Support the Arctic Sea Ice Forum and Blog

Author Topic: Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland  (Read 574556 times)

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #750 on: August 02, 2015, 12:53:47 AM »
some heavy calvings line just behind the present calving front
Yes, most odd. Landsat from yesterday does not have this yet.
« Last Edit: August 02, 2015, 01:12:45 AM by A-Team »

crandles

  • ASIF Upper Class
  • Posts: 1792
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #751 on: August 02, 2015, 12:59:15 AM »
some heavy calvings line just behind the present calving front
Yes, most odd

The gif Espen posted above seems to suggest advancement of glacier not retreat. Could 1 Aug image be taken just at start of calving event?

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #752 on: August 02, 2015, 01:00:43 AM »
some heavy calvings line just behind the present calving front
Yes, most odd

The gif Espen posted above seems to suggest advancement of glacier not retreat. Could 1 Aug image be taken just at start of calving event?

No, the DMI image is from today's Sentinel too.
Have a ice day!

crandles

  • ASIF Upper Class
  • Posts: 1792
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #753 on: August 02, 2015, 01:06:27 AM »
No, the DMI image is from today's Sentinel too.

I am probably being thick, but I don't follow, why does that rule out my suggestion?

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #754 on: August 02, 2015, 01:17:06 AM »
The Sentinel satellite only pass certain dates, and DMI is using the same data.
Have a ice day!

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #755 on: August 02, 2015, 05:13:22 AM »
You bring up a very good question, Crandles. We really need the timestamps on these images to (1) know when they were taken relative to some fixed standard so (2) that it becomes possible to subtract the timestamps to determine how much time elapsed between them (so feature velocity can be determined). Right now, we don't even know the relative order between two on the same Julien calendar date.

DMI does not provide anything but day-month-year (except for NOAA-19, upper right on DMI). They do not even provide the file name that would enable someone to look up the metadata everything is called index.php.gif  In part this may be because they download direct from the satellite (data is not encrypted) and so don't know what file name the satellite center will assign to it.

Now Sentinel file names are themselves metadata, eg S1A_EW_GRDH_1SDH_20150801 T101539_20150801 T101639_etc.tiff, which provides the date (rather than day of year) as well as start-stop times for taking of the image (a bit excessive).

Timestamps in Sentinel is a complex subject but you may be able to subtract two of them to get a time in seconds.. http://publications.jrc.ec.europa.eu/repository/bitstream/JRC92666/lbna27031enn.pdf

Landsat is more easily understood eg  SCENE_CENTER_TIME = "14:54:02.9067400Z" where the 'Z' at the end of that timestamp is interpreted as "zulu time", which means they are in UTC (Universal Time Coordinated) which for our purposes are GMT (Greenwich Mean Time).

All NOAA satellites use zulu timestamping. This would include Terra and Aqua which carry the MODIS instrument. However I don't see timestamps on those at very many sites.

The Z wraps up Alpha Bravo Charlie for fuzzy radio communication ABC's; the Navy, via the U.S. Naval Observatory, is the official US timekeeper.

I have come to prefer conventional dates like 01 Aug 15 to day 213 because of leap year issues and generally want to know the day relative to melt season knowledge. However it is slightly easier to subtract with the latter.

Best practice is to always include the file name somewhere in a post, in the posted image, or in the file name of the posted image. Then the next person can locate the original if need be. However we have a number of rogue individuals posting on these forums.
« Last Edit: August 02, 2015, 05:41:34 PM by A-Team »

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #756 on: August 02, 2015, 10:22:05 AM »
However we have a number of rogue individuals posting on these forums.

I am one of them,and I am not involved in a 100 meter dash or in science for that matter, I will just call it informatism.
Have a ice day!

crandles

  • ASIF Upper Class
  • Posts: 1792
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #757 on: August 02, 2015, 11:25:09 AM »
Looking on Modis between 1/8 and 30/7 I could not detect any differences, so I was tempted to think that whatever happened was after the Modis 1/8 pass but could of course be wrong.

If right, and a calving event takes some time (I think an hour was mentioned on thin ice video), and the latest sentinel image is after the 1/8 Modis pass then catching the calving event in progress doesn't seem impossible, though the image would have to be fairly early in the calving process.

Perhaps too many conditional 'and's in that.

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #758 on: August 02, 2015, 11:29:15 AM »
Looking on Modis between 1/8 and 30/7 I could not detect any differences, so I was tempted to think that whatever happened was after the Modis 1/8 pass but could of course be wrong.

If right, and a calving event takes some time (I think an hour was mentioned on thin ice video), and the latest sentinel image is after the 1/8 Modis pass then catching the calving event in progress doesn't seem impossible, though the image would have to be fairly early in the calving process.

Perhaps too many conditional 'and's in that.




The Modis images in that part of Greenland is sampled later in the day than the Sentinel images.
Have a ice day!

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #759 on: August 02, 2015, 05:58:39 PM »
Right, the order of imaging seems to be NOAA, NOAA, Sentinel, Aqua, Terra. DMI does not do Landsat. That ranges from daily to over a week.

Today the first NOAA came in at 02:24 UTC and the second five hours later at 07:27. These times do form part of the file names used at DMI, eg 201508020727.NOAA.jpg 

The Modis are better than nothing but are marginal for accurate resolution of small features.

It looks like Sentinel missed today up at Petermann. Unless DMI can fill in more of the image with later orbital swaths and Landsat comes through, our ability to measure mean velocities for calved bergs will be minimal. Update: DMI now has a full Sentinel ... I'll post over at the Petermann forum.

Jakobshavn flushes rather fast and even with big distinctive icebergs, tracking is more off than on. (However there is a continuous web cam for this.)

