Pine Island Glacier (PIG) Calving and Discussion

[Wipneus]

In this sequence, 24 Dec - 4 Jan, the "camera" moves with the advancing glacier. This highlights the slight rotation of the "grounded" calving. Two minor calving can be seen that can separate at any time now. Note that the images are not from the same orbital position ( no exact multiple of 16 days  separation), not really useful to detect small changes in crack widths/lengths.

(click required)

[oren]

Beautiful

[Wipneus]

View of the "crack" with different solar lighting. The 4 Jan image (same as in the previous post) is acquired during the "nightly" semi orbit of the Landsat satellite. Sun elevation is only 7.5^o, in the image the light is from the right. The 6 Jan image is taken during the "day", with sun elevation 29.5^o, light comes from the top-left, about the 11 o'clock position.

The low sun gives a dramatic better view on the crack, showing depth where the other does not and shows it extending much further.

Landsat ID's: LC81561312016004LGN00 (4 Jan) and LC80011132016006LGN00 (6 Jan).

Click to see in 7.5 m/pix resolution

[Wipneus]

The minor calving at the PIG front has taken place, sometime between 6 and 9th of January. Again a sequnce moving with the glacier, so we see the stationary rocks and ice moving backward and see the big grounded calving rotate.

Click to see the 30m/pix aniamtion.

[A-Team]

By exploiting the greater bit depth and massively processing all possible image pairs, Landsat-8 can provide an alternative to radar interferometry in providing velocity maps of Antarctica. A new continent-wide Antarctic mosaic is ready but not yet released. However the ‘methods’ paper is out and very instructive. Pine Island Glacier furnished one of the examples, see below.

Because they also did processed Greenland (indeed all land ice globally), I’ve discussed other parts of the paper in three posts to the Store Glacier forum, links below.

People on these forums have measured ice velocity from time to time but not at the extreme sub-pixel levels achieved here. The figure shows speed for a 32-day Landsat-8 band 8 scene pair from Pine Island (longer time separations are used in slower areas). Note the orange asymmetry (white arrow) in the transition from fast red to slow green; this seems not to be an artifact but detected by SAR as well (2nd image).

2.5. Sub-pixel offset determination

To facilitate accurate displacement measurement, we take advantage of the tendency for ice sheet and glacier images to have a smoothly varying correlation surface in the vicinity of a valid match. We use a bivariate cubic spline to fit the peak in the integer-pixel offset correlation surface, and then find the sub-pixel location of the peak of this spline. For computational efficiency, we perform a maximum gradient search of this splined surface in the gridded x and y directions, locating the peak to within ~0.01 pixel.

The resulting offset fields show smoothly varying values, suggesting that the data are not being overfit. Fig. 6a shows the flow speed over Pine Island Glacier in Antarctica, while Fig. 6b shows only the sub-pixel part of calculated offsets. With only the fractional component of the y pixel offset shown, values in this figure wrap at ±0.5 pixels [ie fractions amount to mod 1 modular arithmetic similar to phase wrapping], producing an appearance similar to a differential SAR interferogram. The smoothly varying sub-pixel displacement field, particularly in the slow-moving areas, demonstrates the fidelity of both the internal image geometry and the derived offsets.

Rapid large-area mapping of ice flow using Landsat 8
M Fahnestock, TScambos, T Moon, A Gardner, Terry Haran, Marin Klinger
http://www.sciencedirect.com/science/article/pii/S003442571530211X open access

Previous commentary:
http://forum.arctic-sea-ice.net/index.php/topic,1462.msg68479.html#msg68479
http://forum.arctic-sea-ice.net/index.php/topic,1462.msg68593.html#msg68593
http://forum.arctic-sea-ice.net/index.php/topic,1462.msg68662.html#msg68662

[solartim27]

More minor calving on either side of PIG, as well as a bit of advance.  Dates are from Jan 14 to Jan 26.  Thwaites seems to advance a similar amount in the Sentinel image, but I did not notice any new calving there.

[Wipneus]

In the animation a remarkable darkening of the whole 26 Jan image can be seen. Darker images started mid January, at about the same time that the optical images (MODIS, Landsat) show thick cloud cover.
So are we looking at the effects of clouds/assumed precipitation on Sentinel's 1A radar? Something changing in the processing?

[nukefix]

It's difficult to say whether Polarview-processing has changed with respect to contrast-stretching. Wet snow has a very low backscatter as it absorbs the radar waves very well. Limited surface-melting should be possible at this time of the year, right?

