Jakobshavn Isbræ / Ilulissat Isfjord / West Greenland

[A-Team]

Backing up here to Espen's collection of maximal calving front retreats, that notion may not be optional for characterizing calving front history since our Landsat snapshot are so sparse and irregular. We could use Radarsat/Sentinel radar for recent years to the extent DMI offers them.

Another idea is to make a weighted average rather than try to identify the absolute maximum. This might give better statistics for trending into the future. The idea graphically is to simply trace the calving front in black in a separate layer over each photo date. These traces are otherwise blank so they can be composited (averaged) in Gimp or ImageJ. This has the effect of making a grayscale of the calving front, with the degree of blackness proportional to the time the calving front spent there. This results in a probabilistic representation that is more useful than a one-off description

[A-Team]

The animation compares the region east of the calving front for the 14 and 16th August Landsats. The latter has thin clouds over some of the image. The alignment is anchored in the upper left via a rotation by a rotation of 5.63º and is not globally ideal. It will need a click to get rolling.

The lower three images look at the Radarsat for changes that took place on the 17th. There does seem to have been further calving lobes in the south.

[oren]

The animation compares the region east of the calving front for the 14 and 16th August Landsats. The latter has thin clouds over some of the image. The alignment is anchored in the upper left via a rotation by a rotation of 5.63º and is not globally ideal. It will need a click to get rolling.

The lower three images look at the Radarsat for changes that took place on the 17th. There does seem to have been further calving lobes in the south.

From this superb animation it almost seems like there is another northern branch developing on north wall of the south (main) branch. To my untrained eye it suddenly looks more than just random side calvings and general widening of the channel, but rather like a defined inroad deeper into the cliff.

[Shared Humanity]

The animation compares the region east of the calving front for the 14 and 16th August Landsats. The latter has thin clouds over some of the image. The alignment is anchored in the upper left via a rotation by a rotation of 5.63º and is not globally ideal. It will need a click to get rolling.

The lower three images look at the Radarsat for changes that took place on the 17th. There does seem to have been further calving lobes in the south.

From this superb animation it almost seems like there is another northern branch developing on north wall of the south (main) branch. To my untrained eye it suddenly looks more than just random side calvings and general widening of the channel, but rather like a defined inroad deeper into the cliff.

Most, if not all, of that north wall of the south branch is grounded below sea level. The portion of the ice sheet that is grounded below sea level goes well into the interior. That is a new calving front that will eventually dwarf the north calving front.

[A-Team]

Ok north branch fans ... here is the day after the big event compared to the previous Landsat of the same orbit 16 days earlier. Again will need a click to start. Things are moving but not nearly as fast as with the south branch or north side of the south branch (which is now unblocked).

[Tor Bejnar]

Go, Team, Go!

[A-Team]

And now, the north side of the south branch which seems to have a large cult following (and rightly so!). Same days of year: 228 and 212. I'll put in additional frame dates at the end of season, maybe 2014 and 2013 too. Click to animate.

[sidd]

This is what cliff collapse looks like on a 5km wide, 1Km plus deep grounded front, Bassis in action, together with hydro crevassing and fast flow. I shudder to think whats Thwaites will look like over a 55 Km front and no sideways buttress.

[Adam Ash]

Looking at the terrain models above, the south branch has a few deep pits which the glacier has to climb through - entailing massive dip and rise deflections.

Has there been any finding of ice (with lots of water entrained at depth) fragmenting or stratifying into near horizontal slabs with the upper slabs moving faster than the grounded lower slabs?  I see this as a potential mechanism which would set up the ice to move to some degree independent of the macro-texture of the bedrock, especially where an ice stream has some deep pits to pass over.

The resulting motion of surface slabs running on top of near-horizontal water-lubricated basement ice could allow for significant acceleration of ice movement, especially when supported by large volumes of melt water introduced to the sliding surface via moulins.

[Sleepy]

This is what cliff collapse looks like on a 5km wide, 1Km plus deep grounded front, Bassis in action, together with hydro crevassing and fast flow. I shudder to think whats Thwaites will look like over a 55 Km front and no sideways buttress.

You're not the only one.
What amazes me the most, is the total lack of interest from my countrymen.

A-team, marvellous images above.

[CraigsIsland]

This is what cliff collapse looks like on a 5km wide, 1Km plus deep grounded front, Bassis in action, together with hydro crevassing and fast flow. I shudder to think whats Thwaites will look like over a 55 Km front and no sideways buttress.

You're not the only one.
What amazes me the most, is the total lack of interest from my countrymen.

A-team, marvellous images above.

I'll 2nd exactly what you wrote on both fronts.

Thank you master observers

[A-Team]

Good points, all of them. Meanwhile I'm still trying to figure out just what is going on with the north shore of the south branch. It is an interesting area. The animation shows the 5 cloud-free Landsat path,row 8,11 scenes we've had for 2015.

Adam Ash is bringing in some fresh thinking:

• massive dip and rise deflections from pits? [not manifested in ice stream surface lidar]
• ice stratifying into horizontal slabs, upper moving faster than lower? [all layering lost in ice penetrating radar, ice deforming plastically]
• ice moving independently of bedrock over deep pits [basal meltwater in season, bed may have hydrated till, ice not frozen-on]

The resulting motion of surface slabs running on top of near-horizontal water-lubricated basement ice could allow for significant acceleration of ice movement, especially when supported by large volumes of melt water introduced to the sliding surface via moulins.

Another week of this at JI and there'll be good reason for concern. sidd is likely referring to: http://aoss.engin.umich.edu/people/jbassis

Walker C.C., R. Czerwinski, J.N. Bassis and H.A. Fricker, (2013), Structural and environmental controls on Antarctic ice shelf rift propagation inferred from satellite monitoring, Journal of Geophysical Research- Earth Surfaces, in press.