Looking now at the ESA Sentinel site, I am seeing downloadable items for 01 Aug 15 with UTC timestamps explained as T20:00:29.797Z which for us is T20:00 UTC or thirteen hours later than the second NOAA. Nothing showing at this time for August 2nd.
   
S1A_EW_GRDH_1SDH_20150801T200029_20150801T200133_007074_0099FE_94F3
https://scihub.esa.int/dhus/odata/v1/Products('cc78819a-6b96-4d8e-a794-93549686f977')/$value

Date : 2015-08-01T20:00:29.797Z, Instrument : SAR-C, Mode : EW, Satellite : Sentinel-1, Size : 1 GB

I modified this post to include the DMI Radarsat image of Jakobshavn for 03 Aug 15 ... the unusual calving continues.
« Last Edit: August 03, 2015, 01:48:20 PM by A-Team »

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #760 on: August 04, 2015, 07:20:39 PM »
some heavy calvings line just behind the present calving front
Yes, most odd. Landsat from yesterday does not have this yet.
OK if it is backing off of the under ice ridge then are we back to tabular calving?

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #761 on: August 04, 2015, 10:19:56 PM »
if it is backing off of the under ice ridge then are we back to tabular calving?
No, Jakobshavn is done with tabular calving (unless you count breakup of seasonal melange that builds during the winter).

What I expect to see as Jakobshavn retreats farther out of its fjord back into the narrower winding south channel is reduced heat exchange between warm ocean water and ice at the grounding line because the attack surface is less accessible and the water will have been cooled. However the interface is taller to the extend the ice stream is retreating over a reverse grade (ie down from a sill into a trough).

There will still be plenty of convection (heat transfer by mass transport away) via churning induced by frontal calving and upwelling buoyant meltwater, but the resulting ocean water the glacier faces is then colder than open ocean water because circulation and mixing are reduced in the narrow channel and dominated by ice. (Picture the ice stream 10, 20, or 30 km further inland and the long cul de sac of ocean water.)

The icebergs we see streaming away must be balanced by return flow of deeper waters. However surface transport is fairly shallow (tens of meters) whereas the distance through water to bedrock approaches 1500 m. Thus if this were the only consideration, retreat would slow. The narrowness of the channel could offset this somewhat through turbulent interactions with the wlls.

Jason Box just posted an excellent popular piece on the other considerations which almost all favor faster retreatL
http://www.huffingtonpost.com/jason-e-box/ice-melt-fast_b_7927186.html

plinius

  • ASIF Citizen
  • Posts: 403
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #762 on: August 04, 2015, 10:54:37 PM »
isn't there cliff failure missing in the picture too? Deeper water means higher ice front if the glacier is not to swim up...

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #763 on: August 05, 2015, 02:02:32 AM »
Yes, cliff failure. Rignot is saying now the faces of these glaciers are not flat & vertical but in some cases have had cavities hollowed out by circulating warmer water. It is not that this water is so warm but that it is warmer than freezing and that it is circulating against the ice.

If the water were stationary, it might well freeze on since the temperature profile with depth in Greenland gets to 25º and more below zero. But there is no way the glacier front can cool down any substantial part of the fjord or Baffin Bay.

The underwater portion, notably the grounding line, of the calving front of Jakobshavn Isbrae has never been observed. We do not actually know what is actually going on there in terms of meltwater exiting, sill rock or morraine sediment, currents, tides, temperature, ice rheological properties or geometry of the face.

Jakobshavn behind the grounding line is in no danger of popping up like a cork from intruding seawater. The buoyancy shortfall is graphed in Fig.2 of Joughin 2014. These papers about ice sheets rising up briefly under meltwater hydraulics are to the south and concern land-terminating glaciers.

It is all but impossible even to make similar measurements on Jakobshavn because of severe crevassing extending tens of km upstream and its very rapid motion. I personally don't see a whole lot of meltwater systems draining into JI on Landsat; some but not that noticeable, though it is hard to say where meltwater on the sides ends up ultimately exits.

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #764 on: August 05, 2015, 12:59:51 PM »
With SAR-images the geometry is so well defined that it is quite easily possible to overlay images taken from the same orbit track to the accuracy of 1/1000th of a pixel. So it is not SAR which has problems with geometry, if there are mismatches to optical imagery the problems are at the optical side  :P

That being said, with S-1 the user needs to do some geometric operations in the S-1 Toolbox (for example) in order to transform the images into a suitable map projection. If someone has questions on how to do this I can offer some help during August...once the processing graph is set up it's rather simple to process new acquisitions...

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #765 on: August 05, 2015, 03:12:18 PM »
can offer some help during August...once the processing graph is set up it's rather simple to process new acquisitions...
That would be fantastic if you could walk us through an example with a Jakobshavn file. We (me that is) definitely need to move on to the next level of processing using this Toolbox.

Be a pity if a record retreat happened in between cloudy Landsats and we weren't able to document it. In summer 2014 there were hardly any clear days there. 04 Aug 15 Terra compared to 01 below.

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #766 on: August 05, 2015, 05:57:41 PM »
if it is backing off of the under ice ridge then are we back to tabular calving?
No, Jakobshavn is done with tabular calving (unless you count breakup of seasonal melange that builds during the winter).

What I expect to see as Jakobshavn retreats farther out of its fjord back into the narrower winding south channel is reduced heat exchange between warm ocean water and ice at the grounding line because the attack surface is less accessible and the water will have been cooled. However the interface is taller to the extend the ice stream is retreating over a reverse grade (ie down from a sill into a trough).

There will still be plenty of convection (heat transfer by mass transport away) via churning induced by frontal calving and upwelling buoyant meltwater, but the resulting ocean water the glacier faces is then colder than open ocean water because circulation and mixing are reduced in the narrow channel and dominated by ice. (Picture the ice stream 10, 20, or 30 km further inland and the long cul de sac of ocean water.)

