satellites to come in this Copernicus series ... they are not ready.

What I see here is computer-incompetent upper management that they are not able to fire or replace.

I've found a few cryosphere Sentinel-2A images at ESA. These are very nice but are often unlabelled as to location, image north, resolution, scale, processing level and so forth. It seems like a neat satellite with well-chosen bands but so far it is all a waste.

Quote
Since 03-Dec-2015 the main access point for the Sentinels Scientific Data Hub has been experiencing considerable performance issues. The problems are under continued investigation. To alleviate the problem a temporary access for the Sentinel-2 data has been set up for all users here: https://scihub.copernicus.eu/s2 . Please login to this service using the guest/guest account (no additional registration is needed). New information regarding the return to full service will be provided as soon as possible. 07 Dec 2015 - 17:27

AbruptSLR

  • Multi-year ice
  • Posts: 17520
    • View Profile
  • Liked: 726
  • Likes Given: 236
Re: PIG has calved
« Reply #492 on: December 11, 2015, 05:42:44 AM »
I will be on vacation after tomorrow morning until after Dec 20th.

ASLR
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

A-Team

  • Nilas ice
  • Posts: 2401
    • View Profile
  • Liked: 282
  • Likes Given: 19
Re: PIG has calved
« Reply #493 on: December 11, 2015, 01:23:25 PM »
Quote
on vacation after tomorrow morning until after Dec 20th
Thx for all those excellent posts!

Some progress to report on Sentinel-2A. First, a rare image of never-studied Jungersen Glacier in extreme northern Greenland (82.1º N, -43.1º W) which flows into Nordenskiöld Fjord.

It has some similarities in its arcuate features in 3-4 places to what we've been discussing for upper Pine Island so I'm posting it here. From its features, Jungersen might appear to be faster moving than Pine Island's trunk but is actually not.

Sentinel-2A is similar to Landsat-8 but has improved ground resolution (10 m vs 15 m) for three bands in the visible and band 8 in the near infrared. It has a 10 day return which will drop to 5 when its twin sister 2B is launched. It is primarily for vegetation monitoring and no imagery will be taken beyond 83º for Antarctica.

This works out ok for Pine Island at -75.3º and other coastal glaciers. However no Sentinel-2A imagery has been released yet for any glacier in Antarctica.

ESA has done a rank amateurish job of annotating Sentinel-2A gallery images, not providing ground resolution or band combinations used. Via comparison with band 8 of Landsat-8, the Jungersen image has been inexplicably dumbed down to 15 m and so does not illustrate the full capabilities of their satellite (the entire purpose of the gallery). The image is likely comprised of bands 4,3,2 as RGB but may involve additional undescribed processing steps to correct color. (Landsat provides two tutorials on this.)

The most intriguing feature of Sentinel-2A is 3 channels at the boundary of red visible and near infrared termed the 'red edge'. There is nothing that corresponds to this in Landsat. Whether it will be useful in cryosphere work remains to be determined -- the data server remains broken and no examples have been provided in their image gallery. These are at 20 m and may not differ enough over ice to form a distinctively colored faux RGB.

Here are the bands as (peak, width, resolution):

B2 (490/65/10), B3 (560/35/10), B4 (665/30/10)
B5 (705/15/20), B6 (740/15/20), B7 (775/20/20), B8a (865/20/20)
B8 (842/115/10)

http://www.esa.int/spaceinimages/Images/2015/07/A_slippery_slope
« Last Edit: December 11, 2015, 01:30:20 PM by A-Team »

wili

  • Young ice
  • Posts: 2589
    • View Profile
  • Liked: 163
  • Likes Given: 301
Re: PIG has calved
« Reply #494 on: December 11, 2015, 03:29:41 PM »
"Thx for all those excellent posts! "

Seconded!
"A force de chercher de bonnes raisons, on en trouve; on les dit; et après on y tient, non pas tant parce qu'elles sont bonnes que pour ne pas se démentir." Choderlos de Laclos "You struggle to come up with some valid reasons, then cling to them, not because they're good, but just to not back down."

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 3996
    • View Profile
    • Arctische Pinguin
  • Liked: 619
  • Likes Given: 0
Re: PIG has calved
« Reply #495 on: December 13, 2015, 01:02:46 PM »
Conditions for landsat imagery on the 12th December where almost as good as those one cycle (16d) ago, described in post #466. Elevation of the "mid-night sun" is 7.84 degrees. Here is a detail of those nightly images at a resolution of 15m/pix showing some calving activity in "the notch".

(click to animate)

crandles

  • Young ice
  • Posts: 2512
    • View Profile
  • Liked: 94
  • Likes Given: 47
Re: PIG has calved
« Reply #496 on: December 13, 2015, 01:45:28 PM »
Has the calved iceberg's rotation stopped? Would you expect that once it has rotated, as it has, for it to continue to rotate as it is pushed or is it just as likely to stop and start such rotations? Does this depend mainly on how many places it is grounded at or strength and thickness of ice being driven into the grounding points or other factors?

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 3996
    • View Profile
    • Arctische Pinguin
  • Liked: 619
  • Likes Given: 0
Re: PIG has calved
« Reply #497 on: December 13, 2015, 03:07:41 PM »
Has the calved iceberg's rotation stopped? Would you expect that once it has rotated, as it has, for it to continue to rotate as it is pushed or is it just as likely to stop and start such rotations? Does this depend mainly on how many places it is grounded at or strength and thickness of ice being driven into the grounding points or other factors?

There is about 0.5o rotation between the images. Movement (in a straight line) with the glacier seems to dominate again. I looked at all the cracks visible in that iceberg and cannot see anything changing: no internal stress is building up. So it may be grounded, but it is not sticking.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 3996
    • View Profile
    • Arctische Pinguin
  • Liked: 619
  • Likes Given: 0
Re: PIG has calved
« Reply #498 on: December 14, 2015, 08:30:27 AM »
Forward movement and rotation can be judged by coding with different colors, Red for 12Dec, Green for 26Nov.

(click for a larger image)

oren

  • Young ice
  • Posts: 4478
    • View Profile
  • Liked: 872
  • Likes Given: 1287
Re: PIG has calved
« Reply #499 on: December 25, 2015, 12:10:13 PM »
Forward movement and rotation can be judged by coding with different colors, Red for 12Dec, Green for 26Nov.

(click for a larger image)

Thanks Wipneus. Great way of showing the movement.