Support the Arctic Sea Ice Forum and Blog

Author Topic: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland  (Read 574998 times)

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #450 on: August 21, 2014, 06:16:31 AM »
Oren, thx!  I've made a lot of other improvements in the annotated post-Eemian timeline for Greenland and will replace the previous version in due course. There is a new interpretive column and the 'best' reference for each line (by which i mean the most recent good journal article that is preferably open source, as anyone can chase down historical credits from that bibliography and push into the present with its cites in google scholar).

Some events left a record in a single annual layer in the ice core (suitable for a single line in the time-ordered database but others extended over thousands of years. Extended events can have very detailed internal substructure (depending on whether anyone has examined them in depth) as well as encompass the timing 'extraneous' single-layer events.

The radar stratigraphy cannot be dated nearly as accurately as an ice core from the echogram itself (see image below). However some of them are in fact single-year depositions of tephra over what was then the entire exposed surface of Greenland. With a good core chronology and knowledge of what gave rise to the reflector, they can in fact be dated.

So there are some display issues here along the lines of relational databases. However I think the 125 kyr timeframe can be cleanly partitioned along the lines of geological charts (eon era period epoch age). Not shooting for an authoritative account here, just a framework and initial outline that the next person with an interest can improve.


A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #451 on: August 21, 2014, 04:27:08 PM »
Here are a couple more displays that tie radar horizons onto chronologies of ice cores established by ring counting in the main Greenland ice cores, as checked by correlation with Antarctic cores and intermittently by independently dated volcanic eruptions such as Toba.

This is necessary because radar striations only come with depth below surface but not age. This depth is not linearly proportional to age either because the layers get thinner with depth due to compression and thinning. The images below show how to compensate for this to date radar layers.

The first I made from 804 match points between NEEM and NGRIP; note NGRIP annual layers become thicker from 22,250 years b2k on. The second uses the radar flight path between drill sites to make the correlation. It would be straightforward to extend this to NEGIS, GRIP, GISP2, DYE and other cores; alternatively, once a radar layer is dated that date can be extended by simply following the layer out on other flight tracks.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #452 on: August 22, 2014, 03:48:45 PM »
Here is a carefully constructed view of the age vs depth for NEEM and NGRIP ice cores by the collaboration team, as well as the relative thickness of annual ice layers. These dates can be transferred across the ice sheet using radar horizons (second image).

http://www.clim-past.net/9/2713/2013/

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #453 on: August 23, 2014, 04:22:13 PM »
Some fairly widespread misconceptions arose from one dramatic incident of very rapid drainage of a Greenland melt lake so other researchers were quick to chime in with a more nuanced papers (over a dozen of them). The hydrofracturing moulin mechanism only accounts for a small fraction of observed drainages and even in those, only the most abrupt really overwhelm the subglacial drainage system to the point of buoyant lifting of the ice overcoming its bottom sticking points. 

Overall, short-lived speedups associated with all types of lake drainage amount to only ~ 5% of all summer-specific ice motion, in part because the subglacial drainage system, whether initially distributed sheet or tunnel, is able to increase its capacity in response to inflows, dampening the effect.

To the extent this is true, exacerbating global warming will create more larger and earlier melt lakes; this will increase but not lead to a runaway effect on the overall ice sheet march to the sea. The greater effect may be advection of heat from the glacial surface (from its inefficient contact with the atmosphere) to deeper in the glacier where that energy can drastically reduce ice viscosity.

Here the motion is a few tenths of a km/yr. While that is only a thirtieth that of Jakobshavn, it may pertain to a larger portion of the ice sheet.

The excellent illustration from an excellent article tells the story of two lakes just north of Jakobshavn Isbrae in the variously spelled Paakitsoq region, NE of Swiss Camp and the Luethi study sites. I added the main findings and a few details to make the figure self-contained.

I then chased down three 2014 Landsats to see if the two lakes studied two years earlier were recurrent features with recurrent behavior: sort of, not really, may depend on melt year.

Ice dynamic response to two modes of surface lake drainage
Marco Tedesco et al  2013 Environ. Res. Lett. 8 (2013) 034007
http://iopscience.iop.org/1748-9326/8/3/034007/pdf/1748-9326_8_3_034007.pdf

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #454 on: August 23, 2014, 07:49:04 PM »
Peeking trough the clouds (Radarsat) we may have a calving situation between Aug 18 and Aug 22 2014
It also looks like we are having a depression developing north of the southern branch?:
« Last Edit: August 23, 2014, 07:57:51 PM by Espen »
Have a ice day!