The icebergs we see streaming away must be balanced by return flow of deeper waters. However surface transport is fairly shallow (tens of meters) whereas the distance through water to bedrock approaches 1500 m. Thus if this were the only consideration, retreat would slow. The narrowness of the channel could offset this somewhat through turbulent interactions with the wlls.

Jason Box just posted an excellent popular piece on the other considerations which almost all favor faster retreatL
http://www.huffingtonpost.com/jason-e-box/ice-melt-fast_b_7927186.html
What I'm looking at is this.  Sea water is denser than fresh water.  As you cool sea water with ice you also reduce its salinity.  What wins cooler denser sea water or less dense melt diluted sea water?

What I see currently is sea water (warm) flowing along the bottom of the fjord hitting the calving front grounding line. melting ice becoming less dense and flowing out the surface.  The calving face grounding line backing away from the sill will let the sea water flow an additional 20km inland under the ice.  The water along the bottom of the fjord is pulled all the way to the calving face/grounding line.  I'm currently assuming the two to coincide.

I forget where I read that a 1 degree difference between sea water (warmer) and ice will melt about 2.5 cm a day of ice.  Add that to the 20 km of additional area when the glacier backs off of the sill and the sea water circulation in the fjord should go up.

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #767 on: August 06, 2015, 12:30:21 AM »
can offer some help during August...once the processing graph is set up it's rather simple to process new acquisitions...
That would be fantastic if you could walk us through an example with a Jakobshavn file. We (me that is) definitely need to move on to the next level of processing using this Toolbox.
Happy to help, could you start a topic in the Developers' Corner and I'll jump in? Basically it would be good if you described what are the characteristics of the end product you want (projection etc.).

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #768 on: August 08, 2015, 03:45:42 PM »
Happy to help, could you start a topic in the Developers' Corner and I'll jump in?
Great, most helpful. Let's just use the Sentinel-2 one already started so as not to spread this out all over. I would say, how do we process a Sentinel image so that it co-registers with Landsat images of Greenland which are in local Mercator projection.

https://scihub.esa.int/dhus/
http://earthexplorer.usgs.gov/
https://forum.arctic-sea-ice.net/index.php?topic=1348.0

Which way is up? 

North is not up on Sentinel images as posted, causing us some grief in trying to make a record of the Jakobshavn calving front retreat combining both Landsat and Sentinel views.

In a Landsat mercator download, latitude and longitude lines aren't provided as a mask but they form an ordinary  rectangular grid and north is straight up everywhere on the image. That's not the case in the Earth Explorer viewer which is in Google Earth projection where the lat,lon lines are curved. In Google Earth (and polar stereographic) however, the center-of-image longitude can be rotated to be due north.

Thus there can be no expectation of a single rotation providing a global north on Sentinel images (which are not in mercator) but there should be rotation that provides a local north say midway across the calving front. That would probably suffice for our purposes because residual warping would be second-order error at this scale (though re-projection would be more desirable).

The Sentinel download package does not come with any lat,lon mask that would allow this local rotation to be determined (from a tangent line). However these images do provide quite recognizable traces of coastlines etc so in fact these lat,lon lines would be curved but not pathologically so (ie still smoothly continuous).

I asked the help desk about where in the metadata might I find the precise local anglse of rotation and received this detailed response from eosupport at copernicus.esa.int which I am still digesting:

Sentinel-1 imagery is provided in satellite geometry which is slightly different depending on the type of Level 1 product. For the detected GRD products, the imagery is provided in ground range, x, and azimuth, y, geometry whereas for complex SLC products, the imagery is provided in slant range, x, and azimuth, y, geometry.

This is described at https://sentinel.esa.int/web/sentinel/user-guides/sentinel-1-sar/product-types-processing-levels/level-1. As both GRD and SLC imagery are provided in satellite geometry, its orientation with respect to true north will vary with latitude and incidence angle. Thus it is not possible to accurately orientate the imagery with respect to north using a single rotation as this depends on both the x (range) and y (azimuth) position within the image. [[This could be provided by a simple grayscale mask but isn't.]]

Although Sentinel-1 imagery is not provided in a geocoded orientation (like your Landsat image), the product annotation gives a complete description of the transformation from pixel coordinates (x,y) to latitude & longitude via the geo-location data set record.

This data set is described in the Sentinel-1 Product Specification document S1-RS-MDA-52-7441 available from https://sentinel.esa.int/documents/247904/349449/Sentinel-1+Product+Specification+3.0. In particular see Section 6.3.1.7 for a description of the geolocation grid. Below is an example of first point within a grid from a GRD product:

<geolocationGrid>
<geolocationGridPointList count="483">
<geolocationGridPoint>
<azimuthTime>2015-08-04T22:03:55.263139</azimuthTime>
<slantRangeTime>4.976224778237438e-03</slantRangeTime>
<line>0</line>
<pixel>0</pixel>
<latitude>7.768200950286953e+01</latitude>
<longitude>-9.014424336694947e+01</longitude>
<height>-2.077938625589013e-01</height>
<incidenceAngle>1.937838349149362e+01</incidenceAngle>
<elevationAngle>1.737100868906543e+01</elevationAngle>
</geolocationGridPoint>

Note that an approximate orientation of the image with respect to north can be found by extracting the platformHeading parameter within the product annotation (see Table 6-31 of the above product specification document). Note that this is the orientation of the satellite and not the image. An example of this parameter is given for an image at a high latitude:

<generalAnnotation>
<productInformation>
<pass>Ascending</pass>
<timelinessCategory>Fast-24h</timelinessCategory>
<platformHeading>-4.390998067621028e+01</platformHeading>

This seems to say rotate a high latitude (Greenland) ascending orbit image by 43.9º clockwise to get north more or less up. I would need to look up this parameter in a bunch of Jakobshavn, Petermann and Zachariae to see if it varies much. Offhand, it seems too large for the Jakobshavn image we were considering a few posts above.