[Wipneus]

It is not Polar View, it is visible from unprocessed image downloaded from the data hub as well.
Further, I have seen it before. A  couple of months ago, perhaps October cannot be sure, also with PIG images and visible clouds in the optical data.

[solartim27]

Major rotation of the pinned berg.  Something is going on under the clouds.

[AbruptSLR]

Major rotation of the pinned berg.  Something is going on under the clouds.

Nice catch!  I agree that it looks like something is happening under the clouds offshore of the PIIS.  The first attached image shows the Earth Surface Wind & MSLP map for the Amundsen & Bellingshausen Sea areas for Feb 4 2016, showing a large low pressure system causing a wind pattern contributing to local ocean upwelling, while the second image shows the Earth Ocean Current & SSTA map for Feb 5 2016 showing a large area of warm upwelled water right offshore of Pine Island Bay.  The third image shows a Landsat 8 image of the area offshore of Thwaites for Feb 5 2016, showing a major fracturing of much of the fast sea ice in this area; which could be due to a combination of strong winds and local upwelling of warm CDW.  This combination of recent local atmospheric & oceanic conditions could be contributing to basal ice melting for both the PIIS & the pinned iceberg in-front of the PIIS.

[AbruptSLR]

The attached Aqua image for Feb 6 2016, confirms that not only has the pinned iceberg rotated, but it has also translated away from the calving face of the PIIS:

[Wipneus]

Today's Sentinel 1A image is the first after the rotation event, on the previous image from 2 Feb nothing is visible.
Here is an animation composed of today's image with the one 12 days before.

(yes, a click will get it going)

[Wipneus]

A small break from the main grounded ice berg. Here I took Sentinel images from 10 and 12 February, rotated and moved them for approximate alignment and created a color image using Red=10 Feb and Green=12 Feb. The berg falls slightly outside the 10 Feb image, so on the right only the green color is available.

[oren]

Maybe this stubborn berg got tired of sticking around for so long, and decided to break up piece by piece?

[Wipneus]

In this animated sequence of two Sentinel 1A images, not only the glacier front is advancing. Looking closer, the grounded iceberg has made a tiny clock-wise rotation.

Must click for that.

[steve s]

Thanks Wipneus; the sequences are always interesting. In this case, the ongoing forces on the separated berg seem impossibly strong.

Probably something I am missing in the way of force transmission, but the glacier's movement seems to remain correlated with the bergs rotation although the two are physically separated with the berg pinned. The forces on the berg are sufficient, whatever their sources, to cause rapid growth of peripheral crevasses and so on. If a river were flowing beneath the glacier generating the forces, it would have to be a very vigorous river. I think that can be ruled out.

Any ideas?
 

[AbruptSLR]

Thanks Wipneus; the sequences are always interesting. In this case, the ongoing forces on the separated berg seem impossibly strong.

Probably something I am missing in the way of force transmission, but the glacier's movement seems to remain correlated with the bergs rotation although the two are physically separated with the berg pinned. The forces on the berg are sufficient, whatever their sources, to cause rapid growth of peripheral crevasses and so on. If a river were flowing beneath the glacier generating the forces, it would have to be a very vigorous river. I think that can be ruled out.

Any ideas?

It's not a river but rather the advection of CDW into and then out-of the sub-ice shelf cavity, and which is concentrated in the area of the iceberg.

[steve s]

Thanks ASLR.

I was originally thinking of tides being too weak. Now I realize that the specific gravity of warm water is higher than of cold, so the warm basal waters become an outflow river at the lower surface of the glacier. The strength of the flow (surprises me) is made visible by the ongoing separation between the berg and the glacier. 

[sidd]

" ... the specific gravity of warm water is higher than of cold ... "

wait, what ?

[AbruptSLR]

" ... the specific gravity of warm water is higher than of cold ... "

wait, what ?

The first image shows that the advected warm CDW melts ice at the grounding line, which then due to its low salinity has a low density (even though it is cold) flow up and out of the Pine Island Ice Shelf, PIIS, where it would hit the iceberg.

The second image shows the measured water velocities beneath, and existing from beneath, the PIIS, for both the 2009 (panel C) and the difference from 2009 to 2012 (panel D); which confirm that the water velocities near the iceberg are meaningful.

I note that wind advects the warm CDW moves over the continental shelf into a deep trough that directs the warm CDW beneath the PIIS by a combination of momentum and the siphon effect of the melted ice at the grounding line and basal ice melt from the bottom of the PIIS.

