Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland

[Adam Ash]

...
I wonder where the icebergs that Jakobshaven calves end up..

In the mail room of the Titanic, apparently!

[nukefix]

I like the DEM from I Howat at UOhio/UMinn. They are made from stereo pairs of sub-1m horizontal resolution of commercial WorldView visible images.and have about 2m vertical resolution. I'm recalling they come as 16-bit grayscale with a clever secondary layer that tells where each pixel came from.

We'd need it in a GeoTIFF and in WGS84. 10m resolution would be plenty already for co-locating images in SNAP.

[Cate]

nukefix, here in Newfoundland, we're told that 99% of the icebergs that float past us on the Labrador Current come from "glaciers in Western Greenland", so there's good chance bits of Jakobshavn sail past our NE coast....So yes, they end up in the mail-room of the Titanic, or they melt away on the Grand Banks, or they get grounded, say, off Fogo, or in the harbour at  Leading Tickles, Notre Dame Bay.

Remember the Petermann ice island from a few years ago?

http://www.accuweather.com/en/weather-blogs/anderson/the-travels-of-the-petermann-ice-island-1/55336

[johnm33]

Apart from one decent sized berg a few days ago and a little too-ing and froo-ing very little action here for weeks. High tide at about 19:00 yesterday changed that https://www.lookr.com/lookout/1310041325-Ilulissat

[Wipneus]

 2016/7/14 ( R111) is in, 0% clouds according to metadata. I have uploaded S2A_R111_V20160714T152910.22WEB.B04.tiff

[Espen]

Jakobshavn Isbræ movement update:

[johnm33]

The winds [and tides] over the last few days have brought a lot of warm water north, with the build up of tides and consequential change of the waters in the fjord we should see some major movents by about the 22^nd.  That's to say the peak tide is on the 21^st but peak total flow on the 22^nd easing off slowly for 2 days and then more rapidly.

[oren]

I think I should have phrased my question about the calving front differently.

I'm pretty interested in how much area has been gobbled up along the north wall of the south branch -- JI is far wider, now, isn't it? 

And the same is true of the north branch -- it used to be very noticeably narrower than the south branch.

As to the north wall, it seems JH almost has a new north branch. It is becoming nicely well defined as can be seen in Espen's animation.
And I also get the unconfirmed feeling that JH calving front used to be more symmetric relative to flow direction, but is now slightly skewed (right side is more in retreat) due to loss of buttressing. I tried to find visual proof but failed with my limited resources.

[Wipneus]

2016/7/18 ( R025) is in. Metadata: 14% cloud cover, some over the calving front.
I have uploaded S2A_R025_V20160718T150915.22WEB.B04.tiff

[A-Team]

Some clouds but clear enough for the calving front straightaway to be used in velocity measurements. Its position is unremarkable.

The first image shows three nicebergs at 2.5 m. These possibly represent a near-complete top to bottom column at the calving front.

The animation shows the action on the north branch over the 50 day interval considered. Some regions are moving at a good rate; others have much slower velocities. Could this be the year that the calving front actually retreats or changes character?

Crevasses in the left center seem to be enlarging, these are shown at 3.3 m side-by-side below. Images are shown as negatives (255-pxl value).

S2A_R025_V2016 07 18 T150915.22WEB.B04
S2A_R025_V2016 05 29 T150918.22WEB.B04

[Wipneus]

2016/7/21 ( R068) arrived, metadata says 0.03% clouds. I have uploaded S2A_R068_V20160721T151913.22WEB.B04.tiff

[A-Team]

That image is a beauty; the google drive collection conveniently lists the other R068 Band 4 for animation purposes. Though here again color offered advantages: it appear there is widespread pink snow. That adsorption would be unfavorable for albedo.

The north branch has been producing a lot of bergs and appears to have developed a second plume of rock scour till. The calving frontal area has cleared up but not retreated.

South branch looks like it is setting up for a central calving though predictions haven't worked out notably before. It is fairly far forward in position.