Shared Humanity

  • ASIF Upper Class
  • Posts: 2311
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #455 on: August 24, 2014, 03:38:58 AM »
I'm not sure that is a depression. It might just be a shadow but that area north of the southern branch is grounded below water just like the north branch. It would not surprise me at all if we saw things change there quickly as the south branch  calving front retreats.

I will also not be surprised when that new island in between these two areas grounded below sea level pokes its head out of the ice.
« Last Edit: August 24, 2014, 03:44:29 AM by Shared Humanity »

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #456 on: August 25, 2014, 08:13:21 PM »
Some impressive calving seen at Jakobshavn between Aug. 19 and Aug. 25 2014:

Fits with what is already observed by Radarsat (see above).
Have a ice day!

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #457 on: August 25, 2014, 08:39:35 PM »
With the calving mentioned above, it looks like we are at the same point of retreat as the record set Sep. 27-29 2013, and we still have at least a month left of the traditional calving season for Jakobshavn:

« Last Edit: August 25, 2014, 10:02:44 PM by Espen »
Have a ice day!

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #458 on: August 28, 2014, 04:57:44 AM »
As suggested we had some calvings since last update, Aug. 27 2014 is almost at the maximum retreat set Sep. 27 2013, the northern branch is definitely in the retreat zone. The last frame in the animation below is Sep. 27 2013:
Have a ice day!

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #459 on: August 28, 2014, 02:56:38 PM »
Jakobshaven ice-stream calving-front imaged by Sentinel-1 SAR, 17.8.2014. Note in the lower right corner how differentiation between the ice-stream and ice melange with icebergs can be hard.

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #460 on: August 28, 2014, 04:26:23 PM »
Nukefix, "Note in the lower right corner how differentiation between the ice-stream and ice melange with icebergs can be hard.", is it not more due to the viewing angle?
« Last Edit: August 28, 2014, 05:09:24 PM by Espen »
Have a ice day!

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #461 on: August 28, 2014, 05:37:15 PM »
Nukefix, "Note in the lower right corner how differentiation between the ice-stream and ice melange with icebergs can be hard.", is it not more due to the viewing angle?
Even with the same viewing-angle the ease-of-differentiation depends on environmental conditions (season). Higher resolution would make the task easier but this is what we are stuck with. The use of textural measures might make the task a bit easier for a human operator.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #462 on: August 29, 2014, 02:56:32 AM »
Landsat LC80100112014239LGN00 just came in at EarthExplorer, it is a beauty. No sign of LC80832332014239LGN00 though, sun angle may be low for ascending orbit and they just are not processing that series (even though it might be fantastic for shadows on nearly flat terrain -- as so provide a very high resolution relative DEM!).

This is a very fine reference shot too for Eqip area glaciers, with icestreams standing out clearly from carried abraded material..

No retreat record ... calving some big ones but not quite keeping up with icestream advance. We have no idea on whether a discharge volume is being set -- I can try once again to see if feature tracking over time is possible from Landsat (to see how seasonal velocity compares to past years).

The north branch has remained very active but its front is not going anywhere.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #463 on: August 29, 2014, 03:16:13 AM »
Here are the higher resolution panchromatics from Landsat LC80100112014239LGN00, upsampled to 10m resolution.

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #464 on: August 29, 2014, 10:47:08 AM »

No retreat record ... calving some big ones but not quite keeping up with icestream advance. We have no idea on whether a discharge volume is being set -- I can try once again to see if feature tracking over time is possible from Landsat (to see how seasonal velocity compares to past years).
Feature-tracking with SAR works, lets hope that the S-1 coverage will cover more of Jakobshaven in the future.

BTW, are you aware of any public-domain feature-tracking code?

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #465 on: August 29, 2014, 05:08:27 PM »
"BTW, are you aware of any public-domain feature-tracking code?"

Short answer, no. I could look through more 'methods' sections of Greenland and Antarctica journal articles. Wipneus found a trackable hexagon at Zachariae; that gives velocity relative to fixed rock only at that one point. We are looking for the whole velocity vector field (including z height component).

Landsat scenes even with matching path/row don't have quite the same nadir, azimuth nor time-of-day but they seemed well aligned by the time they get to EarthExplorer. Sun elevation though, that changes illumination angle with the season. I posted that data a while back for cloud-free Landsat days of 2013-14.