It is difficult to find pairs of well-separated, consistently identifiable rocks where the angles can be measured on both Sentinel and Landsat, which would sidestep rooting around in Sentinel metadata. That has to do with snow cover and sun angle in Landsat vs blurry resolution and rock shadowing on Sentinel.
« Last Edit: August 08, 2015, 03:58:50 PM by A-Team »

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #769 on: August 08, 2015, 04:28:34 PM »
Here is the DMI Sentinel from August 8th. It is rotate to the north by an undocumented procedure. It may also be re-projected to polar stereographic, it's a great site but details are sketchy, other than Sentinel-1 satellite are processed by DTU (Technical University of Denmark). http://www.seaice.dk/ has some interesting daily mosaics of the cryosphere.

www.dtu.dk/english

Looks like 'something' is happening on the calving front. There is a path,row 9,11 Landsat from today but the EarthExplorer site is barely working again today and the image is hopelessly clouded over. I am chasing down the full resolution Sentinel to see if that helps.
« Last Edit: August 08, 2015, 04:45:20 PM by A-Team »

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #770 on: August 08, 2015, 05:18:07 PM »
It is in a gaining mood
Have a ice day!

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #771 on: August 08, 2015, 05:50:24 PM »
Image rotated by 15º gives quite a good local fit (to Landsat) using that lake in the lower left and the ice fall dividing north and south branches. Looks like a very large piece has calved or is about to.
« Last Edit: August 08, 2015, 06:09:46 PM by A-Team »

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #772 on: August 08, 2015, 06:16:14 PM »
Image rotated by 15º gives quite a good local fit (to Landsat) using that lake in the lower left and the ice fall dividing north and south branches. Looks like a very large piece has calved or is about to.

Yes a very large piece seems to be sent of soon.
Have a ice day!

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #773 on: August 08, 2015, 06:17:53 PM »
What is the projection of Landsat exactly? UTM zone something?

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #774 on: August 09, 2015, 06:41:37 AM »
Yes. Landsat-8 uses Universal Transverse Mercator (UTM) map projection for everything except south of -63º latitude (ie Antarctica) where Polar Stereographic is used, both relative to World Geodetic System 84 datum (WGS84).

Petermann is UTM_ZONE = 20, less commonly 21 as per the metadata file ending in MTL.txt in each package download 
 
Jakobshavn is UTM_ZONE = 22, less commonly 23

Zachariae is UTM_ZONE = 27 for the area Wipneus and Espen have been looking at

Sptizbergen, Jan Mayen, and Iceland we don't look at too often.
« Last Edit: August 09, 2015, 07:04:46 AM by A-Team »

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #775 on: August 09, 2015, 11:23:25 AM »
This is off-topic here, I'll start a new thread in the Developers' Corner.

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #776 on: August 10, 2015, 12:07:03 AM »
if it is backing off of the under ice ridge then are we back to tabular calving?
No, Jakobshavn is done with tabular calving (unless you count breakup of seasonal melange that builds during the winter).

What I expect to see as Jakobshavn retreats farther out of its fjord back into the narrower winding south channel is reduced heat exchange between warm ocean water and ice at the grounding line because the attack surface is less accessible and the water will have been cooled. However the interface is taller to the extend the ice stream is retreating over a reverse grade (ie down from a sill into a trough).

There will still be plenty of convection (heat transfer by mass transport away) via churning induced by frontal calving and upwelling buoyant meltwater, but the resulting ocean water the glacier faces is then colder than open ocean water because circulation and mixing are reduced in the narrow channel and dominated by ice. (Picture the ice stream 10, 20, or 30 km further inland and the long cul de sac of ocean water.)

The icebergs we see streaming away must be balanced by return flow of deeper waters. However surface transport is fairly shallow (tens of meters) whereas the distance through water to bedrock approaches 1500 m. Thus if this were the only consideration, retreat would slow. The narrowness of the channel could offset this somewhat through turbulent interactions with the wlls.

Jason Box just posted an excellent popular piece on the other considerations which almost all favor faster retreatL
http://www.huffingtonpost.com/jason-e-box/ice-melt-fast_b_7927186.html
https://www.google.com/maps/place/Bosphorus,+Turkey/@41.1194435,29.075278,2200m/data=!3m2!1e3!4b1!4m2!3m1!1s0x14cacaf6a1b454cf:0x7bab8b9dc19261dc

I'm thinking about this place.  Google has it at 16 miles approximately.  The freshwater entering the black sea mixes with the salt water from the Mediterranean this carries the salt water out to sea.  The denser water flows into the black sea at quite a fast speed.  The more fresh water you add the higher the circulation.

Two processes.  One retreat the other thinning.  Both would lead to large scale sea water circulation under the next 20 km of the glacier as the glacier looses contact with the sill. More ice area exposed to sea water gets more melt water, gets more circulation.

Adding heat under the next 20 km of glacier should get it to speed up!

Speeding up gets you thinning and or retreat.

Retreat opens up the sides to attack by sea water.  It also reduces buttressing of the side inflows.  That gets more ice flow.  More ice area exposed to sea water gets you more melt water and more circulation.  (I've driven commercial vehicles.  Go a bit to fast downhill and the brakes over heat.  Smoke, fire and bent metal.  Just a bit too fast is all it takes.)

There are plenty of positive feedback loops.

Are there any negative ones?  Is the system dampened? What keeps the overheated brakes from failing?



A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #777 on: August 10, 2015, 07:39:07 PM »
The analogy to a truck is a good one. Where I live, the driver encounters a long 7% grade coming over the pass so the state built two emergency graveled ramps for runaway vehicles. After reaching a small town, the gradient drops to 0.5%, not enough to coast down to the valley, no escape ramps.