Bassis, J. N., and Jacobs, S., (2013), Diverse calving patterns linked to glacier geometry. Nature Geoscience, 6(10), 833-836.

Duddu, R., J. N. Bassis, and H. Waisman, (2013), A numerical investigation of surface crevasse propagation in glaciers using nonlocal continuum damage mechanics, Geophysical Research Letters., 40, 3064–3068, doi:10.1002/grl.50602.

Bassis, J.N. and C.C. Walker, (2011), Upper and lower limits on the stability of calving glaciers from the yield strength envelope of ice, Proceedings of the Royal Society, doi: 10.1098/rspa.2011.0422, p. 1-19.

Bassis, J.N., (2011), The Statistical Physics of Iceberg Calving and the Emergence of Universal Calving Laws, Journal of Glaciology, (57)201, p. 3-17. 

Bassis, J.N., (2010), Hamilton Type Principles Applied to Ice Sheet Dynamics:  New approximations for the large-scale flow of ice sheets, Journal of Glaciology, (56)197, p. 497-513.

Fricker, H.A., N. W. Young, R. Coleman, J. N. Bassis, J.B. Minster, (2005), Multi-year monitoring of rift propagation on the Amery Ice Shelf, East Antarctica, Geophysical Research Letters, 32, L02502, doi:10.1029/2004GL021036.

[Adam Ash]

'*massive dip and rise deflections from pits? [not manifested in ice stream surface lidar]'

For the diagrams showing the ground profile please refer to post here by A-Team:-  << Last edited: August 17, 2015, 05:35:04 PM by A-Team>>

AND

'*ice moving independently of bedrock over deep pits [basal meltwater in season, bed may have hydrated till, ice not frozen-on]'

To my eye, from these contour maps of the south branch, reading from bottom (calving face) to top (left to right on the diagrams) we see contours of -1300 (at the calving front), -900, -1200, -1100, -1300, -1500, -1300, -1500, -1400, -1300, -1100, -1500, -1000 at the top of the channel.
So analysing those again from left to right (going uphill) we find the following rises of the bed into the face of the ice flow (falls not noted):+300, +200,+200,+200,+400. 

I don't have decent data on the actual surface level of the glacier, but if I flew down the glacier on its fall line along this channel I would be most surprised to see elevation changes of +200 to +400 in the ice surface rising up in front of me where the ice is climbing out of a low section of the channel.

Hence my suggestion that there could be some sliding of layers (of relatively constant thickness) across these pits which are in-filled by relatively immobile ice, rather than the ice stream proceeding at a constant thickness over the downstream rises.

So, in relation to:
'*ice stratifying into horizontal slabs, upper moving faster than lower? [all layering lost in ice penetrating radar, ice deforming plastically]'
I recall ice penetrating radar and cores of Greenland which have located melt water (as liquid water complete with its heat, as mush and as new ice) in lenses at depth in the firn, and presumably this stratification can occur through the ice sheet in depth, persisting in ice which is discharging via the out fall glaciers.  Such lenses could create sliding surfaces that are quite slippery compared with the basement, and could give rise to significantly higher ice velocities (albeit localised by the extent of the fracture plane) than other calculations (which envisage a homogeneous full-depth ice mass whose progress is determined - in part - by base dynamics) would allow.

Just a thought...

[Iceismylife]

The animation compares the region east of the calving front for the 14 and 16th August Landsats. The latter has thin clouds over some of the image. The alignment is anchored in the upper left via a rotation by a rotation of 5.63º and is not globally ideal. It will need a click to get rolling.

The lower three images look at the Radarsat for changes that took place on the 17th. There does seem to have been further calving lobes in the south.

From this superb animation it almost seems like there is another northern branch developing on north wall of the south (main) branch. To my untrained eye it suddenly looks more than just random side calvings and general widening of the channel, but rather like a defined inroad deeper into the cliff.

Most, if not all, of that north wall of the south branch is grounded below sea level. The portion of the ice sheet that is grounded below sea level goes well into the interior. That is a new calving front that will eventually dwarf the north calving front.

(Small potatoes.)

First of thank you for the very nice illustration.

If you follow the main stream upstream a bit you see an off branch on the north side of the main channel that is almost as deep as the main branch headed inland, and not pictured it goes far inland.  When the calving face unplugs that one, by retreating, it will be fun to watch.  (From 4200 ft elevation.)

[Espen]

Looks like Jakobshavn continued the project since our last visit there:

[Lennart van der Linde]

Look who's tweeting:
https://twitter.com/erignot/status/633572684710866948

[A-Team]

The animation below shows the average velocity between day 228 and day 212, the previous Landsat with the same geometry. The center-of-scene times are 5 seconds off from being exactly 16 days apart. The images are modified for contrast and enlarged 2x (to 7.5 m resolution) within their native 16-bit format (actually 12-bit), then matching features are connected at another 2x exaggeration within Gimp.

For example, if the match points have separated by 69.4 pixels, the velocity in meters per day is 32.5=69.4*7.5/16. The sensitivity to a one pixel error gives a range of 32.1-33.0. Note the velocity seems slower towards the sides. Features seem to be holding up quite well over this time span, ie crevasse widening is not a source of artifact even though the points are quite near the front.