Jakobshavn Isbrae moves very fast (peaked at 3 pixels/day at 15 m resolution) which is favorable but the icestream virtually featureless for 100 km upstream and small crevasse features may open and close over time. The flow is east to west with crevasses generally orthogonal; the sun is always to the south so shading is unfavorable.

Low sun angles (ascending orbit) later in day would better shadow crevasses and whatever surface relief may be out there. For example LC80822332014200LGN00 on midnight of 19 Jul 14 had an azimuth of -49.7 and sun elevation 7.4. It could be compared with the 27 Jul 14, LC80812332014208LGN00.

It is easy enough to make a very fast blink comparator in gimp with each date, say at 50% translucency. Alternatively, after masking all but the stream channel, one image can be subtracted from the other until the (local) optimal nudging is discovered for the later date (eg three pixels west, one pixel north). I believe some commercial software does nothing more than script this. In the example above, 8 days x 3 px/day focuses the search around 24 pixels to the west.

False color Landsat band mixes have potential but I haven't gotten anywhere with them to date, the interval between Landsats is too long and erratic. SAR doesn't have solar illumination issues. At Petermann, compositing 3 panchromatic B8 bands from different dates as RGB has given spectacular interferometric imagery with a fetching display of the velocity field (as projected on x,y plane).

Rignot et al  doi:10.1029/2012GL051634, 2012 "use Envisat ASAR, ALOS PALSAR and RADARSAT-1 SAR. The SAR data are processed into single look complex images using GAMMA (www.gamma-rs.ch). The data are then combined into radar interferograms after applying a pixel offset correction calculated using a speckle tracking algorithm [Michel and Rignot, 1999]. The 3-D vector of ice velocity is calculated assuming surface parallel flow using only the speckle tracking results, i.e., the interferometric phase is not employed."

Joughin 2014 http://www.the-cryosphere.net/8/209/2014/tc-8-209-2014.html processes proprietary pairs of DLR TerraSAR-X imagery separated by 11 days with "a set of well-established speckle-tracking techniques" described in an offline 2002 paper http://dx.doi.org/10.3189/172756402781817978 that has 181 follow-up cites.

Joughin 2010 http://www.igsoc.org:8080/journal/56/200/j10j204.pdf has a fine discussion of interferometric InSAR. Earlier in this forum (can't relocate), I posted a fall 2014 abstract that indicates Joaghin's current (advanced) methods for JI.

There are fixed GPS units out on the main ice sheet for slow motion (~50 m/yr) but I'm not aware of public access to real-time data. (SH Doyle Persistent flow acceleration within the interior of the Greenland ice sheet DOI: 10.1002/2013GL058933).

To summarize, the first obstacle (expensive frames) is eliminated with this new satellite though it has missed the summer season and the second obstacle (speckle, interferometric software) is probably surmountable. So I hope you can find a way of doing velocity with Sentinel.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #466 on: August 29, 2014, 05:47:17 PM »
Here is a hasty velocity animation at Jakobshavn over 8 days this summer July 11 v July 19 ... next time I would start with images that are better matched for illumination and at 10 m resolution. So it does look like it would be feasible to determine speeds along the centerline of the channel quite a ways up using this time interval though a given feature may not persist over the whole season. We have 21 cloud-free images for the 123 day interval May through August...

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #467 on: August 29, 2014, 06:52:54 PM »
That's the benefit of radar, the imaging-geometry and illumination are practically identical from observation to observation.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #468 on: August 29, 2014, 07:15:05 PM »
That would be a big plus. How does radar do on snowy scenes though? Seems like surface roughness takes up quite a bit of channel space.

Here is the 09 Feb 14 band 4 Landsat (after a lot of contrast meddling):

Shared Humanity

  • ASIF Upper Class
  • Posts: 2311
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #469 on: August 29, 2014, 11:21:26 PM »
As suggested we had some calvings since last update, Aug. 27 2014 is almost at the maximum retreat set Sep. 27 2013, the northern branch is definitely in the retreat zone. The last frame in the animation below is Sep. 27 2013:

Looking at the animation, it sure looks like the north wall of the southern branch is calving.

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #470 on: August 30, 2014, 08:16:18 PM »
Shared Humanity,

Yes, I also expect a new island showing up in the near future.
Have a ice day!