Although the S2A instrument is producing fabulous initial images, the embedded histogram suggests segmentation techniques could be used on the image so that each peak can be optimally -- and separately -- adjusted for contrast. (ImageJ has a couple dozen ways of doing this.)

[wiki: discovered in NW Greenland] In May 1818, four ships sailed from England to search for the Northwest Passage and chart the Arctic coastline of North America. Severe weather made them finally turn the ships back, but the expedition made valuable contributions to science. Captain John Ross noticed crimson snow that streaked the white cliffs like streams of blood as they were rounding Cape York on the northwest coast of Greenland. A landing party stopped and brought back samples to England. Soil? Meteoric iron?

Chlamydomonas nivalis, a species of green algae containing a secondary red carotenoid pigment (astaxanthin) in addition to chlorophyll. Unlike most species of fresh-water algae, it is cryophilic (cold-loving) and thrives in freezing water.

The secondary pigment protects the chloroplast from intense visible and also ultraviolet radiation, as well as absorbing hea  which provides the alga with liquid water as the snow melts around it. Algal blooms may extend to a depth of 25 cm (10 inches), with each cell measuring about 20 to 30 micrometers in diameter. It has been calculated that a teaspoon of melted snow contains a million or more cells.

The algae sometimes accumulate in "sun cups", which are shallow depressions in the snow. The carotenoid pigment absorbs heat and as a result it deepens the sun cups, and accelerates the melting rate of glaciers and snowbanks.

During the winter months when snow covers them, the algae become dormant. In spring, nutrients, increased levels of light and meltwater, stimulate germination. Once they germinate, the resting cells release smaller green flagellate cells which travel towards the surface of the snow.

Once the flagellated cells reach the surface, they may lose their flagellae and form aplanospores, or thick-walled resting cells, or they may function as gametes, fusing in pairs to form zygotes.
Many species feed on C. nivalis, including protozoans such as ciliates, rotifers, nematodes, ice worms and springtails

https://drive.google.com/folderview?id=0B8NUHa4P2gmlfnhfcUwtRHVzZmRnWnZhQ0UwSkI3ZW12R0dYQ0R4UUR5eElPeEJSYzBsRFk&usp=sharing#list

[Wipneus]

2016/7/24 ( R111) is in, 0% clouds in the metadata. S2A_R111_V20160724T153051.22WEB.B04.tiff has been uploaded.

[A-Team]

Wow, the ablation this year is something to behold. The first image shows exposed surface flow features south of Jakobshavn ... I don't recall seeing anything as vivid as those crevasse flow fields in past years, Lincoln glacier or so. The image needs a click to display properly (recommended) but even that is only a quarter of the S2A image at its full resolution.

The second image shows the outflow area at 50% scale (20m). The 3rd and 4th show two other regions at the nominal maximal resolution of 10 m.

[A-Team]

Same image, a look at the action on the calving fronts in the south branch bay. The weak looking area coming up on the center of the calving front may be gone in a few days.

[Wipneus]

2016/7/28 ( R025) is in, cloudiness: 79.62% according to metadata. Calving front is visible, so uploaded S2A_R025_V20160728T151306.22WEB.B04.tiff

[A-Team]

Over the last ten days, there has been no significant calving along the main front of the south branch. That can be seen from the line of features at the front on 18 July still at the front on the 28th. The forward motion of the glacier is near-record centrally however at 47 meters per day averaged over six sites. The record is 52 m/d.

Over on the south flank, motion is much slower, barely measurable over sixty days at 0.34 m/d, slower by a factor of 138. The ice here is much thinner as well, say by a factor of 20, which makes the volume contribution even less, 1 part 2760.

[Espen]

Over the last 7 days July 24 to July 31 2016 heavy calving and retreat at both the Southern and the Northern branch:

[Shared Humanity]

Wow!

[Wipneus]

2016/7/31 ( R068) is in, clouds according to metadata 0.19% (looks seriously underestimated to me). S2A_R068_V20160731T151913.22WEB.B04.tiff has been uploaded.