There are better automated ways of doing this over thousands of 'chips' so I'll just leave it as giving some idea of how fast the ice stream was advancing. The all-time reported record is something like 52 m/d. However a velocity in the 'upper 30's' fits well within the Joughin seasonal scheme above, as posted by Crandles at http://forum.arctic-sea-ice.net/index.php/topic,154.msg60965.html#msg60965

[A-Team]

Great to see folks doing some independent thinking on the ice physics behind this event -- and where it might lead. After that, might take a look at what seasoned glaciologists consider the core issues for Jakobshavn. The two best recent papers are open source:

Brief Communication: Further summer speedup of Jakobshavn Isbræ
I Joughin et al 03 Feb 2014
http://www.the-cryosphere.net/8/209/2014/tc-8-209-2014.html

Seasonal to decadal scale variations in the surface velocity of Jakobshavn Isbrae, Greenland: Observation and model-based analysis
I Joughin et al 25 May 2012
http://onlinelibrary.wiley.com/doi/10.1002/2014GL062248/full

Eric Rignot, like Joughin originally a Ph.D in electrical engineering, is very familiar with interferometric measurement of ice sheet velocities (having been observing glaciers with SAR since 1989), so it's interesting to see his Twitter take on this Jakobshavn event ("galloping retreat"). We've covered his most recent field work on Greenland glaciers (notably Rink and Store calving front sonar) over at another forum:

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

I wonder if he has access to other types of satellite imagery that he could share through that site. We have maxed out on our time coverage of the event itself and can hardly follow the aftermath with just Modis. Here are Rignot's presentations coming at AGU2015 this December.

Ocean Melting Greenland (OMG) bathymetric survey of northwest Greenland:implications for recent evolution of its glaciers
Eric J Rignot and MW Wood [http://tinyurl.com/paxfb5f]

A bathymetry survey of northwest Greenland  took place July 22-August 19 and Sept 2-Sept 14 from Ilulissat to Thule. We deployed a multibeam Reson 7160 with 512 beams installed on the hull of the Cape Race vessel, with enhanced capabilities for fjord wall and ice face mapping. The survey tracks were optimized with airborne gravity data collected by NASA Operation IceBridge which indicated the presence of troughs, bed topography mapped inland using a mass conservation approach, and the spatial distribution of ice discharge.

The goal was to identify all troughs that are major pathways for subsurface ocean heat and constrain glacier ice front thickness. The data reveal many deep, U-shaped, submarine valleys connected to the glaciers, intercut with sills and over deepened in narrower passages where former glaciers and ice streams merged into larger units; as well as fjords ending in shallow plateaus with glaciers in retreated positions.

The presence of warm, salty water of Atlantic origin (AW) in the fjords is documented using CTD. Some glaciers sit on shallow plateaus in cold, fresh polar waters at the end of deep fjords, while others are deeper and standing in AW. Satellite imagery from 1962 to 2015 illustrate how the evolution of the glaciers under ocean thermal forcing has been modulated by the presence and/or absence of natural pathways for AW.

Observing and Understanding Changes in Polar Ice Sheets and Glaciers Using Airborne and Satellite Remote Sensing
EJ Rignot [session 8074 convener]  https://agu.confex.com/agu/fm15/preliminaryview.cgi/Session8077

Ice-ocean interactions in Greenland and Antarctica
EJ Rignot [session 8074 convener] https://agu.confex.com/agu/fm15/preliminaryview.cgi/Session8074

[sidd]

Yes, i should have said that the Bassis paper i had in mind was the one on Bassis, J. N., and Jacobs, S., (2013), Diverse calving patterns linked to glacier geometry. Nature Geoscience, 6(10), 833-836

I repost Fig 1 b which shows the instability at 1km total height.

[A-Team]

Do we have some people here who already have a clue on how to access and interpret seismic data?

It seems to me the Ilulisat station of the Greenland Ice Sheet Monitoring Network (GLISN) should have recorded portions of the Jakobshavn event, helping us with timing and substructure. Not sure the best way of going about this, but so far have drilled in a ways eventually getting daily Iluli seismographs for our date range:

Greenland Ice Sheet Monitoring Network
http://ds.iris.edu/ds/nodes/dmc/services/seedlink/
http://ds.iris.edu/ds/nodes/dmc/data/#requests
http://ds.iris.edu/mda/DK/ILULI
http://ds.iris.edu/ds/nodes/dmc/tools/stationmonitor/DK/ILULI/

IRIS established the Greenland Ice Sheet Monitoring Network (GLISN) of seismic and geodetic sensors to observe the dynamic behavior of the Greenland Ice Sheet as it interacts with the atmosphere, oceans, and solid Earth. GLISN offers the unique ability to record, understand, and respond to changes in the size and scope of the Greenland Ice Sheet’s behavior as it reacts to climate change. Glacial-earthquakes are increasing in occurrence, and their spatiotemporal patterns continue to evolve. Rates of crustal seismicity are also expected to grow as ice melts and withdraws, but the background rates are poorly known.

Models of the Greenland Ice Sheet are hampered by limited knowledge of its basal conditions including the geothermal heat flux, and characterization and interpretation of mass-loss signals observed using GRACE and GPS data are complicated by limited knowledge of the viscoelastic properties of the crust and mantle underlying Greenland.

Seismic imaging techniques using observations of the propagating seismic wavefield can characterize the subsurface structure, permitting inference of heat flux and the space- and time-varying response to surface loading and unloading. Similarly, analyses of seismic source signals due to ice and water motion, rock fracturing, and ocean loading provide constraints on processes including iceberg calving, deformation within the ice mass and at the bed, changes in the bedrock stress state, and variability in sea state and sea-ice cover. Such observations provide key input for the effort to understand the effects of surface melt and ice-ocean interactions on ice-sheet and glacier dynamics.

All data and metadata are freely and openly available through the IRIS Data Management Center. GLISN allows the Arctic science community full access to year-round recordings of seismic and geodetic signals produced by the Greenland Ice Sheet.