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #471 on: August 31, 2014, 09:21:41 AM »
That would be a big plus. How does radar do on snowy scenes though? Seems like surface roughness takes up quite a bit of channel space.
Snow and snow wetness affect the level of backscatter. However, feature-tracking is based on tracking visible features (crevasses), so changing the level of backscatter should not matter that much especially if suitable pre-processing is used to mitigate differences in backscatter-levels.

If anyone would have some feature-tracking code, even in pseudocode, I could try to convert that to a usable open-source feature-tracking module.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #472 on: August 31, 2014, 03:07:40 PM »
"... feature-tracking code."

One thing to consider at the outset is measurement error vs available sensor resolution, especially whether final output will allow year on year comparison at say monthly resolution.

For Landset 15 m resolution on the Jakobshavn icestream, suppose a crevasse is moving at 30 m per day. That would be 2 pixels over in a region where the movement is pure east to west (ie compatible with the rectangular pixel grid). However movement confined to the JI channel is varying from that to almost diagonal.

One pixel up, one pixel over amounts to 1.4 pixels of diagonal motion, yet this binning, on top of inevitable Landsat scene to scene registration error, adds up. So here we would not be looking at better than monthly velocities where there are 60 pixels of movement for the 11 km/yr velocity assumed above.

The table below shows how measured km/yr varies with a meter change in meters per day. It follows that Landsat panchromatic might be able to detect a speedup/slowdown relative to 2013 over the melt season but would struggle with month to month comparison.

m/d   px/d   px/m   km/yr
25   1.7   50.7   9.1
26   1.7   52.7   9.5
27   1.8   54.7   9.9
28   1.9   56.7   10.2
29   1.9   58.8   10.6
30   2   60.8   11
31   2.1   62.8   11.3
32   2.1   64.9   11.7
33   2.2   66.9   12
34   2.3   68.9   12.4
35   2.3   70.9   12.8

One thing I've pondered (no doubt definitively addressed by previous ponderous scholarly articles) is whether the two most common methods for upsampling resolution (bicubic and sinc) merely provide delusional eye candy or -- on some terrains at least -- do a decent job of getting at the same result had the instrument resolution actually been better.

I suppose NASA already pursued this when deciding to post Landsat-8 at 15 m, yet that may have been for land whereas the homogeneity and predictability of the Greenland icesheet might make this more favorable (indeed across the board, for all manner of Greenland data). Here the 15 m could be blurred to 30 m (or band 4 30 m used) and then upsamped for subtractive comparison to the original 15 m. The bottom line will be we are better off with Sentinel radar.

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #473 on: September 01, 2014, 10:16:53 AM »
Ideally the image posting should satisfy Nyquist's sampling theorem so a 30m resolution sensor should be sampled/posted at a 15m pixel-size. Feature-tracking is performed over a window and sub-pixel accuracy in the speed-measurements is certainly possible.

Radar-images have a much more stable geometry so it's possible to co-register the scenes to 0.1 pixel (or even 0.001 pixel) accuracy. However, ionospheric disturbances cause some distortion in radar-images when the Sun is very active, these problems show up as banding in the velocity-images and can be largely mitigated with suitable filtering.
« Last Edit: September 01, 2014, 03:02:00 PM by nukefix »

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #474 on: September 18, 2014, 09:21:42 PM »
Due to weather conditions, we have not received Landsat images for some time, but we still have Radarsat, and we may have some heavy activities at Jakobshavn during September 11 and September 13 2014, will report more on this later when more reliable data is available:
Have a ice day!

AbruptSLR

  • ASIF Emperor
  • Posts: 13273
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #475 on: September 25, 2014, 07:23:05 PM »
The attached Sentinel 1-A image of the Jakobshavn Glacier was taken on Sept 23 2014 and shows some more details of the recent calving on the southern branch.

See the following link to get the high or low resolution image yourself:

http://www.esa.int/spaceinimages/Images/2014/09/Jakobshavn_Glacier_Greenland
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #476 on: September 25, 2014, 09:39:51 PM »
Thanx AbruptSLR for keeping me update! :) ;)
Have a ice day!

AbruptSLR

  • ASIF Emperor
  • Posts: 13273
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #477 on: September 25, 2014, 09:59:14 PM »
Espen,

If A-Team was still on-line I sure that he could pull some imagining magic to show all kinds of details that my posted image misses; but it looks like there has not been much (if any) new calving since the event you identified on Sept 13.

