Support the Arctic Sea Ice Forum and Blog

Author Topic: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland  (Read 1199996 times)

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1250 on: February 11, 2016, 10:59:17 PM »
It is a pity that Sentinel releases their images twisted in different angles and various resolutions, yes very user friendly so to speak!

Have a ice day!

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1251 on: February 11, 2016, 11:44:12 PM »
Same Sentinel at full resolution, inset in upper left from 03 Feb 16. Seems like most of the action is in the north corner of the calving front. Icebergs ... the big dark ones look new. Iceberg attribution is getting harder because of north side activity but I would say most of these are from the main ice stream. Again, we're thinking the darker ones have rolled on their side being tall and thin (unstable) whereas the whiter ones have not rolled and still have the original crevasse field on top.

johnm33

  • Guest
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1252 on: February 12, 2016, 12:35:42 AM »
It is another tidal peak so maybe we'll see a big flush tonight, there's a big crack on 79N worth a look too.

sidd

  • First-year ice
  • Posts: 6774
    • View Profile
  • Liked: 1047
  • Likes Given: 0
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1253 on: February 12, 2016, 05:37:51 AM »
"the end of calving seasons, now it is a year around business"

Like tornado season in the midwest. Used to have one tornado season a year, then two, and now it all year.

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1254 on: February 13, 2016, 09:04:30 PM »
Adding to Solartim27 observations on February 11 2016, Jakobshavn is continuing massive calvings, seen in this not aligned animation due to Sentinel obsession to do things more difficult than necessary, hope you all can be happy with that?

https://twitter.com/Ecoverycom



Click on image to enlarge and animate!
« Last Edit: February 13, 2016, 09:50:08 PM by Espen »
Have a ice day!

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1255 on: February 13, 2016, 11:48:25 PM »
A better aligned version of the calving session between February 11 and February 13 2016:
Have a ice day!

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1256 on: February 14, 2016, 12:08:12 AM »
Blasting away in the dead of winter. I wish there was data from previous years to compare it to, but it seems rather massive.
I wonder about the current calving line position compared to where it was at the end of summer.

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1257 on: February 14, 2016, 12:14:14 AM »
Eyeballing stuff from September it seems the calving front is still quite advanced since then, despite the wintertime calving. I would still definitely expect another record retreat this year.

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1258 on: February 14, 2016, 12:50:01 AM »
Eyeballing stuff from September it seems the calving front is still quite advanced since then, despite the wintertime calving. I would still definitely expect another record retreat this year.

Yes, there is still some more to go before we are in the retreat-zone, but the southern branch seems wider now than earlier.
Have a ice day!

TenneyNaumer

  • New ice
  • Posts: 76
    • View Profile
    • Climate Change: The Next Generation
  • Liked: 2
  • Likes Given: 22
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1259 on: February 14, 2016, 06:59:17 AM »
Does anyone think it possible that sea water may have begun to infiltrate the ice in the below-sea-level areas that are in the general area to the east of what used to be the "fork" between the north and south branches? I believe someone posted a map of the topography, and it shows deeper areas in that region.

johnm33

  • Guest
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1260 on: February 14, 2016, 12:48:12 PM »
Does anyone think it possible that sea water may have begun to infiltrate the ice in the below-sea-level areas that are in the general area to the east of what used to be the "fork" between the north and south branches? I believe someone posted a map of the topography, and it shows deeper areas in that region.
It doesn't have to be sea water, the calving front is quite a way from the sea, but the tides will have some effect backing up the outflow and causing some movement wherever there's water beneath the ice, despite the thickness of the ice. It's probably no coincidence that recent movements occured around the periods of peak tidal range. http://www.dmi.dk/fileadmin/user_upload/vandstand_txt_pdf/2016/Ilulissat2016.pdf 2.6m on both the 11th feb and jan. Blowing up hycom there seems to be a more or less constant flow of fresher water escaping from Disko bay.

AbruptSLR

  • Multi-year ice
  • Posts: 19703
    • View Profile
  • Liked: 2268
  • Likes Given: 286
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1261 on: February 14, 2016, 04:39:26 PM »
Does anyone think it possible that sea water may have begun to infiltrate the ice in the below-sea-level areas that are in the general area to the east of what used to be the "fork" between the north and south branches? I believe someone posted a map of the topography, and it shows deeper areas in that region.

The attached re-posted figure is from Gladish et al. (2015, Part I), and shows how the modified Irminger Current water reaches the calving front of the Jakobshavn Glacier and accelerates cliff failure calving events, while the fresher water floats on top of the deeper ocean water.