Greenland is almost always displayed at extreme vertical scale exaggeration. I used to think wow 10,000 feet from summit to sea level, no wonder Jakobshavn is like a truck barreling over the pass. Later I realized that it is so big and flat that if a truck were set down there, the driver would have no idea how to drive down to the sea.

For a while there, I was posting radar transects at true 1:1 scale but they took up so many horizontal feet of monitor space and got so few views that, like everyone else, I went back to vertical scale exaggeration.

The thing to remember with Jakobshavn is the main ice stream could care less about erosion at the calving front/ocean front that just amounts to bugs hitting the windshield. It is driven by gravitational forces, in proportion to the gradient, and braked by stiff rheology and friction with the sides and bedrock.

However this doesn't explain -- given so many other Greenland glaciers originating off the same summit ridge -- why Jakobshavn so much faster and its trough gouged so much deeper all the way to the interior, with offshore moraines showing it has been exceptional throughout the Pleistocene.

Is Jakobshavn potentially a runaway truck?

The short answer is no one knows because the very fact that it is so deep, fast, and crevassed means it has been impossible to get temperature profiles and basal hydraulics by drilling, see the bedrock very well with radar, measure meltwater inputs, or approach the bathymetry and grounding line with sonar. Nearby glaciers such as Rink, Store and Epiq are much better understood, even though there is some irony in their lesser contribution to sea level rise. See post #215 and #217 on the Hansen article forum.

Somewhere in the many Joughin papers and press commentary on Jakobshavn, I recall him giving good reasons why it is never going to get more than 3x its current velocity. I'll take another look at the assumptions underlying that in view of the Rignot 2015 results to see if those are still valid in the face of future surges in basal meltwater plumes coming out at the calving front.

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #778 on: August 10, 2015, 10:11:45 PM »
The ice flows have been doubling every 5 years for how long?  The main flow doesn't have to go much faster than it currently is to keep up the doubling rate as long as the calving face keep retreating.

The friction of ice in water is approximately zero at zero speed.  The melt water driven surface current is flowing past the calved ice out to sea.  The sea water path inland is very big.   Fast shallow water on top slow deep water going the other direction.  We could see the speed of the calved ice flow measured in km per hour not m per day.

The big question I have is how far inland the calving face will retreat.  It could be stable on the sill for 5 to 10 years before retreating farther inland.  Then make up for lost time very quickly.  As it retreats the grade goes up.  As it thins the grade goes up, and the base friction goes down as it floats off of bed rock.

Up stream a ways it branches into three branches.  The shallowest and narrowest is the fastest flowing.  Why?  My read is steepest grade from summit to main channel and it was the smallest and so experienced the fastest acceleration when the flow went up last at the end of the last ice age.  A lot of the main channel currently has or should have melt water on the bottom and so is floating the ice.

Thank you for going back and looking at the assumptions driving the 3X the current limit.

Lots of positive feed back loops, I don't see a runaway truck lane.

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #779 on: August 10, 2015, 10:44:15 PM »
Thinning slows flow.  Negative feed back. :-\

http://www.the-cryosphere.net/8/209/2014/tc-8-209-2014.pdf


A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #780 on: August 10, 2015, 11:15:39 PM »
A lot of papers over the years ... the future of Jakobshavn seems to be word salad. It would be better if people would simply lay out their predictions in a format that could be evaluated for prediction skill as time goes by, starting with peak retreat predictions for 2015, 2016, 2017. But that is studiously avoided by everyone in the field.

A Joughin interview online finds him more or less in synch with IPCC on the Hansen matter, with 10' rise this century viewed as exceedingly remote. Yet he has a 2014 paper on the irrevocable collapse potentially underway for Thwaites.

http://www.sciencemag.org/content/344/6185/735

http://news.nationalgeographic.com/2015/07/150721-james-hansen-sea-level-rise-climate-change-global-warming-science/

Further summer speedup of Jakobshavn Isbræ
I. Joughin and B. E. Smith
http://www.the-cryosphere-discuss.net/7/5461/2013/

The transient summer speeds we observe for 2012 (>17000myr−1) appear to represent the fastest observed speed for any outlet glacier or ice stream in Greenland or Antarctica. This yields a speed up by more than a factor of four relative to the 1990s, while the mean annual speedup is by just under a factor of 3. If, as the glacier recedes up the trough, it is able to maintain the peak speeds year round, then a sustained speedup by a factor of 4 of 5 is conceivable based on recent behavior, which is about half of the ad hoc tenfold upper limit on speed proposed by Pfeffer et al. (2008). Nevertheless, these speeds would occur in a trough roughly twice as deep as prior to the speedup.

Hence, a tenfold increase in ice flux may be possible for Jakobshavn Isbræ if the trough does not narrow substantially with distance upstream. Equivalently, while the increase in terminus speed and the glaciers overall maximum speed may remain under a factor of five, as the terminus retreats farther inland where the speeds now are comparatively slow, the relative speedup is much greater (e.g., if the terminus retreated to M26 with a speed of 16000myr−1, this would represent a twelve-fold speedup). Thinning by hundreds of meters to a terminus near flotation, however, yields something closer to a ten-fold flux increase. It is unlikely that such retreat could be sustained for more than a few decades because the terminus would rapidly retreat ∼ 50 km to shallower depths (Joughin et al., 2012).

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #781 on: August 11, 2015, 02:18:37 AM »
Here is the Radarsat-2 image for today, a curious X at the calving front and seemingly a melting fringe. It is a proprietary Canadian satellite with vaguely similar capabilities to Sentinel 1A. I don't believe there is public access to higher resolution than what is provided below by DMI.