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

Lennart van der Linde

  • ASIF Middle Class
  • Posts: 697
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #478 on: September 28, 2014, 07:42:35 AM »
This was posted by ASLR in the Antarctica-folder, but it may be of interest here too:

The link leads to an open access pdf that provides valuable information about the bed topography of both Jakobshavn (Greenland), and Byrd (Antarctica), Glaciers.

S. GOGINENI, J.-B. YAN, J. PADEN, C. LEUSCHEN, J. LI, F. RODRIGUEZ-MORALES, D. BRAATEN, K. PURDON, Z. WANG, W. LIU, & J. GAUCH, (2014), "Bed topography of Jakobshavn Isbræ, Greenland, and Byrd Glacier, Antarctica", Journal of Glaciology, Vol. 60, No. 223, 2014 doi: 10.3189/2014JoG14J129

http://www.igsoc.org/journal/60/223/j14j129.pdf

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #479 on: September 28, 2014, 11:50:51 AM »
Hold on AbruptSLR,

I do think we have a new development, according to Modis we passed the previous max. retreat set in September 2013, looking forward to study the next usable landsat image.
Have a ice day!

AbruptSLR

  • ASIF Emperor
  • Posts: 13273
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #480 on: September 28, 2014, 03:55:50 PM »
Espen,

It looks like you will have a lot of new information to report in your October summary on Greenland Glaciers.

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

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #481 on: September 29, 2014, 04:35:43 AM »
Here is the confirmation, we passed the previous retreat record set in September 2013, both the southern branch (slightly) and northern branch (especially) retreated further:

Ps. Notice how the snow/ice retreats on the rocks.
« Last Edit: September 29, 2014, 05:31:49 AM by Espen »
Have a ice day!

sidd

  • ASIF Upper Class
  • Posts: 1704
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #482 on: October 06, 2014, 06:43:22 AM »
Prasad Gogineni  and his henchmen have a paper out on subglacial topo under Jacobshawn (and Byrd, in Antarctica), including juicy ice layer data from CRESIS.

doi: 10.3189/2014JoG14J129

Mr. A-Team should be all over this when he reappears.

sidd

icefest

  • ASIF Citizen
  • Posts: 258
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #483 on: October 06, 2014, 12:14:16 PM »
Prasad Gogineni  and his henchmen have a paper out on subglacial topo under Jacobshawn (and Byrd, in Antarctica), including juicy ice layer data from CRESIS.

doi: 10.3189/2014JoG14J129

Mr. A-Team should be all over this when he reappears.

sidd

I can't get more than: http://www.igsoc.org/journal/60/223/j14j129.swf
If anyone can find a copy, I'd be appreciative.

-icefest


EDIT: found it!: http://www.igsoc.org/journal/60/223/j14j129.pdf
Open other end.

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #484 on: October 22, 2014, 08:36:55 PM »
As seen with Sentinel-1 on 19.10

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #485 on: October 24, 2014, 10:56:57 AM »
Here's a quick-and-dirty 1st guess for the GL location based on the S-1 data above:

Shared Humanity

  • ASIF Upper Class
  • Posts: 2311
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #486 on: October 24, 2014, 09:12:18 PM »
The heavily crevassed north flank of the south branch is all grounded below sea level. I believe the surface condition of this region is evidence of water melting the underside of the glacier.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #487 on: November 02, 2014, 04:32:45 PM »
"The heavily crevassed north flank of the south branch is all grounded below sea level ...evidence of [sea] water melting the underside of the glacier."

A lot of Greenland is below sea level. However that does not mean that sea water has infiltrated the interior; quite the contrary, fresh water is flowing out. Warming of Baffin Bay primarily affects the calving front of JI and there mostly because of calving induced turbulence which improves heat transfer.

Retreat this year again stopped short of the big sill shown in figure 20b of the Gogineni paper cited above. According to published predictions, this sill will reduce heat exchange with sea water in coming years.

However, you would be correct to say that the bottom of Jakobshavn is not yet well characterized by ice-penetrating radar (despite heroic efforts): the resolution leaves a lot to be desired in terms of resistive knobs; the height distribution of 'temperate' ice is not well known and the composition, height  and hydration state of bottom till have not been determined; and the sole temperature drill profile in the main fork (Iken 1993) never reached bedrock.

While modeling is quite sophisticated these days, at some point it must make contact with experimental observation. Unfortunately the key parameters are precisely those characterizing the lower depths of the channel (softness of ice, resistance to basal slipping). Tight flight patterns of the new radar drone may help in coming seasons, as would new high speed drilling techniques (AGU abstract C53A-0279: ten days for a 3300 m core that includes bedrock).