Carl V. Gladish, David M. Holland, Aqqalu Rosing-Asvid, Jane W. Behrens, and Jesper Boje, 2015: Oceanic Boundary Conditions for Jakobshavn Glacier. Part I: Variability and Renewal of Ilulissat Icefjord Waters, 2001–14. J. Phys. Oceanogr., 45, 3–32, DOI: 10.1175/JPO-D-14-0044.1
“It is not the strongest or the most intelligent who will survive but those who can best manage change.”
― Leon C. Megginson

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1262 on: February 16, 2016, 06:04:12 PM »
Winter, what winter?

Big movements and calving at Jakobshavn:

Please click to enlarge and animate!

https://twitter.com/Ecoverycom
« Last Edit: February 16, 2016, 06:16:25 PM by Espen »
Have a ice day!

johnm33

  • Guest
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1263 on: February 16, 2016, 06:43:16 PM »
I can't get that to download at the moment, I'll try later but looking at these, yesterday looked like something might be happening but todays looks huge, if that is a bow wave outside the bay.
http://ocean.dmi.dk/arctic/images/MODIS/Disko/20160215s01a.ASAR.jpg
http://ocean.dmi.dk/arctic/images/MODIS/Disko/20160216s01a.ASAR.jpg
added. Its a similar story just north at Uummannaq, yesterday some hints, today massive outflow, eyeballing those waves they look around 4k apart, so unless it's some sort of cloud effect absolutely massive.
http://ocean.dmi.dk/arctic/images/MODIS/Uummannaq/20160215s01a.ASAR.jpg
http://ocean.dmi.dk/arctic/images/MODIS/Uummannaq/20160216s01a.ASAR.jpg
« Last Edit: February 16, 2016, 06:59:17 PM by johnm33 »

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1264 on: February 16, 2016, 07:37:43 PM »
looked like something might be happening but todays looks huge, if that is a bow wave outside the bay.

 it's some sort of cloud effect
Can just barely see the new calving activity through some haze.  Looks like some serious wind blew away the ice in the bay.  I was struck by the mid channel opening seen between the 11th and 16th
(Edit:  I zoomed out and scanned back over a few days.  The ice that was in the bay isn't blown out of the bay, it mostly melts!)
« Last Edit: February 16, 2016, 07:46:14 PM by solartim27 »
FNORD

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1265 on: February 16, 2016, 10:44:59 PM »
Hmm, we need to take a look at the new Sentinel !A... I am doing the two below, download takes forever as they don't mirror and almost always drop the connection, not remembering where they left off. Clueless, as in 1995 skill set.

Unzip, open in ImageJ. Crop to reduce file size. Flip to get them right-side up. Rotate to get north up. See how bad contrast is. Dumb down to 8-bit gray. Crop to 700 pixel forum width if possible. Save as png. Repeat hh and hv polarizations, make RGB if appropriate. 

The HH is at full scale, the HV at 2x. The IW usually has better resolution, still downloading (the last 15 seconds is really dicey, takes 2-3 minutes).
« Last Edit: February 16, 2016, 11:46:46 PM by A-Team »

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1266 on: February 16, 2016, 10:47:19 PM »
Winter, what winter?

Big movements and calving at Jakobshavn:

Please click to enlarge and animate!

https://twitter.com/Ecoverycom

Quite a large calving, and the northeast corner of the south branch totally crumbled. It seems JH decided to calve mainly on the right-hand side of the flow.
See how the whole fjord is full of overturned icebergs.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1267 on: February 16, 2016, 11:59:57 PM »
Here is what Sentinel 1A is capable of, run in IW GRDH mode. It is a day earlier than the images two posts back. (Hours of separation from the images below can also be figured from the times in the file name.)

S1A_IW_GRDH_1SSH_2016 02 15 T:204702
S1A_EW_GRDH_1SDH_2016 02 16 T:100727


I added the north branch at half-resolution to show cracks in its melange and the flow line of south branch calving products that keep the north branch penned in.
« Last Edit: February 17, 2016, 11:09:44 AM by A-Team »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1268 on: February 17, 2016, 03:27:01 PM »
The animation below shows the 10 available high resolution Sentinel images over the winter, from 18 Oct 15 to 15 Feb 16. There is a lot of back-and-forth of the calving front so I've set the animation speed fairly slow at 1200 ms.

There are two additional mid-February images taken from a very different orbital perspective which can't be readily aligned with the others (ideally everything should be put into Landsat mercator) but give almost daily coverage of the calving front (16th, 15th, 13th) rather than the 12 day orbital spacing of the animation.

11 Feb 16 s1a-iw-grd-hh-20160211.png
13 Feb 16 s1a-iw-grd-hh-20160213.png

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1269 on: February 17, 2016, 05:45:02 PM »
The animation below shows the 10 available high resolution Sentinel images over the winter, from 18 Oct 15 to 15 Feb 16. There is a lot of back-and-forth of the calving front so I've set the animation speed fairly slow at 1200 ms.