Here is what we are looking for (legend for first image):

Spectrogram (bottom) and waveforms (top) of of 14 s of seismic data showing one intermediate depth (marked with red box) and three shallow icequakes (Röösli et al., 2012). Spectrum of the intermediate depth icequake is characterized by high frequency content in the 20-80 Hz range; shallow icequakes are characterized by much lower frequency content in the 10-20 Hz range. Waveforms at the top are bandpass filtered in the 1-80 Hz range.

Here is one of a dozen or so seismic studies in West Greenland:

Glacier, fjord, and seismic response to recent large calving events, Jakobshavn Isbræ, Greenland
JM Amundson et al
http://onlinelibrary.wiley.com/doi/10.1029/2008GL035281/full
The recent loss of Jakobshavn Isbræ's extensive floating ice tongue has been accompanied by a change in near terminus behavior. Calving currently occurs primarily in summer from a grounded terminus, involves the detachment and overturning of several icebergs within 30–60 min, and produces long-lasting and far-reaching ocean waves and seismic signals, including “glacial earthquakes”. Calving also increases near-terminus glacier velocities by ∼3% but does not cause episodic rapid glacier slip, thereby contradicting the originally proposed glacial earthquake mechanism. We propose that the earthquakes are instead caused by icebergs scraping the fjord bottom during calving. The recent loss of Jakobshavn Isbræ's extensive floating ice tongue produces long-lasting and far-reaching ocean waves and seismic signals

[plinius]

naive suggestion by my side - get the data plus at least one reference station about 1000km or so away (and _NOT_ in Iceland...). Then look for the following: Low Hz waveforms (iceberg singing, should be lower frequency than the normal noise), strong and _short_ bursts, because the longer ones are usually far away earthquakes, and avoid notable s-->p structure (that selects close events in addition, if you do not have directional information you recognize those with a lower amplitude precursor that quite suddenly jumps into a far larger amplitude that then decays, s waves travel faster/more direct than the transversal p-waves in the crust). Once you have those, you check with your reference station that the event is far weaker there. No fun, getting through that forest...

P.S.: Can also cross-check with that:
http://www.emsc-csem.org/Earthquake/

[AbruptSLR]

The first figure is from Holland et al. (2008) and shows how when the atmospheric/oceanic conditions are correct the Irminger Current (the pink curve in panel a) conveys relatively warm water to Disko Bay and from there into Ilulissat Icefjord.

David M. Holland, Robert H. Thomas, Brad de Young, Mads H. Ribergaard & Bjarne Lyberth (2008), "Acceleration of Jakobshavn Isbræ triggered by warm subsurface ocean waters", Nature Geoscience 1, 659 – 664, doi:10.1038/ngeo316

The second figure is from Gladish et al. (2015, Part I), and shows how the modified Irminger Current water reaches the calving front of the Jakobshavn Glacier and accelerates cliff failure calving events.

Carl V. Gladish, David M. Holland, Aqqalu Rosing-Asvid, Jane W. Behrens, and Jesper Boje, 2015: Oceanic Boundary Conditions for Jakobshavn Glacier. Part I: Variability and Renewal of Ilulissat Icefjord Waters, 2001–14. J. Phys. Oceanogr., 45, 3–32, DOI: 10.1175/JPO-D-14-0044.1

[TenneyNaumer]

What great animations!  Thanks so much for all that hard work!

Things really are hoppin' 'n poppin' at the Jakobshavn Isbrae (what is plural for Isbrae?  or should we get a translation of "forked tongue"?)

I also saw that the Loch Ness Monster is alive and well and having a fine time up in the north branch.

It is interesting that since at least 2009, the first two weeks in August have been a time when a lot of the moulins up and disappear en masse. 

[Espen]

That is Isbræer! (not easier)

[Carex]

All this excitement and no one is lamenting that Espen's proposed camp-out on the rocks, to watch the calving, never materialized!

[A-Team]

Lamenting

It looks to me from that June 2015 phone-filming of a large calving event that much of the front is no longer visible from the last rocks on the SE side of the south channel. That would be ever so much more the case so today -- the calving lobe is around the corner. It would be very dangerous to go out on the adjacent crevasse fields, even slow moving ones. Still, the noise and bergs floating by would be impressive even if the primary calving was not visible or several km away.

I re-located DM Holland's web cams and weather station. It has a better view from the north shore and is a whole lot easier to get to from town. They seem stuck on 29 Jul 15; so far I have not obtained any information about the mid-August calving event. Their photo is awesome at 2842 x 1602 resolution (shown reduced below).
http://efdl_ems.cims.nyu.edu/aws_jig/overview.html

The best vantage point really is the divider between north and south branches. While this looks like a crazy unstable icefall, several years of Landsat show that it is very stable. It would take a helicopter to get there and, if Greenland is anything like going out to Baja islands, you would not want to pay the return fare in advance.

I've attached the Radarsat from yesterday whose orbit, like the Sentinel's, missed the calving front itself. Those semi-circular waves of debris spreading out may or may not be artifactual. There is no way of enhancing these images further.

[A-Team]

Abstracts are closed already for the next AGU meeting, Dec 2015 San Francisco. Hopefully we will have a few live bloggers this year. A few keyword searches turns up the Greenland talks. However the 'abstract' links are dead; AGU search requires a password but oddly googling the titles often recovers the abstract (ie the AGU site is still broken but differently from last year). https://agu.confex.com/agu/fm15/preliminaryview.cgi/start.html

Abstracts can be found by putting 'http://fallmeeting.agu.org/2015/abstract/' in front of the title (spaces replaced by dashes). The alphabetic author search is quite effective: https://agu.confex.com/agu/fm15/preliminaryview.cgi/peopleindex_y

Calving and velocity variations observed by Terrestrial Radar Interferometry at Jakobshavn Isbræ in 2015
S Xie http://tinyurl.com/q889dmr