[Espen]

Still moving on, heavy calving activity seen between July 31 and August 2 2016:

[Shared Humanity]

Where is the current calving front in relation to the historic minimum?

[A-Team]

relation to the historic minimum?

Nowhere near a record. Have to have lobe well below the island on left. However if the forward velocity should slow down but the calving keep up its current pace, then we could see new territory reached.

[nukefix]

Lake-drainage around the trunk of Jakobshaven is visible in GREEN in this RGB made of S1 IW acquired on 20.7.2016 & 1.8.2016.

I have the same image in 15m resolution but at 4MB it should be too large for this forum...?

[CraigsIsland]

Where is the current calving front in relation to the historic minimum?

for me it's more about the things that lead to the minimum of the calving front - darker snow, more snow melt, bigger lakes etc. I'm really curious how those other metrics stack up against other years and time frames. I'm still a newb at understanding the dynamics at play, but the positive feedback loops appear to be occurring and we should be seeing more sea level rise contributions year over year.

[Wipneus]

2016/8/7 ( R025) came in. Cloud pct. 85.73% but does not look that bad. S2A_R025_V20160807T150914.22WEB.B04.tiff has been uploaded.

[Espen]

Jakobshavn Isbræ update: Another large calving! Aug 11 Aug 12 2016

[Wipneus]

S2A 2016/8/13 ( R111) has come in. Cloudiness (from metadata) is 0%. Uploaded S2A_R111_V20160813T152912.22WEB.B04.tiff.

[Tealight]

Jacobshavn-20170813_Sentinel-2A_B432_10m.gif (969.22 kB, 700x700 - viewed 96 times.)

Welcome to 2017 everyone


There is a large ice berg in the fjord from a recent calving. Its a neat rectangle so we can be sure that one side is the thickness. This gives us a minimum thickness for the glacier front. When the dimensions of ice bergs increase it is a confirmation that the front reached the deeper sections of the fjord. The record retreat from last year was at the edge of the deeper parts according to NSIDC BedMachine.

[nukefix]

Off-topic: Why do certain images in this thread reach 300+ views rapidly while others do not? Are these clicks coming from the forum or perhaps from Twitter or similar?

edit: One of Wipneus' images on this page have 2100+ views...certainly some of the clicks must have come from outside the forum...but how did they find the image? hmm...

[Shared Humanity]

Tealight....great image of the underlying bed topography and the minimum for last year. I believe this image holds the key to the future with regards to the retreat of the calving face. On the northern wall of the main ice stream we see two elevated areas of the underlying topography (in white) which are where the flow of the sheet slows down. In between, where the topography is considerably lower (in yellow), the ice moves rapidly towards the main ice stream. It is clear that this ice moving towards the main ice stream serves to pin the glacier and restrict calving along the northern edge. If the calving face ever retreats past this ice and into the deep trough behind it, we may then see a rapid retreat of the calving face south and then east.

[Tealight]

Tealight....great image of the underlying bed topography and the minimum for last year. I believe this image holds the key to the future with regards to the retreat of the calving face. On the northern wall of the main ice stream we see two elevated areas of the underlying topography (in white) which are where the flow of the sheet slows down. In between, where the topography is considerably lower (in yellow), the ice moves rapidly towards the main ice stream. It is clear that this ice moving towards the main ice stream serves to pin the glacier and restrict calving along the northern edge. If the calving face ever retreats past this ice and into the deep trough behind it, we may then see a rapid retreat of the calving face south and then east.

I had the image lying around in a folder for 2-3weeks. but never managed to publish it. It is my first published bedrock overlay of Jakobshavn with correct mathematical scaling. Previously I just eyeballed it. Hard to say if it is more accurate, but at least it is more consistent. At the northern and southern shore the bedrock overlay shows 300m below sea level where it clearly isn't.