There are two additional mid-February images taken from a very different orbital perspective which can't be readily aligned with the others (ideally everything should be put into Landsat mercator) but give almost daily coverage of the calving front (16th, 15th, 13th) rather than the 12 day orbital spacing of the animation.

11 Feb 16 s1a-iw-grd-hh-20160211.png
13 Feb 16 s1a-iw-grd-hh-20160213.png


Thanks for the great animation.
While the left side of the flow has quite advanced, the right side is more or less back to where it was in mid-November. And so is the north wall.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1270 on: February 17, 2016, 07:01:38 PM »
Here is that closely spaced Sentinel !A sequence of 11 13 15 Feb 2016 ... not really in the same geometrical projection but good enough for tracing icebergs and calving events. Click to full width which is 75% of full resolution.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1271 on: February 18, 2016, 09:10:30 PM »
The animation below of Jakobshavn looks at the feasibility of winter velocity determination using iw-grd-hh Sentinel imagery. This has not been possible previously using open source imagery.

For this to work, the images have to aligned with pixel-perfect accuracy (which I did to cracks in fixed rocks in the SW corner) and the Jakobshavn ice stream has to have stable features that can be tracked from scene to scene. This could get quite dodgy in the case of a crevasse that was widening at it approached the calving front, so I took the cut out and beyond the 76º elbow.

This made the animation too large so I dumbed down the resolution 2x (which is actually a good thing). Visually, feature flow is quite apparent (don't look at the calving front, been there done that above). Note there is a single 12 day gap where they did not take a scene on 30 Oct 15.

This took just under an hour to make during which I could also delete the day's emails. The processing steps are quite complicated though, especially in contrast-matching as the originals are all over the place; the slightest mistake on a file with 429,346,958 rgb 16-bit pixels can be very costly.

It needs a click to animate and few seconds to get up to speed. I've left in the north bank of the south branch as quite a few folks are following developments there.

The second animation is bumped back up to Sentinel resolution, cropped to the straightaway, and is rotated to pure horizontal motion to simplify pixel displacement counting (a feature moving 12 pixels to the left over 12 days has a velocity of 120 m/12 days x 365 ÷ 1000 m/km = 3.65 km/yr). Averaging over a longer period would be more accurate provided the particle being tracked didn't originate around the elbow.

The third animation below is a crop of same data shown in the first. Here I am just seeing if wavelet denoising makes it easier to track blob velocities. Wavelets are a warmed-over version of fourier bandpass filtering nicely implemented in G'mic and Gimp. It should display without a click

In terms of predictions for end-of-season 2016, the two products are position of maximal calving front retreat and the bounding line of ice in the 10 Oct 2015 destined to be calved off by 10 Oct 2016.

In terms of ominous developments, the two things to watch for are bulk disintegration in from the calving front like we saw last August and increasingly high velocity during the winter months.

s1a-iw-grd-hh-20151018t204655-20151018t204720-008212-00b8da-001.tiff
s1a-iw-grd-hh-20151111t204649-20151111t204714-008562-00c22e-001.tiff
s1a-iw-grd-hh-20151123t204649-20151123t204714-008737-00c70e-001.tiff
s1a-iw-grd-hh-20151205t204705-20151205t204730-008912-00cc0c-001.tiff
s1a-iw-grd-hh-20151217t204704-20151217t204729-009087-00d0d5-001.tiff
s1a-iw-grd-hh-20151229t204704-20151229t204729-009262-00d5db-001.tiff
s1a-iw-grd-hh-20160110t204703-20160110t204728-009437-00dad1-001.tiff
s1a-iw-grd-hh-20160122t204703-20160122t204728-009612-00dff2-001.tiff
s1a-iw-grd-hh-20160203t204703-20160203t204728-009787-00e501-001.tiff
s1a-iw-grd-hh-20160215t204702-20160215t204727-009962-00ea20-001.tiff
« Last Edit: February 18, 2016, 11:05:13 PM by A-Team »

solartim27

  • Grease ice
  • Posts: 599
    • View Profile
  • Liked: 42
  • Likes Given: 75
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1272 on: February 18, 2016, 09:30:46 PM »
The animation below of Jakobshavn looks at the feasibility of winter velocity determination using iw-grd-hh Sentinel imagery.
Stunning, well worth the click.
FNORD

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1273 on: February 18, 2016, 10:07:39 PM »
The animation below of Jakobshavn looks at the feasibility of winter velocity determination using iw-grd-hh Sentinel imagery.
Stunning, well worth the click.