Radarsat swings by Jakobshavn at irregular intervals from a few days to a week. I tried animating the last half-dozen images to make some sense out of today but the geometry varies greatly with no real way of warping them onto fixed rock anchors. There is no Sentinel or Landsat for today ... so what you see is what we get.

crandles

  • ASIF Upper Class
  • Posts: 1792
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #782 on: August 11, 2015, 02:59:59 AM »
Looks like a really hard to read draft mark made to be really difficult for a bot to read.

I imagine I can make out

C386Dd3H3CCH
D7723176521515PEC52M
___________________X

The x being on the calving front with the M immediately above. Besides all that on the glacier, there are other areas that seem to have hints of lettering/numbers as well. Quite bizarre.


A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #783 on: August 11, 2015, 05:07:14 AM »
Quite bizarre.
Attached is a cut-out of the original DMI posting (inset) and a 2x bilinear enlargement. I cannot see the letters myself but your interpretation of an extensive watermark ruining the image makes sense.

Launched in December 2007, Canada's next-generation commercial radar satellite offers powerful technical advancements that will enhance marine surveillance, ice monitoring, disaster management, environmental monitoring, resource management and mapping in Canada and around the world.

This project represents a unique collaboration between government and industry. MacDonald, Dettwiler and Associates Ltd. (MDA) will own and operate the satellite and ground segment. The Canadian Space Agency (CSA) helps fund the construction and launch of the satellite and will recover this investment through the supply of RADARSAT-2 data to the Government of Canada during the lifetime of the mission.

My grasp of higher economic theory is rather minimal but how can the govt of Canada 'recover its investment' by buying and selling to itself? Meanwhile it seems that MDA has totally privatized a public satellite. I've heard that Radarsat just orbits idly (not taking images) until someone pays for a scene order. The Canadian govt may have a standing order -- indeed the only order -- and share these with DMI  for iceberg monitoring and marine safety[?] Again, I will have to defer to someone with more insight into private enterprise to explain how this can make sense to the Canadian taxpayer.

MacDonald, Dettwiler and Associates Ltd. (TSX: MDA) is a Richmond, British Columbia global communications and information company. MDA has locations throughout Canada, and the United States operating under the MDA brand name.MDA provides operational solutions to commercial and government organizations worldwide, including:

    Airborne Surveillance Solutions
    Satellite Ground station
    Maritime Information Systems
    Aviation Information Systems
    Geospatial information services, including Satellite Data Distribution, Value-Added Information Services
    Robotic surgery research via its NeuroArm development program
    Space robotics, satellite information and payload systems
    Radar and optical satellite imagery, and, remote sensing

Rubikscube

  • ASIF Citizen
  • Posts: 247
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #784 on: August 11, 2015, 12:32:03 PM »
What I still don't get about Jakobshavn is why there is so little interest in the areas east of the currently active main channel. Morlighem shows there are multiple channels hundreds of meters deep stretching for hundreds of kilometers inland, even though those areas appear inactive today there must surely be possible to extract some useful information about the past and future behavior of Jakobshavn by taking a closer look at them. I mean, how can scientist make an even remotely accurate estimate of JHs future contributions to SLR if something like 90% (probably more) of its drainage is ignored? Have they virtually just given up on JH because the main channel more or less inaccessible?

nukefix

  • ASIF Citizen
  • Posts: 407
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #785 on: August 11, 2015, 01:55:46 PM »
S-1 IW on 7.8.2015 in Landsat projection (UTM zone 22 + 15m pixel size)

ablair

  • ASIF Lurker
  • Posts: 5
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #786 on: August 11, 2015, 03:30:45 PM »
Nice shot, if you zoom into "cap ice" there seems to be a promontory /embayment forming on it's southern shore just past the corner

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #787 on: August 11, 2015, 04:04:20 PM »
Nice work, nukefx. I received some additional help from Copernicus this morning and dug into Landsat geocoding a bit,  both described on your new Sentinel forum. The pipe dream here is gesticulating at Landsat and Sentinel urls of the same date and having a script co-register them at 15 m pixels, 16-bit files cropped to specified lat,lon corners for viewing as an ImageJ stack. Interesting concept: pan-sharpening of Sentinel.

http://forum.arctic-sea-ice.net/index.php/topic,1355.msg60306.html#msg60306

"To complete our previous message, the user should look at the link below. This even describes how to superimposes a Landsat and S1A image. The blog also describes the ESA tool SNAP (http://step.esa.int/main/toolboxes/snap/)."

https://scientiaplusconscientia.wordpress.com/2015/08/06/working-with-sentinel-1-data-pre-processing-georeferencing-and-exporting-with-snap/


A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #788 on: August 11, 2015, 04:16:48 PM »
Nice shot, if you zoom into "cap ice" there seems to be a promontory /embayment forming on it's southern shore just past the corner
I'm thinking you mean the area indicated with the white arrows below. That would be quite interesting to track over the season and indeed the three Landsat-8 years we have now. It may be a new development enabled by buttressing withdrawal of the main ice stream.

http://forum.arctic-sea-ice.net/index.php/topic,154.msg56631.html#msg56631

ablair

  • ASIF Lurker
  • Posts: 5
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #789 on: August 11, 2015, 04:37:23 PM »
Very interesting there too but I meant right near the point, further West

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #790 on: August 11, 2015, 04:44:32 PM »
What I still don't get about Jakobshavn is why there is so little interest in the areas east of the currently active main channel... how future contributions to SLR if something like 90% (probably more) of its drainage is ignored? Have they given up because the main channel more or less inaccessible?
From watching AGU meeting abstracts and posters from Ph.D dissertations that never seem to emerge as journal articles, I'd guess quite a few approaches haven't panned out.

It is mildly disturbing to see JI research lurch from one trending topic to another. For a while it was all about melange buttressing, then ablation and dynamic thinning, frictional softening on the sides, meltwater moulins and basal lubrication, sills and troughs, and today emergent meltwater plumes entraining warm Atlantic Water that erodes ice at the grounding line.