The Autumn of break-ups: When Jakobshavn Isbrae lost its floating tongue
A Aschwanden http://tinyurl.com/phr23cr

Bed topography of Jakobshavn Isbræ, Greenland from high-resolution gravity data
Lu An http://tinyurl.com/ou3qm35

Seismic Imaging of Sub-Glacial Sediments at Jakobshavn Isbræ and NEEM Greenland
GP Tsoflias http://tinyurl.com/nwbkzoh

Quantifying Sub-Glacial Abrasion at Jakobshavn Isbræ: A Novel Approach Using In Situ 10Be Measurements
NE Young [abstract not found]

Basin-Wide Mass Balance of Jakobshavn Isbræ during 1880-2100
IS Muresan http://tinyurl.com/psckmkz

Seasonal and interannual evolution of Jakobshavn Isbrae, Greenland from a 2008-2015 high-res DEM and velocity time series
DE Shean http://tinyurl.com/qjnuyrj

Paleo-reconstruction of the Jakobshavn Glacier during the late Holocene using ISSM and Paleo-data of Margin Migration
http://tinyurl.com/oqjlrry

Spatial Elevation Changes Observed at Helheim Glacier and Comparison with Jakobshavn Isbræ
C Roberts http://tinyurl.com/np27or5

Subaqueous melting in Zachariae Isstrom, Northeast Greenland combining observations and an ocean general circulation model
C Cai http://tinyurl.com/oq62mje

Deformation Studies of NEEM, Greenland Basal Folded Ice
K Keegan http://tinyurl.com/pwgdbrl

Ocean observations from below Petermann Gletscher
Andreas Muenchow [no abstract, work ongoing]

Reverse glacier motion during iceberg calving and the cause of glacial earthquakes
Tavi Murray http://tinyurl.com/opduxky
Glacier seismicity update to 2015 Science article free full http://tinyurl.com/ozrkfub

[Rubikscube]

The best vantage point really is the divider between north and south branches. While this looks like a crazy unstable icefall, several years of Landsat show that it is very stable. It would take a helicopter to get there and, if Greenland is anything like Mexico, you would not want to pay the return fare in advance.

Whoa, hold your horses... you mean like camping on Cape Ice? Those crevasses aren't harmless just because they aren't moving. I presume you will have to scout a location in advance at least (that would be some additional expenses on worldview imagery i suppose). Fortunately, I imagine every country in the world to be more like Mexico than Greenland .

[A-Team]

those crevasses aren't harmless... to scout a location in advance

Fine, stay home then and read the adventure blog. WorldView for 25 Aug 2012 below.

[A-Team]

Some of these AGU talks will shed experimental light on perennially speculative discussions on our forums, such as the true bedrock topography, sediments on top of that, responses to tide and timing of south branch collapse. NEEM, good grief, they could have kept drilling and obtained a 50 m sediment core

Bed topography of Jakobshavn Isbræ, Greenland from high-resolution gravity data
Lu An, UC Irvine [Rignot group?]

Jakobshavn is one of the largest marine terminating outlet glaciers in Greenland, feeding a fjord about 800 m deep in the west coast. It sped up more than twofold since 2002 and contributed nearly 1 mm of global sea level rise during the period from 2000 to 2011. Holland 2008 posit that these changes coincided with a change in ocean conditions beneath the former ice tongue, yet little is known about the depth of the glacier at its grounding line and upstream of the grounding line and the sea floor depth of the fjord is not well known either.

Here, we present a new approach to infer the glacier bed topography, ice thickness and sea floor bathymetry near the grounding line using high-resolution airborne gravity data from AirGRAV collected in August 2012 from a helicopter platform. The data combined with radio echo sounding data, discrete point soundings in the fjord and the mass conservation approach on land ice.

AirGRAV acquired a 500m spacing grid of free-air gravity data at 50 knots with sub-milligal accuracy, i.e. much higher than NASA Operation IceBridge’s 5.2km resolution at 290 knots. We use a 3D inversion of the gravity data combining our observations and a forward modeling of the surrounding gravity field, and constrained at the boundary by radar echo soundings and point bathymetry. We reconstruct seamless bed topography at the grounding line that matches interior data and the sea floor bathymetry.

The results reveal the true depth at the elbow of the terminal valley and the bed reversal in the proximity of the current grounding line. The analysis provides guidelines for future gravity survey of narrow fjords in terms of spatial resolution and gravity precision. The results also demonstrate the practicality of using high resolution gravity survey to resolve bed topography near glacier snouts in places where radar sounding has been significantly challenged in the past. The inversion results are critical to re-interpret the recent evolution of Jakobshavn and reduce uncertainties in projecting its future contribution to sea level.


Calving and velocity variations observed by Terrestrial Radar Interferometry at Jakobshavn Isbræ, Greenland, in 2015
Surui Xie, U South Florida Tampa

We observed the highly dynamic terminus of Jakobshavn Isbræ in Greenland by using a Terrestrial Radar Interferometer (TRI) during a 5 days’ period in early June, 2015. Calving and ice surface velocity variations were captured by our continuous measurements with a sampling rate of 90 seconds. Our terrestrial-derived time series show that calving events are characterized by suddenly fluctuations in surface velocities, which is very distinct in the mélange and less distinct on the glacier. Except for the relatively fast and steady motion, the glacier also moves in response to the semidiurnal ocean tides, and the impact of tides decreases rapidly upstream from the terminus.

Seismic Imaging of Sub-Glacial Sediments at Jakobshavn Isbræ and NEEM Greenland
Georgios Tsoflias,  UKansas [Cresis?]

Sub-glacial sediment conditions can have a major control on glacier flow yet these are difficult to measure directly. We present active source seismic reflection experiments that imaged sub-glacial sections at Jakobshavn Isbræ and at the NEEM drill site.