Stunning indeed. And ominous. It looks like the whole ice sheet is flowing into the fjord.
If anyone manages to predict the bounding line of ice to be calved, it will not be complete without marking the ice coming in from the sides as well. It's slower velocity (and lower height I guess) but coming over a wider field.
I don't think I've seen such a marking of ice to be calved over the year, even in hindsight. Very interesting to see such. Area drained in a single year. Comparisons between years. Some PhD should be doing this right now.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1274 on: February 18, 2016, 10:15:57 PM »
Quote
will not be complete without marking the ice coming in from the sides as well. It's slower velocity (and lower height I guess) but coming over a wider field. notseen such a marking of ice to be calved over the year  Area/volume drained in a single year. Comparisons between years. Some PhD should be doing this right now.
You got that right from start to finish.

The first Landsat-8 of the season just came in. It is a beauty -- though doesn't quite have coverage west enough to pick up the calving front. Notice how narrow the actual fast ice stream is -- no more than a third of total width. Needs a click to get to full 30 m resolution.

I added the Sentinel from 15 Feb 2016 that more or less covers the same region as this Landsat. It needs reprocessing to UTM23 mercator with their SNAP Toolbox to have Landsat-matching geometry.

Quote
Jakobshavn Isbrae: Velocity variations from hourly to decadal time scales at Greenland's fastest tidewater glacier
DB Podrasky UA Fairbanks dissertation, 2013, 138 pages 47.8 MB free full (data from 2006-10)
Advisor: Martin Truffer Geophysics
http://gradworks.umi.com/36/07/3607056.html

Outlet glaciers in Greenland, and elsewhere, have recently shown large variations in terminus position and ice flux. One example is the tidewater retreat of Jakobshavn Isbræ, which began in the late 1990s with high thinning rates, acceleration and collapse of the floating glacier tongue. The retreat has continued to the present, with glacier speeds more than doubling in two decades' time.

A campaign of in-situ measurements was initiated in 2006 with the aim of determining the importance of short-term forcing as a control on the continuing evolution of the glacier. Three years of continuous GPS measurements along the centerline of Jakobshavn Isbræ reveal seasonal velocity variations due to seasonally varying terminus position. The relationship between glacier speed and surface melt is complex, with both speed-up and slowdown events in response to variations in the rate of surface melt.

During a particularly long and intense melt season in 2007, a series of melt-driven slowdowns effectively reduced the mean ice flow over the whole year. On shorter timescales, the response to surface meltwater input is more predictable with diurnal velocity variations of 1--2 % that closely match changes in meltwater input.

The influence of iceberg calving and tidal forcing is restricted to the lower 10 km of the glacier, imposing an upper limit on longitudinal stress coupling length of a few ice thicknesses. The response to these forcings does not exceed 5 % of mean flow. This is consistent with a glacier operating under high driving stresses...

Ice sheet velocities as far as 120 km inland of the margin have responded to the continuing retreat with increases in speed. Flow lines in adjacent ice have rotated toward the centerline of the main channel. Ocean-driven changes have led to increased ice flux far inland on the Greenland Ice Sheet, implying a continuing large-scale evolution of the Jakobshavn Isbræ drainage basin.
« Last Edit: February 22, 2016, 12:07:34 PM by A-Team »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1275 on: February 23, 2016, 02:16:16 PM »
Jakobshavn has proven too complex to experimentally characterize or productively model so we're stuck with observation.

In the Jan 2016 paper below, just listing the dozen weakly motivated assumptions and poorly constrained experimental quantities and reviewing the ever-changing historic explanations make it clear that continuum mechanics has nothing to offer Jakobshavn at this time.

Indeed, watching the first motion animation in #1273, the disconnect between daisy world and actual ice stream is so extreme I wonder if we even want to talk about the same glacier. However it is the real Jakobshavn and its future behavior that concerns us. There was no guidance in this paper nor any testable predictions. 

Maybe that's just as well. Modelling of El Nino tele effects ran afoul of 'primum non nocere' this year, the predictions being worse than worthless, were actually harmful from a planning perspective.

Crandles and Oren ask us (#1234 #1275) to predict the Oct 2016 calving retreat line and year-on-year ice discharge isopleths. These won't be provided in the scientific literature as models cannot see even 8 months out. The discharge line has never been determined even though it is more important overall than fluctuations of the calving front.

To find all the ice that will make it out to the fjord this year requires line integrals of the seasonally-varying velocity field along flow lines -- the groundwork for that has only been laid at Zachariae by the Rignot group.

However we can estimate this for Jakobshavn by carefully watching the 2015 Landsat summer animation, ie watch a blob, see if it goes into the sea, if not follow a feature closer in on the flow line, if so follow one farther upstream, mark the limiting point, repeat on a nearby flow line, connect the dots.

This is not thrown off, like velocity, by thinning or crevasse widening towards the front and gives ice volume discharged integrating the bounded area over the bedrock DEM.

Your predicted maximal calving front retreat might be drawn over the first 2016 Landsat of the year or the 27 Feb 16 high resolution Sentinel, with one eye of course on the maximal retreat line history during the grounded era (post-2004).