Edit: For circulation pattern see:
http://nora.nerc.ac.uk/502263/1/1244341RevisedManuscript2.pdf

Caption for second image: "Fig. 3. Observed and simulated hydrography and circulation in 2009 and 2012. A. Section of observed and simulated 2009 potential temperatures (color) and salinity (black contours) along the eastern Amundsen Sea trough and underneath the PIG ice shelf. White lines show the surface-referenced 27.47 and 27.75 isopycnals. The panel shows observations outside the PIG cavity, and simulation results within it. Observations are linearly interpolated from profiles (black triangles) indicated in figure 1B. B. Same as A but for the 2012 observations and simulation. C. Modeled potential temperature (color) and velocity (black vectors, every fifth vector is shown) averaged within 50 m of the seabed for the 2009 simulation. White vectors show the corresponding velocity observed by Autosub (binned on the model grid, see also Fig. S2A). The cyan line indicates the position of the section used in panels A and B. The white line indicates 750 m seabed depth. D. Same as C but for the difference between the 2012 and the 2009 simulations."

[steve s]

" ... the specific gravity of warm water is higher than of cold ... "

wait, what ?

Sorry I was imprecise Sidd. Fresh water's density is greatest at about 4 degrees C, so ice forms at the water's surface, where the water is at 0 degrees. Dissolved salt depresses the freezing point and increases the density, but leaves the general relationship unchanged.

Advective outflows, having lost heat to the ice, are less dense and above the inflows. The density change is due to both cooling and freshening.

The motion of the berg indicates that a large volume water is traveling at the top of the water column -- along the bottom of glacier. To my amateur eye, the strength of the outflow at the top of the water column seems either (1)  too great to be explained by melt-water alone (as ASLR's figure suggests is the case); or (2) the volume of melting is higher than seems to be indicated by apparent volume changes in the glacier.

[AbruptSLR]

The motion of the berg indicates that a large volume water is traveling at the top of the water column -- along the bottom of glacier. To my amateur eye, the strength of the outflow at the top of the water column seems either (1)  too great to be explained by melt-water alone (as ASLR's figure suggests is the case); or (2) the volume of melting is higher than seems to be indicated by apparent volume changes in the glacier.

I think that it is a mistake to assume that the motion of the pinned iceberg is only impacted by the out-flow of relatively fresh discharge along the base of the ice shelf, as it is probable that the keel of the iceberg is deep enough to also be impacted by the relatively warm CDW discharge shown in the first of my previous two images.  I note that this relatively warm CDW discharged from the PIIS is substantial stream (significantly larger than the meltwater stream) and is directed towards the Thwaites Glacier/Ice Shelf, where it retains sufficient heat to contribute to accelerated basal ice melt there [as I have extensively documented in numerous threads in this folder].

Edit: I also note that the relatively warm CDW discharge stream is likely also continuing to melt the basal ice of the iceberg, which could be contributing to the behavior of the iceberg.

[sidd]

i was under the impression that sea water had no density maximum above freezing ...

[solartim27]

I think the motion of the pinned berg has been greatly limited by the northeast corner impacting the glacier next to it as it rotates

[AbruptSLR]

i was under the impression that sea water had no density maximum above freezing ...

Sea water density changes with temperature, salinity and pressure and the freezing point is also a function of pressure & salinity; therefore, it is best to talk specifically about the water under the PIIS.  Perhaps the linked reference & associated images provide some insight as to the case near the iceberg:

Stanley S. Jacobs, Adrian Jenkins, Claudia F. Giulivi & Pierre Dutrieux (2011), "Stronger ocean circulation and increased melting under Pine Island Glacier ice shelf", Nature Geoscience, Volume: 4, Pages: 519–523, doi:10.1038/ngeo1188


http://www.nature.com/ngeo/journal/v4/n8/abs/ngeo1188.html


Abstract: "In 1994, ocean measurements near Antarctica’s Pine Island Glacier showed that the ice shelf buttressing the glacier was melting rapidly. This melting was attributed to the presence of relatively warm, deep water on the Amundsen Sea continental shelf. Heat, salt and ice budgets along with ocean modelling provided steady-state calving and melting rates. Subsequent satellite observations and modelling have indicated large system imbalances, including ice-shelf thinning and more intense melting, glacier acceleration and drainage basin drawdown. Here we combine our earlier data with measurements taken in 2009 to show that the temperature and volume of deep water in Pine Island Bay have increased. Ocean transport and tracer calculations near the ice shelf reveal a rise in meltwater production by about 50% since 1994. The faster melting seems to result mainly from stronger sub-ice-shelf circulation, as thinning ice has increased the gap above an underlying submarine bank on which the glacier was formerly grounded. We conclude that the basal melting has exceeded the increase in ice inflow, leading to the formation and enlargement of an inner cavity under the ice shelf within which sea water nearly 4 °C above freezing can now more readily access the grounding zone."