At Jakobshavn Isbræ we re-processed an existing 9.8 km-long high-resolution seismic line using an iterative approach to determine seismic velocities for enhancing sub-glacial imaging. The seismic profile imaged sediments ranging in thickness between 35 and 200 meters over underlying bedrock.

Based on the geometry of the reflections we interpret three distinct seismic facies: a basal till layer, accreted sediments and re-worked till. The basal till and accreted sediments vary in thickness from less than 5 m to nearly 100 m thick and are interpreted as the zone of most recent deposition.

A reflection polarity reversal observed at a low topographic region along the ice-sediment interface suggests the presence of liquid water spanning approximately 200 m along the profile.

At NEEM we acquired a 5.8 km long-offset shot gather. Seismic imaging revealed two prominent reflections at the base of the ice. The upper reflection is interpreted at the base of ice – top of till interface whereas the lower reflection is interpreted as the base of till – top of bedrock. The thickness of the subglacial sediment section at NEEM is estimated to approximately 50 m using seismic imaging. The NEEM ice core drilled through the upper part of this section and ceased drilling before reaching bedrock.


Basin-Wide Mass Balance of Jakobshavn Isbræ (West Greenland) during 1880-2100
Ioana Muresan, DTU Space Denmark

Here we use a 3-D PISM modeling approach... thinning and retreat is mostly controlled by a loss of resistive stresses at the terminus through glacier dynamics induced calving rather than by changes in oceanic temperatures.

We find that under atmospheric/no ocean RCP 4.5 and RCP 8.5 forcing an increase in ocean temperatures of just 0.7 °C (relative to 1880-2012) is enough to trigger a collapse of the JI’s southern tributary by 2050 which further destabilizes JI and unleashes a major glacial collapse of ~25 km.

JI’s contribution to SLR is 2.8 mm for the period 1880 to 2014, from which the contribution between 1997 to 2014 represents 27%. By the end of the century contributions to SLR as high as 11 mm under RCP 8.5 can be expected from Jakobshavn Isbræ alone.

[AbruptSLR]

The linked reference by Hughes et al (2015) presents state-of-the-art analysis about the Jakobshavn Effect (Hughes, 1986), focused on progressive ice-bed uncoupling due to such factors as: basal meltwater, buoyancy friction (particularly with changing surface elevation indicated in the attached image for Jakobshavn), boundary constraints of the fjord.  This work has relevance to multiple marine-terminating, and marine, glaciers in both Greenland and Antarctica:

Hughes, T., Sargent, A., Fastook, J., Purdon, K., Li, J., Yan, J.-B., and Gogineni, S.: Sheet, stream, and shelf flow as progressive ice-bed uncoupling: Byrd Glacier, Antarctica, and Jakobshavn Isbrae, Greenland, The Cryosphere Discuss., 9, 4271-4354, doi:10.5194/tcd-9-4271-2015, 2015.

http://www.the-cryosphere-discuss.net/9/4271/2015/tcd-9-4271-2015.pdf

Abstract. The first-order control of ice thickness and height above sea level is linked to the decreasing strength of ice-bed coupling alone flowlines from an interior ice divide to the calving front of an ice shelf. Uncoupling progresses as a frozen bed progressively thaws for sheet flow, as a thawed bed is progressively drowned for stream flow, and as lateral and/or local grounding vanish for shelf flow. This can reduce ice thicknesses by 90 % and ice elevations by 99 % along flowlines. Original work presented here includes (1) replacing flow and sliding laws for sheet flow with upper and lower yield stresses for creep in cold overlying ice and basal ice sliding over deforming till, respectively, (2) replacing integrating the Navier–Stokes equations for stream flow with geometrical solutions to the force balance, and (3) including resistance to shelf flow caused by lateral confinement in a fjord and local grounding at ice rumples and ice rises. A comparison is made between our approach and two approaches based on continuum mechanics. Applications are made to Byrd Glacier in Antarctica and Jakobshavn Isbrae in Greenland.


Extract: "Warming in high polar latitudes can, in principle, trigger a succession of positive feedback mechanisms called the Jakobshavn Effect (Hughes, 1986). Buoyancy fraction phiB combines the two dominant mechanisms, reduced ice-bed coupling when surface meltwater floods the bed under an ice stream and reduced ice-shelf buttressing when an ice shelf disintegrates beyond the ice stream. For Greenland, the Jakobshavn Effect would move northward along the east and west coasts, affecting all calving ice streams."

[A-Team]

Here is what can be done with better access to imagery than what we have:

Seasonal and interannual evolution of Jakobshavn Isbrae, Greenland from a 2008-2015 high-res DEM and velocity time series
David Shean, UWashington [Joughin group?]

Greenland’s large marine-terminating outlet glaciers have displayed marked retreat, speedup, and thinning in recent decades. Jakobshavn Isbrae, one of Greenland’s largest outlet glaciers, has retreated ~15 km, accelerated ~150%, and thinned ~200 m since the early 1990s.

Here we present a comprehensive analysis of high-resolution elevation (2-5 m/px) and velocity (100 m/px) time series with daily and monthly temporal coverage.

The Jakobshavn DEM time series consists of over 70 WorldView 1-3 stereo DEMs and 11 TanDEM-X spanning 2008-2015. Complementary point elevation data from Operation IceBridge and ICESat-1 GLAS extend the surface elevation record to 1999 and provide essential absolute control data, enabling sub-meter horizontal/vertical accuracy for gridded DEMs.

Velocity data are primarily derived from TerraSAR-X/TanDEM-X image pairs with 11-day interval from 2009-2015.
These elevation and velocity data capture outlet glacier evolution with unprecedented detail during the post-ICESat era.