The issues here are whether the retreat will be greater than the usual 1-2 km, whether the retreat line will again be lopsided (asymmetric about the central flow line), whether the north side of the south branch will have a substantive impact, and whether there will be a repeat of last August's massive event.

The latter scenario has serious potential for unwelcome outcomes. Interpreting dark as smooth, light as rough in Sentinel 1A images (first image below), these consistently show highly crevassed ice extending back from the calving front well past the elbow, eventually transitioning to two dark streams of the principal ice streams.

The question is, how deep do these crevasses go today and how far back? Looking at oblique air photos, the surface ice appears in total shambles: not orderly sheared crevasses but randomly shattered brittle ice that couldn't adapt to newly higher velocity with viscous deformation.

If this shattered ice extends deeper down and farther back than before, then the unprecedented calving event of mid-August 2015 could be a harbinger of things to come this summer.

Quote
Basal resistance for three of the largest Greenland outlet glaciers
DR Shapero IR Joughin K Poinar M Morlighem F Gillet-Chaulet
http://onlinelibrary.wiley.com/wol1/doi/10.1002/2015JF003643/full

Resistance at the ice-bed interface provides a strong control on the response of ice streams and outlet glaciers to external forcing, yet it is not observable by remote sensing. We used inverse methods constrained by satellite observations to infer the basal resistance at Jakobshavn, Kangerdlugssuaq, and Helheim [which account for 40% of excess discharge from the GrIS].

In regions of fast ice flow and high driving stresses, ice is often assumed to flow over a strong bed. We found, however, that the beds of these three glaciers provide almost no resistance under the fast-flowing trunk. Instead, resistance to flow is provided by the lateral margins and stronger beds underlying slower-moving ice upstream.

The mechanistic explanation of fast flow at Jakobshavn has changed over the years from (1) vertical shear in ice flowing over a strong bed resisting  horizontal sliding of ice above it, (2) to a transition from no-sliding to fast-sliding regime over a weak bed of a deep, narrow trough, to (3) resistance to flow from lateral margins with an inconsequental role for the downstream bed though a stronger bed underlying slower-moving ice upstream. Fast flow in Greenland has been attributed more to internal deformation, facilitated by high strain heating that softens the ice, with the beds supporting a large proportion of the driving stress.
« Last Edit: February 23, 2016, 02:39:36 PM by A-Team »

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1276 on: February 23, 2016, 10:39:05 PM »
After all the amazing animations we've seen here recently, modeling JH as an ice stream with a deep channel doesn't really cut if for me. At least visually it seems like about half the ice flowing through the elbow doesn't even come from the deep channel extending eastward, but rather from the "staging area" south of the channel. This was quite a surprise for me when I watched the animations for the first time.

Adam Ash

  • Frazil ice
  • Posts: 311
    • View Profile
    • The 100 metre line
  • Liked: 10
  • Likes Given: 23
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1277 on: February 23, 2016, 11:11:20 PM »
The bed cross sections and long-sections show significant variations in the elevation of the bedrock at the interface under the glacier from top to bottom.  These result in portions of the glacier having down to only 50% of the cross section area over the bed high spots compared with sections up and down stream. 

The velocity profiles suggest fairly constant downhill progression of the ice surface.  In order for this constant volume of flow to occur through the sections of reduced cross section, the observed (surface) velocity should almost double.  There is no evidence of that doubling.

So my hypothesis is that the glacier is actually moving with a horizontal shear line at about the level of the high spot(s), with ice above moving at the observed rate, while ice below is stalled, or even possibly flowing in an eddy with the upper surface driven down stream by friction across the shear line, and (since its an eddy) the basal ice actually moving upstream in some situations. 

The shear line could be lubricated by water intrusion and thermal stresses.

This mode of operation would provide a comparatively constant cross section for the flowing ice, which then enables the observed fairly constant mass flow and velocity to prevail.

A review of volume versus velocity versus cross section on the south branch would be telling.


A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1278 on: February 24, 2016, 03:00:21 PM »
Quote
#1279 AA The bed cross sections and long-sections show significant variations in the elevation of the bedrock at the interface under the glacier from top to bottom. These result in portions of the glacier having down to only 50% of the cross section area over the bed high spots compared with sections up and down stream.

In order for this constant volume of flow to occur through the sections of reduced cross section, the observed (surface) velocity should almost double. There is no evidence of that doubling. A review of volume versus velocity versus cross section on the south branch would be telling.
Indeed, it is hard to see how the massive amounts of incompressible ice of variable thickness coming in from the sides at variable velocity-with-depth can be smoothly accommodated into the ice stream -- and at what price of internal heating, layer shearing and ice fracturing. (Layercake isochrons are completely obliterated the last 60 km or so.)