We might wonder if warmer fjord water in winter will mean later frozen melange and so less resistance to late-season calving. The Landsat season ends this week after two fine low angle shots in late October so we can't monitor open water with that satellite. Sentinel will still work but there we have no public domain imagery from previous years for melange comparison. The fjord web cam might have that data depending on how well it works in dim light.

The images below show the last available Landsat 15 m panchromatic. The extreme western portion of the fjord is still open water.
« Last Edit: November 02, 2014, 04:39:18 PM by A-Team »

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #488 on: November 02, 2014, 06:19:59 PM »
Jakobshavn gained some weigth since our latest update, when Jakobshavn reached the maximum retreat (Sep. 28  2014) for the season and overall:

Have a ice day!

oren

  • ASIF Upper Class
  • Posts: 1588
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #489 on: November 02, 2014, 07:12:54 PM »
Interesting: the south branch advanced quite a lot while the north branch did not advance at all during this period

Rubikscube

  • ASIF Citizen
  • Posts: 247
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #490 on: November 03, 2014, 01:15:52 AM »
Interesting: the south branch advanced quite a lot while the north branch did not advance at all during this period

The northern branch, especially its eastern part, is virtually stuck on land, so it should not come as a big surprise that it advances very slowly even though it is noteworthy. I would love to see some physical depth measurements taken from the edge of the north branch, but I suspect that anyone taking on such a task might run into some serious problems (such as health and safety requirements).

It is also noteworthy that the "third branch" appears to be advancing fairly rapidly, although the crevasses and apparently low lying ice edge makes it a little difficult to establish exactly how much ice is exiting there.

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #491 on: November 03, 2014, 01:34:58 PM »
Right, the north branch is mostly up against a cliff as can be seen from the bedrock map or from midsummer oblique imagery. From the velocity map, prospects are dim for significant ice volume discharge (relative to the south branch) in coming decades. The middle branch may have better prospects (below).

Much farther to the east, the higher velocity dark blue lobe does not correspond well with the red active tributary of the south branch. So it doesn't really work to assume past glacial gouging of bedrock serves as a proxy to ice volume contributions today or later in the century. The images below are adapted from the latest Gogenini pdf cited above.

The animation below shows a surprising amount of ice movement in the south branch between the Oct 30 and Nov 1 Landsat scenes. Here the scene geometry was not easy to align: the satellite was at the same latitude and azimuth for both but scene-center longitudes (nadirs) were quite different, -47.80º vs -50.89º, or 120 km apart.

The latter has quite remarkable shadows, though the sun elevation at 5.99º is not that much lower than that of the 30th, 6.63. Recall shadow length goes as 1/tan, tan small angles ~ angle itself in radians, and radians convert to degrees by a constant so the ratio of degrees 6.63/5.99 indicates only a 10.7% increase in shadow length in the two day later photo.

A vertical 1.00 meter stick would cast shadows of 8.60 and 9.53 meters at the respective angles (using radians). These scenes thus have considerable value in determining very slight variations in ice surface topography.

The final Jakobshavn Landsat from 2013 was taken on Nov 7 and things did not resume until Feb 9. The lowest acceptable sun elevation was 5.41º for the November scene. I'll post the final database of clear images in a week or so.
« Last Edit: November 03, 2014, 01:46:59 PM by A-Team »

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #492 on: November 03, 2014, 10:14:36 PM »
Should you ever wonder exactly what experimental data underlies the pretty bedrock elevation picture at a given location, the first image below shows all the radar flight segments available for the north and south branches of Jakobshavn from 1993-2013. Very few other sites in Greenland have such a high density of flight lines.

There's a whole lot of flight lines but still a whole lot of sq km regions with no coverage whatsoever. What could possibly be a rational basis for interpolation in fairly irregular bedrock terrain? All that's available is the bounding polygon.

We've been wondering what underlies the middle branch so I selected a recent track that sliced through the area (newer radar usually represents better technology), namely a P3 overflight from 2013 that Cresis staff marked up for ice surface (does not include firn) and bedrock, namely 20130404_02_032_2echo_picks, available at https://data.cresis.ku.edu/data/rds/. The 1echo_pick file has raw unannotated data.