I attach another forgotten image from 18 July 2016 which shows the greater area and a bedrock legend. Click for full resolution

In the deeper parts we can at least expect a higher volume loss, but I don't know if the calving front retreats faster. After all the calving front cuts into a thick icesheet with ice supply from north, east and south. Previously it only had ice supply from the east.

Today is the one year anniversary of cloud free Sentinel 2A images of Jakobshavn. I plan to do some comparisons of land ice retreat. Especially the ice free areas near the calving front which are used as control points.

@nukefix
Maybe there are some people who only follow certain users and not a whole thread.

[nukefix]

Sentinel-1 13.8.2016 in 10m UTM22. It seems to me that a big calving is going to happen sometime soon as there are big crevasses opening 600m+ upstream from the current calving front. These are also visible on the S-2 image posted above by Wipneus.

[Espen]

Dramatic surface ice retreat seen around Jakobshavn Isbræ over the last 3 years (2013-2016) watch the red encircled areas in the animation below, a very clear sign of ice sheet thinning:

[Darvince]

Off-topic: Why do certain images in this thread reach 300+ views rapidly while others do not? Are these clicks coming from the forum or perhaps from Twitter or similar?

edit: One of Wipneus' images on this page have 2100+ views...certainly some of the clicks must have come from outside the forum...but how did they find the image? hmm...

For images which are small enough to not need a click to view at full resolution, every view of the page counts as a view, while for images that are bigger, they need to be clicked to count as a view of the image.

[Adam Ash]

Well spotted Espen 1735!  I have long wondered when the ocean is going to get around the back of the perimeter island rim of Greenland.  Once that starts then its all on as the side support of the ice sheet disappears. 

[nukefix]

Thinning at the margin is happening pretty much everywhere in Greenland I think.

[Espen]

Thinning at the margin is happening pretty much everywhere in Greenland I think.

Yes, that is written in many reports etc. all over the place.
But I find it more interesting and important to show that it is actually happening.
I think the retreat of  surface ice around Jakobshavn is dramatic, since it can easily be seen over a very short period of only 3 years.

[Tealight]

Here is my report for land surface ice retreat using S2A images from 16 Aug 2015 and 13 Aug 2016. The surface melting was pretty average without any melt spikes during summer. See post: http://forum.arctic-sea-ice.net/index.php/topic,1465.msg86782.html#new

Exacly next to Jakobshavn the ice retreated around 40m and 20-25m in the vicinity. Further to the south the ice only retreated 15m. One marine terminating ice tongue retreated 160m, but this value wasn't counted towards the total land retreat. Over the whole 110km tile length the lost area equates to around 2.61km².

All distances were obtained with the distance measurement tool in SNAP trying to get the average value for an area.

[Rubikscube]

Superb work Tealight.

[Wipneus]

2016/8/17 ( R025) is in. Cloudiness (from metadata) is 0.95%, and calving front is clear.
S2A_R025_V20160817T150912.22WEB.B04.tiff uploaded.

[Espen]

Tealight great work, good imagery is better than 1000 written reports

[RoxTheGeologist]

I was wondering if it were possible to equate the horizontal distance loss with a vertical loss?

If the slope angle is 90° then there might be seeing significant mass lost with no visible satellite change. However, if we calibrate the horizontal losses with slope angle then we could produce a map of average height lost and get to the mass lost.

Anybody have good topo maps of the region?... hmmm, tricky, as the interesting areas were covered in ice last year.

[Tealight]

I was wondering if it were possible to equate the horizontal distance loss with a vertical loss?

If the slope angle is 90° then there might be seeing significant mass lost with no visible satellite change. However, if we calibrate the horizontal losses with slope angle then we could produce a map of average height lost and get to the mass lost.

Anybody have good topo maps of the region?... hmmm, tricky, as the interesting areas were covered in ice last year.

The DMI already calculates the vertical loss from surface melting:
http://beta.dmi.dk/en/groenland/maalinger/greenland-ice-sheet-surface-mass-budget/

In areas with just 15m retreat almost all losses should come from surface melt. There are no marine terminating glaciers nearby to accelerate ice loss, only shallow melt water lakes.