I am at a loss to understand how a model that hangs a thin curtain of ice down from a single central Jakobshavn flowline can result in a publishable scientific product.

An east to west sequence of those vertical cross sections would make for a most excellent animation. The data is readily available from bedrock and surface DEMs though it is challenging to find an archived dataset that has not been heavily kriged, re-gridded and dumbed down to islandwide resolution from what is known far better for the Jakobshavn region.

The temptation here is to go straight to the data -- the myriad radar cross sections flown at Jakobshavn. If displayed as direct images, that would require rescaling of each frame to a common unit and tedious longitudinal cropping to a common boundary. Alternatively Cresis furnishes bedrock and surface as densely sampled cvs elevation files graphable from Excel. It seems a mess at first (image below) but the cross sections could probably be pulled by sorts on longitude.

These cross sections could be 8-bit grayscale gradients to show depth and tinted with color to show surface speed. There is no experimental information on how speed varies with depth. The tricky part is indicating velocity as direction is almost always discarded in Jakobshavn 'velocity' maps. Yet here we are primarily interested in flow convergence and layering, not generic westward movement off Greenland's central ridge.

I've not been able to locate software that can operate transitively on feature tracking across multiple frames (yielding successive points along each flowline (that could be readily fit with splines). Nor have I found software that can display a fast animation underneath a transparent layer on which I can trace the flowlines revealed by the animation. I've considered taping a clear mylar layer over the screen and scanning it afterwards, a big step backwards technologically speaking but gets a product out the door.

There is also the issue of whether these cross sections should be taken perpendicular to the fastest flow centerline or simply along meridians. The former means sections cut through each other unless the centerline is first straightened in software which only raises other issues.

The trough bottom, till, and overdeepenings present special problems, though radar does ok with onset of the walls. We have become so indoctrinated to a narrow and deep Jakobshavn channel from products with (rarely stated) 10-50x vertical exaggeration that it's important to remember that a depth of 1 km over a width of 5-6 km is not all that dramatic given the usual parabolic cross section fit.

I'll look back to see where I put the big multi-frame 2015 animations and re-post it here with a pruned flight track overlay -- the most favorable instance is a large volume of ice coming in from the southeast, not too far past the elbow. That should suffice to illustrate the issue you raise with a deficiency in area not accompanied with an increase in speed.
« Last Edit: February 24, 2016, 09:41:26 PM by A-Team »

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1279 on: February 26, 2016, 12:31:02 PM »
Looks like Jakobshavn has also been imaged from another track on 25.2 in IW-mode. Here's a screenshot from Polarview:

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1280 on: February 26, 2016, 05:30:16 PM »
Here's an annotated image from 25.2 - I circled the area on the north wall of the south branch that I do not understand. Any ideas what the darker features might be?

the arrows are pointing at radar shadows that can be used to estimate the height of the calving phase. At the north end of the north branch the shadow is up to 300 meters wide. Looking at the shadow of the main trunk it seems clear that the ice-stream is considerably thicker in the middle...what do you think is causing this?


Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1281 on: February 26, 2016, 08:17:01 PM »
I will suggest the encircled area is a "high rise bedrock"?
Have a ice day!

Laurent

  • Young ice
  • Posts: 2546
    • View Profile
  • Liked: 13
  • Likes Given: 50
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1282 on: February 26, 2016, 08:31:15 PM »
Yes, I think it is the elevation encircled in red.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1283 on: February 26, 2016, 09:46:02 PM »
Quote
Looks like Jakobshavn has also been imaged from another track on 25.2 in IW-mode.
The IW Sentinel of Feb 25th is presumably is a followup of that odd one from Feb 13th shown in #1272. The Feb 11th was from another track too, there may or may not have been a followup IW on the 23rd. The main track series may continue tomorrow.

Quote
the encircled area is a "high rise bedrock"?
The black feature is discharging but regenerating at its upper source according to the first click-requiring animation in #1273. If this were Landsat, that would fit with abraded till scraped off a local rumple. (We see that all season from the east-central region of the north branch.)

The Landsat expected for Feb 24th has not been posted. However there is that very nice oblique airplane window shot from #1240 which is shown below at larger scale for the calving front region. That can be compared with the declassified 1962 ARGON satellite image of May 1962 posted by jimb in #1229.

I re-located the season finale animation back in #1102, 27 days of Landsat, bottom animation shows the upper half.

Watching these animations, there are a good dozen stationary features in, above and around lower south Jakobshavn as discussed in the annotated images of #1151. A couple of them appear to be eddies, quiet areas behind high rises that deflect ice around themselves. There is quite a bit of topography almost exposed and with the south branch still thinning vertically, it won't be too long before some prime camping sites become available.
« Last Edit: February 26, 2016, 09:59:57 PM by A-Team »

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1284 on: February 27, 2016, 12:43:18 AM »
Yes, I think it is the elevation encircled in red.