[AbruptSLR]

I think the motion of the pinned berg has been greatly limited by the northeast corner impacting the glacier next to it as it rotates

It is actually the northwest corner that has impacted the ice shelf next to it.

[steve s]

ASLR: "... as it is probable that the keel of the iceberg is deep enough to also be impacted by the relatively warm CDW discharge shown in the first of my previous two images."

I'm confused by the first part of this comment. The berg's keel should be little deeper than the PIG's front. The second part may offer a clue. The berg's pinned, so return advective water might be a factor as it flows around the pin. There may be also orographic lifting of fluids over the eastern side of the berg, dependent on the local terrain, and about that I am ignorant. However the advective inflows must be west of the berg where the depths are greater for any of this to work out.

[AbruptSLR]

ASLR: "... as it is probable that the keel of the iceberg is deep enough to also be impacted by the relatively warm CDW discharge shown in the first of my previous two images."

I'm confused by the first part of this comment. The berg's keel should be little deeper than the PIG's front. The second part may offer a clue. The berg's pinned, so return advective water might be a factor as it flows around the pin. There may be also orographic lifting of fluids over the eastern side of the berg, dependent on the local terrain, and about that I am ignorant. However the advective inflows must be west of the berg where the depths are greater for any of this to work out.

I base my comment on the first attached image in Reply #525, that shows that seaward of the PIIS calving face the top of the warm CDW profile bumps upward.  It is possible that this figure is out of date as the location of the calving face has changed and the amount of advected CDW varies with the location of the ABSL, which in turn varies with the ENSO and SAM phase.

[Andreas T]

........ Fresh water's density is greatest at about 4 degrees C, so ice forms at the water's surface, where the water is at 0 degrees. Dissolved salt depresses the freezing point and increases the density, but leaves the general relationship unchanged.
.......

according to this https://nsidc.org/cryosphere/seaice/index.html
that is not correct
as salinity increases the maximum density approaches the freezing point and in sea water with salinity above 24 psu density is highest at freezing temperature

[sidd]

So the max density temperature below the freezing point refers to supercooled salt water ?

[AbruptSLR]

So the max density temperature below the freezing point refers to supercooled salt water ?

The density of sea water has the same pressure and temperature relationships as fresh water but, with the addition of "salt", its mass is increased (see image).

[plinius]

You are missing out a point, ASLR, but as long as you are not an urchin, that's not lethal...

[Andreas T]

So the max density temperature below the freezing point refers to supercooled salt water ?

that is how I too interpret that graph, but I don't have other information.
 Obviously because salt forms no part of ice crystals and is frozen out as enriched brine, frozen sea water has a lower density than sea water at freezing temperature.
The graph shown by ASLR shows that salty sea water behaves like less salty sea water with added salt, but it also shows that neither has a density maximum at 4deg C.
The salinity axis in that graph just doesn't go as low as in the NSIDC graph

[Wipneus]

Not much to report from here. The gap between the glacier and the grounded ice berg is closing. A month or so.

[oren]

That iceberg is very stubborn 

[Wipneus]

Update: it will not take long to close the gap.

[crandles]

What then?

A) Break off bits at end of iceberg (end that is away from PIG)?
B) Rotate iceberg further?
C) Iceberg breaks into pieces and most pieces able to float away?
D) Iceberg pushed off pinning point so can remain largely intact and float away?
E) Something else?

A and B seem most likely in the short term?

[AbruptSLR]

What then?

A) Break off bits at end of iceberg (end that is away from PIG)?
B) Rotate iceberg further?
C) Iceberg breaks into pieces and most pieces able to float away?
D) Iceberg pushed off pinning point so can remain largely intact and float away?
E) Something else?

A and B seem most likely in the short term?

While the initial contact itself might only lead to A & B in the short-term, this contact will change the compressive stress field in the PIIS, which in my opinion will increase the probability of another major calving event in the July - August 2016 timeframe as noted in my Reply #366 (see below).  If so, another major calving event might dislodge the currently pinned iceberg.

Extract from Reply #366: "... the new major calving will occur along the crack highlighted by A-Team in Reply #357; which more or less means that I am "predicting" another major calving event in next year's July-August timeframe."

[Wipneus]

An animation at high resolution (IW, 10m/pix) shows convincingly that the grounded ice island is rotating slightly. Remaining time for contact is about 24 days (two Sentinel 1 cycles).

[Wipneus]

Real soon now.