If you want to do the extra work you can calculate the slope from my measurements and their model calculations.

[nukefix]

Wow, I see clear waves on the surface of the main stream. The radar-facing side of the bumps are brighter and one can see huge crevasses opening up at the top of the bumps. I wonder does this mean that the main channel is going calve big soon? 

S-1 25.8.2016 in 10m UTM zone 22.

[Iceismylife]

Wow, I see clear waves on the surface of the main stream. The radar-facing side of the bumps are brighter and one can see huge crevasses opening up at the top of the bumps. I wonder does this mean that the main channel is going calve big soon? 

S-1 25.8.2016 in 10m UTM zone 22.

You said it Wow.

[johnm33]

" I wonder does this mean that the main channel is going calve big soon? "
Wow indeed, maybe it just calved and another one is on? Confirms it's sitting in water too imho.
Meanwhile last night there was a big breakout at the outer cill, just when I was thinking maybe it's there for the winter.
https://www.lookr.com/lookout/1310041325-Ilulissat

[A-Team]

Sounds like we need to keep on close eye on Sentinel 2AB and Landsat-8 visible products. The Sentinel 2A public image browser has these dates for the latter part of August: 27, 26, 24, 23, 20, 17, 16. Wipneus last added a band 4 16-bit tiff to the gedeeld Jakobshavn collection on August 17th. However jp2 format will load ok in gimp but as 8-bit.

The latest cloud-free Landsat was acquired on 18-AUG-16, LC80080112016231LGN00. The 25th was totally clouded over; the next image is not due until Sept 1st, a path,row of 10,11. The next S1A is not due until Sept 4th (?). It's best if before and after have the same geometry.

Looking at the two Sentinel 2A's available after nukefix's remarkable S1A of the 25th, both were as cloudy as the Landsat, ie uninformative.

http://earthexplorer.usgs.gov
http://sentinel-s2-l1c.s3-website.eu-central-1.amazonaws.com/#tiles/22/W/EB/2016/8/
https://drive.google.com/drive/folders/0B8NUHa4P2gmlfnhfcUwtRHVzZmRnWnZhQ0UwSkI3ZW12R0dYQ0R4UUR5eElPeEJSYzBsRFk

[Shared Humanity]

Looking at the current calving face and the associated crevices, I am struck by the angle of the face and the fact that these large crevices align very closely to it. I don't recall the calving face having such a sharp angle in relationship to the sides of the main ice stream. Are we seeing something different, a shift in the calving dynamics that might be related to the flow of ice from the sides relative to the main stream? Could increased flow of the ice from the north wall result in this angle?

[Wipneus]

Wipneus last added a band 4 16-bit tiff to the gedeeld Jakobshavn collection on August 17th. However jp2 format will load ok in gimp but as 8-bit.
(...)
The latest cloud-free Landsat was acquired on 18-AUG-16, LC80080112016231LGN00. The 25th was totally clouded over; the next image is not due until Sept 1st, a path,row of 10,11. The next S1A is not due until Sept 4th (?). It's best if before and after have the same geometry.
(...)
Looking at the two Sentinel 2A's available after nukefix's remarkable S1A of the 25th, both were as cloudy as the Landsat, ie uninformative.

Yup, I scan the amazon cloud every day. Tail of current output:

dag : 2016/8/17 ( R025 , 100.0 , 0.95)
dag : 2016/8/20 ( R068 , 100.0 , 95.42)
dag : 2016/8/23 ( R111 , 77.13 , 97.68)
dag : 2016/8/24 ( R125 , 45.73 , 96.34)
dag : 2016/8/26 ( R011 , 1.89 , 100.0)
dag : 2016/8/27 ( R025 , 100.0 , 98.16)

Last two numbers are coverage and cloud percentage. None of the latest was of interest for ice watchers.