Thanks for that awesome bedrock map. It explains much. What is the source? Is there anything covering more area around JH?

Laurent

  • Young ice
  • Posts: 2546
    • View Profile
  • Liked: 13
  • Likes Given: 50
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1285 on: February 27, 2016, 10:35:56 AM »
Sorry being not scientist I do not care much about the sources. Even if I wanted to, I would have a hard time taking track of them. That mean I don't know how to store them even if I had the info. I am not roaming somewhere else than the Neven forum so this is some materials that have been posted here.
I join an other one that is a bigger picture.

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1286 on: February 27, 2016, 10:37:25 AM »
Sorry being not scientist I do not care much about the sources. Even if I wanted to, I would have a hard time taking track of them. That mean I don't know how to store them even if I had the info. I am not roaming somewhere else than the Neven forum so this is some materials that have been posted here.
I join an other one that is a bigger picture.

Thank you! I must have been missed that or forgotten about it.

AbruptSLR

  • Multi-year ice
  • Posts: 19703
    • View Profile
  • Liked: 2268
  • Likes Given: 286
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1287 on: February 27, 2016, 10:54:26 AM »
Sorry being not scientist I do not care much about the sources. Even if I wanted to, I would have a hard time taking track of them. That mean I don't know how to store them even if I had the info. I am not roaming somewhere else than the Neven forum so this is some materials that have been posted here.
I join an other one that is a bigger picture.

Thank you! I must have been missed that or forgotten about it.

sidd posted Laurent's base map in Reply #603; however, A-Team provides many links to Morlighem's bedrock data including the following:

"I provided above goes to a page linking to an NSIDC datasheet, from there you can bypass optional registration and get to the thickness, bedrock, and elevation data ftp page and download Morlighem's 2.1 GB .nc file (NetCDF) last updated 5/19/15 for which we have found open source viewers. (Although life would be so much easier if they just posted the data in a csv text file for Excel etc. like they do for ice cores.)"

ftp://sidads.colorado.edu/DATASETS/IDBMG4_BedMachineGr/MCdataset-2015-04-27.nc

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

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1288 on: February 27, 2016, 12:31:16 PM »
Yes, I think it is the elevation encircled in red.
Not sure, as it seems to form in the middle of a fast-flowing part. Perhaps it's the 300m deep trench just west of the elevation you marked?

crandles

  • Young ice
  • Posts: 3379
    • View Profile
  • Liked: 239
  • Likes Given: 81
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1289 on: February 27, 2016, 01:32:31 PM »
Yes, I think it is the elevation encircled in red.
Not sure, as it seems to form in the middle of a fast-flowing part. Perhaps it's the 300m deep trench just west of the elevation you marked?

I agree, there is a mountain ridge casting a shadow just outside the red circle which I think is the elevation.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1290 on: February 27, 2016, 01:51:54 PM »
Quote
not care much about the sources. Even if I wanted to, I would have a hard time taking track of them.
Perhaps it's the 300m deep trench just west of the elevation you marked?
On this one, I would recommend looking at the pictures before they got 'photoshopped', just like you would do for an online dating site. Otherwise you are just taking your chances.

The maps above come from me coloring sidd's contouring of Morlighem's refinement of Gogineni's interpretation of ice penetrating radar data. The true horizontal ground resolution is not good enough relative to the size of a buried nunatak or pothole to identify everything that could really affect flow of the ice.

Furthermore AGU2016 talks put accuracy of the bottom into question with new seismic data finding up to 100 m of glacial flour built up in places -- which is a lot relative to the conceptualized bottom contours shown.

Overland tracks are actually rather far and few between as we learned trying to prove the NS dividing knob to be island. Cresis already the bottom already traced for you in red so you don't need a background in rocket science to see what the land looks like underneath the ice.

The problem all goes back to trying to cover a two dimensional surface with one dimensional flight tracks ... the plane just doesn't have the fuel or the time to fly a space-filling curve like the ones suggested by Peano in 1890.
« Last Edit: February 27, 2016, 02:08:24 PM by A-Team »

Espen

  • Young ice
  • Posts: 3705
    • View Profile
  • Liked: 420
  • Likes Given: 4
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1291 on: February 27, 2016, 05:29:23 PM »


On this one, I would recommend looking at the pictures before they got 'photoshopped', just like you would do for an online dating site. Otherwise you are just taking your chances.


Photoshopped?
Have a ice day!

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1292 on: February 29, 2016, 10:58:19 AM »
Not sure, as it seems to form in the middle of a fast-flowing part. Perhaps it's the 300m deep trench just west of the elevation you marked?
I agree, there is a mountain ridge casting a shadow just outside the red circle which I think is the elevation.
The thing I don't understand is what makes the center of the dynamic feature so dark in the SAR-image. The resolution is not high enough to make out the details...perhaps the optical images will solve this conundrum..