These mostly fall into 'it's the plug at the bottom' rather than 'it's less resistance to upper driving stress'.

One interesting observational program on upper Jakobshavn has been the PARCA stakes (acronym works in forum search) at about 2000 m. People are also watching the march of melt ponds upslope.

One bit of research that forum folks could do is compare ice volume above and below 2000 m (using the ice thickness DEM). I would guess that melting all of Greenland below 2000 m (leaving, impossibly, a vertical cliff) would not have that much effect on SLR. Nothing much happens without mobilization of higher elevation ice.

The upper velocity field is also regularly measured with SAR for acceleration. However subtracting small numbers in slow-moving regions gives higher uncertainty. Work-in-progress using 16-bit Landsat and Sentinel may improve this situation markedly.

It might be feasible to drill just down from the triple tributary junction. Drill holes never end up truly vertical even on the flats so some motion can be accommodated. The key would be new rapid drilling techniques. They don't need to recover a core necessarily, just get temperature, tilt, basal hydraulics and so forth along the lines of what Luethi has been doing at Swiss Camp.

Tor Bejnar

  • ASIF Upper Class
  • Posts: 1515
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #791 on: August 11, 2015, 05:39:13 PM »
Melting all ice below 2000 m (and nothing about it) will create (impossible) floating ice mountains, touching the ground only where the bedrock extends above 2000 m (on the east coast). ;D (Sorry about that.)
Arctic ice is healthy for children and other living things.

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #792 on: August 11, 2015, 06:15:54 PM »

Further summer speedup of Jakobshavn Isbræ
I. Joughin and B. E. Smith
http://www.the-cryosphere-discuss.net/7/5461/2013/

The transient summer speeds we observe for 2012 (>17000myr−1) appear to represent the fastest observed speed for any outlet glacier or ice stream in Greenland or Antarctica. This yields a speed up by more than a factor of four relative to the 1990s, while the mean annual speedup is by just under a factor of 3. If, as the glacier recedes up the trough, it is able to maintain the peak speeds year round, then a sustained speedup by a factor of 4 of 5 is conceivable based on recent behavior, which is about half of the ad hoc tenfold upper limit on speed proposed by Pfeffer et al. (2008). Nevertheless, these speeds would occur in a trough roughly twice as deep as prior to the speedup.

Hence, a tenfold increase in ice flux may be possible for Jakobshavn Isbræ if the trough does not narrow substantially with distance upstream. Equivalently, while the increase in terminus speed and the glaciers overall maximum speed may remain under a factor of five, as the terminus retreats farther inland where the speeds now are comparatively slow, the relative speedup is much greater (e.g., if the terminus retreated to M26 with a speed of 16000myr−1, this would represent a twelve-fold speedup). Thinning by hundreds of meters to a terminus near flotation, however, yields something closer to a ten-fold flux increase. It is unlikely that such retreat could be sustained for more than a few decades because the terminus would rapidly retreat ∼ 50 km to shallower depths (Joughin et al., 2012).

 "the terminus would rapidly retreat ∼ 50 km to shallower depths"

60 km upstream there is a three way split in the main channel.  Jakobshovn is flowing the shallowest narrowest channel.  Get the other two flowing and then we would have fun.

If it keeps its current retreat average of 1 mile a year up.  Making up lost time for getting off of the sill.  It could be there in 40 years.

It is definitely politically incorrect to predict large sea level rises soon.  But is that what is going to happen?  Joughin said that just looking at doubling times is less than entirely accurate... But what else is there for Jakobshavn?

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #793 on: August 12, 2015, 01:20:38 AM »
60 km upstream there is a three way split in the main channel.  Jakobshovn is flowing the shallowest narrowest channel.
It's one thing to locate this area on the bed topography, ice thickness, surface elevation maps ... another to co-locate it accurately on surface velocity, Landsat and Sentinel maps. I've posted extensively on the confluence area before ...
http://sites.uci.edu/morlighem/dataproducts/mass-conservation-dataset/

Melting all ice below 2000 m (and nothing about it) will create (impossible) floating ice mountains
Not talking about that. Find the 2000 m contour. Dig a trench straight down to bedrock with your backhoe. Now measure the ice volume downhill. Then measure the ice volume uphill. Divide. Get small number (?)

Or, for a more refined graph of cumulative ice volume vs elevation, select each key color below, count the pixels lit up in the map, get a point for your plot. (Elevation contour: sum thickness of ice for each pixel at that contour = volume of ice at that contour.) Label the other axis for mm of sea level rise if everything at that contour and below melted.
« Last Edit: August 12, 2015, 01:25:50 AM by A-Team »

Rubikscube

  • ASIF Citizen
  • Posts: 247
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #794 on: August 13, 2015, 12:12:01 AM »
What could have been a walk in the park becomes a whole lot messier when every thickness map within reach features a myriad of different color contours rather than a nice and tidy little color palette. Or is it just me looking at the wrong places? Well fortunately, when taking a closer look there turned out to be more system in this color anarchy then what seemed at first ("anarchies" often work that way), so it was eventually possible to narrow things down to a handful of specific contours, each representing a thickness.

However, there are problems with this approach, most notably; the pure colors (red, yellow, cyan and blue in this instance) merged with and were eaten by their less pure neighbours so to speak. This effect becomes particularly visible at low elevations where the slope is steeper and the different colors are crammed tighter together. Then, for some reason (I'm tempted to say god knows why, but I'm not sure he does this time) there turns out to be 36 different contours, for one 32 would have fitted perfectly into their color system, as opposed to the current solution which tries to divide 256 on 9, but more importantly it would allow each contour to represent a leap of exactly 100 meters, as opposed to 88,1m which I got from pixel counting their color scale (which by the way measures 241, a prime). Is this because 88 meter fits some margin of error or something like that, or do they just want to mess up the diagrams of excel noobies like me?