After much back-and-forth between Cresis kml, Google Earth pins, Cresis images and Gimp, it's possible to place the desired radar data segment along the track as displayed over the satellite imagery that GooglE uses at this resolution (second image below). Interior Greenland does not have any visible features in satellite imagery (uniform snow).

The dotted green lines show possible troughs gouged out by past ice stream movement. The three minor dips would have to be traced farther east in other flight lines for confirmation.

The P3 flight, being oblique both N-S and to JI main channel, is rotated to horizontal so the precious radar data can be displayed at its native resolution. (It's more complicated but still doable when the track is curved.) The Cresis image here spans 52 km but only provides 4 internal lat,lon coordinates; none of these correspond to internal GPS way points of the Cresis kml file.

I'll skip over other crazy-making details unless there is a public clamor. (Cresis pdfs still have lat,lon way points as text; these can be grepped out site-wide into GooglE path pins, along with internal way points of the kml files themselves, and the whole enchilada distributed as a one-time GooglE import, over an elevation contour base map.)
« Last Edit: November 04, 2014, 01:22:26 PM by A-Team »

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #493 on: November 04, 2014, 05:13:44 PM »
Thanks A-Team, that explains a lot! :)
Have a ice day!

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #494 on: November 05, 2014, 02:30:21 PM »
Lennart, ASLR, sidd, and icefest ask above for comment on the new Gogenini paper at http://www.igsoc.org/journal/60/223/j14j129.pdf . This is a beautifully written paper that provides significant new information on the Greenland radar program. It would take several posts to cover the ground.

I first puzzled over why the paper used only radar data from 2005-09 for Jakobshavn, whereas a great many overflights with presumably better technology took place over the subsequent years (first image below shows 2010-14 tracks).

It has to do with pretenders to the ice-penetrating radar throne (emphasis added):

'Since 1990, several attempts have been made to sound Jakobshavn and other fast-flowing glaciers, with very limited success. Fully coherent radars operating over the frequency range 1–450 MHz are used for these measurements (Gogineni 2001, 2012; Dall 2012; Morlighem 2014). We first succeeded in sounding Jakobshavn Isbræ ... in 2005.'
Chasing this down, 'Dall 2012 P-band radar sounding in Antarctica' concerns a threat from satellite radar, which could provide more comprehensive coverage of big continents if only the surface clutter issue could be resolved. That radar (ESA Polaris) on airplanes not only reached bedrock in 3300 m ice on the Greenland summit ridge (possibly detecting basal melt as well) in 2008 but also obtained spectacular results (images below) in Antarctica using a remarkably effective enhancement (first derivative of nadir return power) to better get at internal stratifications.

As you might expect, ImageJ is all over gradients, laplacians and hessians of digital imagery, eg http://bigwww.epfl.ch/thevenaz/differentials . Gimp offers Filters → Artistic → Van Gogh (line integral convolution), various edge detectors, Difference of Gaussians, or roll your own algorithm (bump one pixel up and down, subtract grayscales, take max).

However derivatives would be better taken directly, possibly even in hardware, on incoming 12 bit return data -- in fact the whole Cresis effort should be integrated from antenna design right through digital enhancement product -- too often the latter is an afterthought or salvage effort after the goal post moves (eg from bedrock to radar stratigraphy, internal deformations and water pockets).

Which is exactly what happened here: five years after electrical engineers design the radar instrument and IceBridge conducts the flights, they bring in some poor computer science guy (11th author http://www.csce.uark.edu/~jgauch/) to recover the latent data via histogram equalization, adaptive median and bas relief edge-detection filtration, all of which are standard menu items in Gimp.

Morlighem 2014 combined sparse radar ice-thickness data with high-resolution ice motion interferometery in a computer-intensive mass conservation optimization model, finding 'deeply incised submarine glacial valleys beneath the Greenland ice sheet' (doi: 10.1038/ngeo2167, paywalled, have copy) to determine ice thickness and bed topography along the entire periphery of the Greenland ice sheet, detecting widespread overdeepenings in the faster moving ice streams that extend 'significantly deeper below sea level and farther inland than previously thought'.

Omg, 'previous thinkers' must mean Gogenini and co-authors of his many papers?!? The response establishes that while Cresis did not invent ice-penetrating radar, they certainly built ever-improving scientific instruments, pioneered their use on a survey-level scale (unlike Polaris) and succeeded in reaching bedrock observationally in very difficult deep ice by 2006 (whereas Morlighem 2014 merely models it).