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1293 on: February 29, 2016, 11:44:16 AM »
Quote
what makes the center of the dynamic feature so dark in the SAR-image. The resolution is not high enough to make out the details...perhaps the optical images will solve this conundrum
The animation below, at 7.5 m resolution (2x band 8), shows the 24 Landsats available for this region (in UTM22) from day 140 of 2013 to day 251 of 2015. The animation pauses on the latter date. The scenes are rotated 35º CCW to keep file size low; the divider island is just shown on the left.

Note the flow is convergent (not parallel), with a full 90º spread from due south to due west and quite a range of surface velocities. The region under discussion appears to be a concave trough (which may partly answer your question about radar darkness).

I also looked at the dividing line separating flow into the north branch from that into the south branch. Here the dividing island acts like the horn of a saddle. The ice flows in at 45º (towards the SW), which is not the westerly flow of the overall ice sheet nor, until the last few km, that under the influence of the ice stream. Other than that, the dividing line is physically unnatural, not corresponding to a ridge, with only 5 km or so of horn 'anticipation'.

I am getting a little concerned about missing 2016 Landsats ... something may have changed for the worse in terms of orbital coverage of Jakobshavn, disrupting the time series we need for measurement of velocity in slower regions as well as for year-on-year acceleration. (If a feature moves 1 pixel (15 m) over this span of 841 days, the sensitivity is 6.5 m/yr.)
« Last Edit: February 29, 2016, 12:10:14 PM by A-Team »

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1294 on: February 29, 2016, 12:09:10 PM »
Hmm I wonder if the dark areas are places where the crevasses have been filled with random debris due to serac-falls etc. I cannot imagine that the ice-fall/stream ever becomes "smooth" but perhaps there are situations where more of the incoming energy gets absorbed.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1295 on: February 29, 2016, 12:19:09 PM »
The feature forms at a shear zone where faster ice heading west meets slower ice going south, in effect a standing wave from the reflection. This is a fairly complex situation physically when considered in 3D especially considering the topographical cross section and buttressing variations. However here as elsewhere, mother nature effortlessly solves the non-linear differential equations involved in real time at full precision.

The animation, rotated at 45º to make bulk upper velocities horizontal, shows the march to the sea in a larger context. It is a little jumpy because of geometric variations between path,row 9,11 and 10,11 orbits over the three years and the sole ground control point of the dividing island. It is also hard to match contrasts between snowy scenes and mid to late season times of maximal ablation.
« Last Edit: February 29, 2016, 12:29:58 PM by A-Team »

nukefix

  • Grease ice
  • Posts: 546
    • View Profile
  • Liked: 27
  • Likes Given: 9
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1296 on: February 29, 2016, 03:34:38 PM »
Here's a photo of the calving front and North Wall from 4.6.2015

Shared Humanity

  • Guest
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1297 on: February 29, 2016, 04:13:21 PM »
Here's a photo of the calving front and North Wall from 4.6.2015

Great photo. Thank you. I love to be able to use and connect different senses when studying something.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1298 on: February 29, 2016, 04:27:12 PM »
Same image, enlarged. It may not go quite far enough to the west to pick up the feature we've been discussing but does show some good topography nearing the surface of the ice, which can be seen to be splitting and spreading as it flows  over the hill in the right-center below.

If anyone is in the mood to determine velocities by tracking feature displacement on any of these animations, the downloaded gif provides the dates as frame names, eg 13 284 means the 284th day of 2013. There will be a unique Landsat for that date at EarthExplorer with all the rest of the metadata.

You can see that many of the published SAR velocity maps are rubbish. The ice in this area is moving in all sorts of directions at all sorts of speeds. All that information is getting tossed. Nothing good on the physics side can emerge after that.
« Last Edit: February 29, 2016, 08:52:36 PM by A-Team »

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland
« Reply #1299 on: February 29, 2016, 08:58:46 PM »
The feature forms at a shear zone where faster ice heading west meets slower ice going south, in effect a standing wave from the reflection. This is a fairly complex situation physically when considered in 3D especially considering the topographical cross section and buttressing variations. However here as elsewhere, mother nature effortlessly solves the non-linear differential equations involved in real time at full precision.

The animation, rotated at 45º to make bulk upper velocities horizontal, shows the march to the sea in a larger context. It is a little jumpy because of geometric variations between path,row 9,11 and 10,11 orbits over the three years and the sole ground control point of the dividing island. It is also hard to match contrasts between snowy scenes and mid to late season times of maximal ablation.

Your imaging abilities are omg-amazing. This is fascinating.

And this animation definitely shows a bear in profile.