The lower trunk of Jakobshavn displays significant seasonal velocity variations, with recent rates of ~8 km/yr during winter and >17 km/yr during summer. DEM data show corresponding seasonal elevation changes of -30 to -45 m in summer and +15 to +20 m in winter, with decreasing magnitude upstream.

Seasonal discharge varies from ~30-35 Gt/yr in winter to ~45-55 Gt/yr in summer, and we integrate these measurements for improved long-term mass-balance estimates. Recent interannual trends show increased discharge, velocity, and thinning of 15-20 m/yr which is consistent with long-term altimetry records.

The DEM time series also reveal new details about calving front and mélange evolution during the seasonal cycle.
Similar time series are available for Kangerdlugssuaq and Helheim Glaciers. These observations are improving our understanding of outlet glacier dynamics, while complementing ongoing efforts to constrain estimates for ice-sheet mass balance and present/future sea level rise contributions.


Spatial Elevation Changes Observed at Helheim Glacier, Southeast Greenland, and Comparison with Kangerlussuaq Glacier and Jakobshavn Isbræ
Carolyn Roberts [Csatho group?]

Mass loss from Southeast Greenland is about 50% of total ice loss from the entire Greenland Ice Sheet with Helheim and Kangerlussuaq the main contributors...

Helheim experienced a dramatic thinning event of 50 m in 2004 with an abrupt cessation in 2006, followed by a sequence of thickening/thinning events of lesser magnitude from 2006 – 2012. This complex behavior is further investigated using multi-sensor surface elevation data over a broad region of the glacier trunk during the period of 1981 – 2012. The onset of the rapid thinning event occurred over most of the glacier trunk rather than at the calving front.

 At the commencement of the rapid thinning event, Helheim Glacier’s calving front appeared to be grounded. This is in contrast to Jakobshavn Isbræ which was approaching flotation prior to its rapid thinning events in 1999.

Additionally, the new dataset reveals curvilinear bands (1 km long, 100 m wide) of 10 m elevation change corresponding to lateral shear zones and tributary confluence junctions.

[A-Team]

Uh-oh. How many hundreds of ice sheet model articles used a seriously wrong form of Glens Law? Almost all of them. How many will be fixed, withdrawn or retracted in the next year: zero.

While these experiments are not easy (compared to hitting 'return' at a model command line), it just goes to show (once more) that theory decoupled from experiment is a colossal waste of time.

Dislocation Creep of Ice At Glaciological Pressures and Temperatures
Chao Qi, U Pennsylvania
https://fallmeeting.agu.org/2015/abstract/dislocation-creep-of-ice-at-glaciological-pressures-and-temperatures/
https://en.wikipedia.org/wiki/Ice-sheet_dynamics#Flow_dynamics
http://tinyurl.com/q6u5nnw and http://tinyurl.com/pobf6kc free full to original 1952 and 1958 articles

The Glen law, a power law between effective strain rate and effective stress of the form Atauⁿ, where A is a [[poorly known]] temperature-dependent parameter and n is the stress exponent of value 3, attributed to dislocation creep, has underpinned models and calculations of glacier flow for over six decades.

Compilations of ice creep data from tests at ambient and elevated confining pressures, however, suggest that dislocation creep of ice is characterized by a value of n=4, not 3. While high-pressure experiments on ice provide the best constraints on the dislocation creep regime and have consistently yielded a stress exponent of ~4, most of these tests have been conducted at much-lower-than-glaciological temperatures.

To investigate dislocation creep of ice at glaciological conditions, we deformed samples at temperatures  264 K and elevated confining pressures up to ~30 MPa, the maximum cryostatic pressure in the ice sheets...Plots of strain rate vs. both peak stress and flow stress yield a value of n=4, consistent with previous data from higher-pressure, lower-temperature tests, from some ambient pressure experiments, and with models of climb-limited dislocation creep (Weertman, 1968).

At stresses <3 MPa, tests on the finer-grained samples show a slight decrease in n to a value below 4, while data for the coarser-grained samples show no such transition, consistent with the onset of dislocation-accommodated grain boundary sliding (GBS) in the finer-grained samples at low stresses.

Our data demonstrate that Glen's law is an average of two creep regimes – dislocation creep proper, with n=4, and dislocation-accommodated GBS, with n about 2 – and fails to accurately describe the rheological behavior of ice over the broad range of strain rates in natural ice bodies.

[Neven]

There's a great article up on the Washington Post website linking to the ASIB blog post on this, with reactions from Alley, Box and Rignot.

[Lennart van der Linde]

Neven, did you change your name from Akropolis to Curlin (as the WP-article calls you)?

[Neven]

Neven, did you change your name from Akropolis to Curlin (as the WP-article calls you)?

No, Acropolis is/was my pseudonym. Don't tell anyone.

[Lennart van der Linde]

Ah, i see...

[Espen]

There's a great article up on the Washington Post website linking to the ASIB blog post on this, with reactions from Alley, Box and Rignot.

Thanks Neven,

1. Typical scientist reaction from one corner.

2. How big is a calving, and how long a break before a new calving is starting?

3. Why on earth dont they pool part of their respective budgets to fund some serious surveillance equipment up there?

[Neven]

2. How big is a calving, and how long a break before a new calving is starting?

3. Why on earth dont they pool part of their respective budgets to fund some serious surveillance equipment up there?

I left a comment echoing your sentiment, Espen:

"Thanks for the link, and thanks for asking the experts about this, great article. If people want to know what a calving looks like in 3D, here's an excerpt from the Chasing Ice documentary:
 
We obviously need to develop rules and definition for what constitutes a calving. In what timespan? 24 hrs? 48 hrs? Oh, and a dedicated satellite that just hovers over Jakobshavn Isbrae and a couple of drones, just in case. 
 