Note however this modelling does fill in the blanks of bedrock polygons bounded by inevitably sparse radar tracks (see post above), so it's worth a more detailed look as interpolative method.

The 2014 model paper chose to cite a 2001 paper for purposes of invidious comparison (ref 7: Bamber, Gogenini 2001). It is difficult to comprehend how the Rignot group could be unaware of the intervening updates, notably the widely cited replacement (Bamber, Gogenini 2013 doi:10.5194/tc-7-499-2013) given Rignot is listed as co-author. That went online 15 Nov 2012, whereas the Morlighem article was submitted a full year later, 20 Nov 2013.

Lennart van der Linde

  • ASIF Middle Class
  • Posts: 697
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #495 on: November 05, 2014, 10:36:45 PM »
Thanks A-Team for your comments on the Gogineni paper.

What I keep wondering about: how important is it that the trough of Jakobshavn Isbrae is at least 100m (?) below sea level all these 200 km inland? How much more vulnerable to melting does this make this glacier compared to glaciers grounded above sea level? Or is this less important due to the relatively small width of this trough, which would cause much tighter kinematic constraints compared to for example Thwaites glacier?

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #496 on: November 06, 2014, 12:38:05 PM »
Lennart, good question.  Indeed what keeps sea water from rushing in and the whole mountain of ice bobbing up, like submerging a ice cube tray in a sink of warm water? So much of north-central Greenland is below sea level that it is seems like one big grounded iceberg.

We should get around to a buoyancy discussion at some point -- for lower JI, it is provided but not discussed much  in the Joughin 2014 paper. It's clearly explained in Martin Lüthi's course "Physics of Glaciers I"  free online at http://people.ee.ethz.ch/~luethim/teaching.html and in a 2002 paper "Hydraulic conditions at the base of Jakobshavn Isbrae, West Greenland" (http://people.ee.ethz.ch/~luethim/pdf/40A013.pdf). Also relevant: their new paper on boreholes Foxx and Gull on a flow line below Swiss Camp (http://www.the-cryosphere-discuss.net/8/5169/2014/tcd-8-5169-2014.html).

It is also key to get a grip on the complex hydraulic connectivity of subglacial water. For that, the 2 Oct 14 Lüthi paper at http://www.nature.com/nature/journal/v514/n7520/full/nature13796.html. Two of the co-authors expand on the paper at http://www.sciencedaily.com/releases/2014/10/141001185934.htm

As far as I can tell, the bottom line on Jakobshavn acceleration is that there is no bottom line, yet. One causal mechanism after another has been ruled out. Only the initial phase of speedup is understood, collapse of its floating ice shelf portion from warming ocean water of Baffin Bay. That was years ago. There's minimal ice shelf today, unless you count the seasonally frozen mixture of fjord and calved products (mélange).

Back-pressure from mélange could account for some portion of calving seasonality. If so, then later freeze-up and earlier melt-out (again, warmer ocean underneath) would increase winter calving episodes of Jakobshavn Isbrae and overall velocity.

It is not feasible to monitor calving in winter at visible wavelengths, though a combination seismometers, web cams, calving seiches, Sentinel radar and Modis SST might be up for the job. I'm seeing a 2010 AGU abstract on this very topic but, four years later, not seeing a published paper. That makes me disinclined to rush into winter monitoring.

Winter MODIS observations of West Greenland fjord ice activity
Abstract C51A-0472 presented at 2010 Fall Meeting, AGU
Cassotto, R. K., M. A. Fahnestock and J. M. Amundson
http://www.science.gov/topicpages/f/fjord+west+greenland.html
« Last Edit: November 07, 2014, 12:26:06 PM by A-Team »

Lennart van der Linde

  • ASIF Middle Class
  • Posts: 697
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #497 on: November 06, 2014, 01:17:00 PM »
That's some nice homework for me, thanks :)

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #498 on: November 06, 2014, 09:34:13 PM »
I am a bit confused about the present (southern) calving area, the images (data) above do not show that, but I suspect there is a "doorstep /threshold" in that area, resulting in a pause of further retreat?
« Last Edit: November 06, 2014, 09:56:39 PM by Espen »
Have a ice day!

sidd

  • ASIF Upper Class
  • Posts: 1704
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #499 on: November 06, 2014, 11:05:26 PM »
"...what keeps sea water from rushing in and the whole mountain of ice bobbing up ...?"

A good phrase to look for is Volume Above Flotation (VAF)