[Tor Bejnar]

There are hints of change (widening?) in the fractures across the iceberg (and ice shelf), but these might just be differences in the electromagnetic spectrum ("light" if the images had been taken with an old Earthling's camera) at image-taken times.  Whether fractures have grown or not already, I rather expect the iceberg to break into about 3 pieces when pushed seaward, given that parts of it are grounded.

[Wipneus]

There are hints of change (widening?) in the fractures across the iceberg (and ice shelf), but these might just be differences in the electromagnetic spectrum ("light" if the images had been taken with an old Earthling's camera) at image-taken times.  Whether fractures have grown or not already, I rather expect the iceberg to break into about 3 pieces when pushed seaward, given that parts of it are grounded.

The medium resolution Sentinel images (EW extra wide, 40m/pix) make it difficult to judge such effects. You must look at the features over longer time spans and so.
Here is a high resolution (IW, Interferometric Wide, 10m/pix) animation, scaled to 40m/pix, same time of observation difference, to compare.

[solartim27]

Not quite there yet, but looks like you could almost jump across.  May 1 to May 13.  Gif seems to not be working, I just see the May 13 image.
http://www.polarview.aq/images/105_S1jpgfull/S1A_EW_GRDM_1SSH_20160513T044346_2666_S_1.final.jpg

and  S1A_EW_GRDM_1SSH_20160501T044346_1B97_S_1.final

[Wipneus]

I'd say contact is made. Curious that this 20160513 image appeared in time on Polar View, but not in ESA's data hubs. I expect next image on 18 May, will the grounded ice island move or break?

[Shared Humanity]

There are a lot of very visible fractures on that ice island. My guess is it will shatter.

[theoldinsane]

Bump

What then?

A) Break off bits at end of iceberg (end that is away from PIG)?
B) Rotate iceberg further?
C) Iceberg breaks into pieces and most pieces able to float away?
D) Iceberg pushed off pinning point so can remain largely intact and float away?
E) Something else?

A and B seem most likely in the short term?

While the initial contact itself might only lead to A & B in the short-term, this contact will change the compressive stress field in the PIIS, which in my opinion will increase the probability of another major calving event in the July - August 2016 timeframe as noted in my Reply #366 (see below).  If so, another major calving event might dislodge the currently pinned iceberg.

Extract from Reply #366: "... the new major calving will occur along the crack highlighted by A-Team in Reply #357; which more or less means that I am "predicting" another major calving event in next year's July-August timeframe."

I guess that if the shelf is thicker than the island it will be another outcome compared to if the shelf and the island are of near equal thickness. Maybe anyone of the experts can comment? Anyway, it is exciting and maybe interesting for future bigger events of its kind.

[AbruptSLR]

I guess that if the shelf is thicker than the island it will be another outcome compared to if the shelf and the island are of near equal thickness. Maybe anyone of the experts can comment? Anyway, it is exciting and maybe interesting for future bigger events of its kind.

I point-out that another major calving event will occur for the PIIS along existing crevasses; it is only a question of when, and what will trigger such a calving (as impacting the grounded iceberg is not a requirement for such a major calving event).

[theoldinsane]

I guess that if the shelf is thicker than the island it will be another outcome compared to if the shelf and the island are of near equal thickness. Maybe anyone of the experts can comment? Anyway, it is exciting and maybe interesting for future bigger events of its kind.

I point-out that another major calving event will occur for the PIIS along existing crevasses; it is only a question of when, and what will trigger such a calving (as impacting the grounded iceberg is not a requirement for such a major calving event).

Yes, but suppose you at some point in the future have a lot of grounded islands at the same time. Will they slow down the shelf via some kind of buttress effect (negative feedback) or will they help to speed up the calving through "change the compressive stress field in the PIIS" (positive feedback). Or has it no significance?

[AbruptSLR]

I guess that if the shelf is thicker than the island it will be another outcome compared to if the shelf and the island are of near equal thickness. Maybe anyone of the experts can comment? Anyway, it is exciting and maybe interesting for future bigger events of its kind.

I point-out that another major calving event will occur for the PIIS along existing crevasses; it is only a question of when, and what will trigger such a calving (as impacting the grounded iceberg is not a requirement for such a major calving event).

Yes, but suppose you at some point in the future have a lot of grounded islands at the same time. Will they slow down the shelf via some kind of buttress effect (negative feedback) or will they help to speed up the calving through "change the compressive stress field in the PIIS" (positive feedback). Or has it no significance?

Each case is different.  While potential buttressing from a mélange is real, it is also temporary, and the amount and influence of the buttressing is dependent on geometry and dynamics.