Either way I have managed to produce a couple of diagrams that resembles those who were requested, although it is not volume vs elevation, which I agree would be better, but rather volume vs thickness. I'll do the elevation as well, but if someone happens to jump across such a map that doesn't feature every possible RGB combination, that would be helpful and it would enable me to do a bit more than just setting a crude line at approximately 2000m (the same goes for the thickness map of course). Note that I have been treating the individual coutures as ranges which means the volume calculations are based on the medium of these ranges, please let me know if I've gotten that wrong. In this calculation the 2000 meter boundary splits the volume 35/65, give or take a percentage or two.

Click to enlarge.
« Last Edit: August 13, 2015, 12:21:20 AM by Rubikscube »

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #795 on: August 13, 2015, 12:54:40 AM »
Mighty fine effort, Rcube.

Yes indeed, one of the great unsolved mysteries in glaciology is why so few of them have a clue how to post maps keyed in accord with elementary principles of GIS. Howat seems to be the only one familiar with more useful grayscale.

Here the link 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

Much of interior Greenland is below sea level but not by much on average. This means that ice thickness is approximately the same as elevation. I located the 2000 m contour on your graphic as well as Csatho's 2000 m boundary for the PARCA stake flow survey. About 38% of the ice volume lies seaward of an imaginary curtain hanging down from the 2000 m contour.

If this melted, it would produce about 2.8 m of sea level rise, rather respectable. Right now the GrIS is contributing about 0.007 m/yr.

Of course this is not at all how the ice sheet actually melts and flows. Although that massive late-season rain event of 2011 did manage to lift up the whole lower west central ice sheet for a while. There is nothing worse than a rain event at a time when the basal drainage conduits have re-frozen for the year -- and these events are expected to become more common.

Amplified melt and flow of the Greenland ice sheet driven by late-summer cyclonic rainfall
SH Doyle et al
http://www.sciencedaily.com/releases/2015/07/150713113447.htm

Intense rainfall events significantly affect Alpine and Alaskan glaciers through enhanced melting, ice-flow acceleration and subglacial sediment erosion, yet their impact on the Greenland ice sheet has not been assessed. Here we present measurements of ice velocity, subglacial water pressure and meteorological variables from the western margin of the Greenland ice sheet during a week of warm, wet cyclonic weather in late August and early September 2011.

We find that extreme surface runoff from melt and rainfall led to a widespread acceleration in ice flow that extended 140 km into the ice-sheet interior. We suggest that the late-season timing was critical in promoting rapid runoff across an extensive bare ice surface that overwhelmed a subglacial hydrological system in transition to a less-efficient winter mode. Reanalysis data reveal that similar cyclonic weather conditions prevailed across southern and western Greenland during this time, and we observe a corresponding ice-flow response at all land- and marine-terminating glaciers in these regions for which data are available.

Given that the advection of warm, moist air masses and rainfall over Greenland is expected to become more frequent in the coming decades, our findings portend a previously unforeseen vulnerability of the Greenland ice sheet to climate change.
« Last Edit: August 13, 2015, 03:00:10 PM by A-Team »

A-Team

  • ASIF Upper Class
  • Posts: 1911
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #796 on: August 13, 2015, 03:08:16 PM »
Here is the low resolution Sentinel from DMI ... looks interesting, be good to pursue this with Nukefx's protocol for getting it geo-oriented. A Landsat is due tomorrow and then again two days later (16th).

Offhand, I would say the calving front retreat remains stalled. Jakobshavn may have reached 'confinement' status, ie a near steady-state range balancing glacial advance and ocean augmentation of calving, just like its neighbors Store and Rink which have not retreated since the 1940's.

http://forum.arctic-sea-ice.net/index.php/topic,1327.msg60183.html#msg60183
http://forum.arctic-sea-ice.net/index.php/topic,1327.msg60197.html#msg60197

The fourth image shows that the calving line today is not up to what it was on Day 212, though it has some interesting pre-calving features.
« Last Edit: August 13, 2015, 07:14:24 PM by A-Team »

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #797 on: August 13, 2015, 05:39:25 PM »


However, there are problems with this approach, most notably; the pure colors (red, yellow, cyan and blue in this instance) merged with and were eaten by their less pure neighbours so to speak. This effect becomes particularly visible at low elevations where the slope is steeper and the different colors are crammed tighter together. Then, for some reason (I'm tempted to say god knows why, but I'm not sure he does this time) there turns out to be 36 different contours, for one 32 would have fitted perfectly into their color system, as opposed to the current solution which tries to divide 256 on 9, but more importantly it would allow each contour to represent a leap of exactly 100 meters, as opposed to 88,1m which I got from pixel counting their color scale (which by the way measures 241, a prime). Is this because 88 meter fits some margin of error or something like that, or do they just want to mess up the diagrams of excel noobies like me?




Click to enlarge.

Try exactly 100 yards instead of meters.

Espen

  • ASIF Governor
  • Posts: 3078
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #798 on: August 13, 2015, 07:38:02 PM »
Just adding to what mentioned above:
Have a ice day!

Iceismylife

  • ASIF Citizen
  • Posts: 182
    • View Profile
Re: Jakobshavn Isbræ / Sermeq Kujalleq / Ilulissat Icefjord
« Reply #799 on: August 14, 2015, 07:03:06 PM »
Thank you A team for the lovely reference on ice thickness etc.

Looking at ice thickness up Jakoshavn and assuming an average retreat of 1.5 km or 1 mile a year for the next 50 years, it looks like Jakobshavn could raise sea level world wide by 150 to 300 mm.  (6 inches to 1 foot)  That would be by 2065.

What does the IPCC say the sea level rise for the world should be by then?