All kidding aside, this was yet another big calving (it's difficult to compare as there isn't as much info on individual calving events as there is on the glacier and its retreat in general), the retreat line has moved to a new position, and might move some more in the next couple of weeks."

[A-Team]

I dunno, the article was lame, creating controversy when there is none. The right question to ask: "does this huge event herald a new stage in the world's fastest glacier?" All these open source animations and the reporter just pastes up irrelevant stock photos.

Curlin could be just another round of aliasing ... suspiciously close to Canadian shuffleboard played on an ice sheet.

Holland's site, as I noted above in #877, could settle the time structure if the web cams had been working which they currently are not. No one has ever tracked volume calved vs time as far as I know. The easiest way to do that is integrate velocity over time just up from the calving front as distance moved is a proxy for volume (as D Shean does at Dec 15 AGU post #867 above).

But there the problem is the sketchy monitoring. As Espen says, they have not invested diddly into decent surveilance equipment. They come up to get a paper's worth, turn around and go home mid-season. This is going to look really really stupid if this turns out to be a historic moment when it all began to unravel.

The Aqua, followed by Terra 5 hours later (?), may show the lobe deepening but that is uncertain.

Holland's site, as I noted above, could settle the time structure if the web cams had been working which they currently are not. No one has ever tracked volume calved vs time as far as I know. The easiest way to do that is integrate velocity over time just up from the calving front as distance moved is a proxy for volume (as D Shean does at Dec 15 AGU post #867 above).

But there the problem is the sketchy monitoring. As Espen says, they have not invested diddly into decent surveilance equipment. They come up to get a paper's worth, turn around and leave mid-season. This is going to look really really stupid if this turns out to be a historic moment when it all began to unravel.

The Aqua, followed by Terra 5 hours later (?), may show the lobe deepening but that is uncertain.

[Rubikscube]

Fine, stay home then and read the adventure blog. WorldView for 25 Aug 2012 below.

Hehe, I certainly won't be dragging my sorry ass into that minefield, not sure your helicopter guy will agree to ditch you down there either, even if you pay him ever so much (like I said this is not Mexico).

Say whatever you want about Google, but they do actually got street view from the outer island towards the south branch, they have even bothered to walk up several "streets" as well, and from what I can see, the main ice stream remains pretty much visible all the way up to the bend (the light makes it a bit difficult to judge). A camera there should be able to monitor the calving front for many years from now given that JI doesn't start retreating at absurd speeds.

Back to the volume vs elevation plot, I promised an estimate, but I'm not able to draw even a crude 2000 m line in their elevation plot. I will see what I can do with the cdf, but most likely I'm too much of a programming noob to make any sense of it.

[A-Team]

1. Typical scientist reaction from one corner.

Really. Staking out the middle ground, not go a micron beyond the data?  Box, Rignot, Hansen have a better idea -- it's past time to get the message out and stop the equivocating blather.

The fact is, this event was wholly without precedent in terms of volume of grounded glacier calved at the key geometric sill that is so influential a driver of future retreat. I've looked at every Landat-8 ever taken of Jakobshavn -- nothing else comes close. The burden of disproof thus falls on those who haven't looked at the data. In terms of time-stamps, consecutive Modis shots define a 24 hour window, shorter than the 48 hour Landsats.

Why would I want to compare this event to tabular calving volumes of a long-gone floating ice shelf 20 years ago that don't contribute to sea level rise? It's all about where this event occurred and what happens next (if anything).

There are three ways of looking at the Dec 2015 AGU abstracts I collected above: 

-1- Safe to ignore, it's not published yet, wait for eventual IPCC consensus, more studies are needed;

-2- Isn't it great how science just marches on, smoothly progressing to ever greater understanding!

-3- Wtf is with these clowns, how did they get it so dead wrong for so long?!%*?

I am in the third camp, being very familiar with meeting abstracts/posters and what they mean and do not mean. Here  the specificity (level of detail) is very alarming. We had no idea whether Jakobshavn sat on bedrock, till, sediment, or  meltwater. We spent 22 years on radar flyover without ever really getting down to the bottom of the elbow. We dinked around not observing the interface of glacier front and ocean as if sonar imaging had not been invented in WWII. Seasonal volume discharge integration, who knew? Model after model deforming the glacier with the wrong Glen-Nye flow exponents and temperature profile.

All that's changed now -- see posts #881, #883, #884. And it could have been done years and years ago.

[sidd]

Didnt Schoof have a glen's law exponent of 4 way back in 2007 ? or someone else had evidence that it was closer to 4.

[Neven]

I want those pants! Brush, baby, brush!

[TenneyNaumer]

Hey Espen, at least it is not going to be  Jakobshavn Isbræeren -- LOL!

[TenneyNaumer]

A-Team, I noticed that white line sticking out of the latest image of the south branch -- it's just a contrail, isn't it?

[nukefix]

The southern side of the fjord has a good view of the calving front as Google street-view shows and it would be a good place for more instrumentation for understanding calving processes. However I'd monitor the calving front from space, now that S-1 IW every 12 days is virtually guaranteed this can be done year round and perhaps in the future we'll get images even more often once the 2nd satellite in the constellation is up. Good times!

[skanky]

EOB - http://earthobservatory.nasa.gov/IOTD/view.php?id=86436&eocn=home&eoci=iotd_readmore

[Neven]

Thanks, Skanky! Wow, that image comparison tool is so beautiful.

[A-Team]

monitor the calving front year round from space, now that S-1 IW every 12 days is virtually guaranteed.., even more often with the 2nd satellite

Yes, the Sentinel program is a huge step forward in terms of really systematic year round monitoring of Jakobshavn. At Petermann, 81ºN, there can be 3-4x daily coverage by Landsat, Sentinel, Modis and NOAA. However available daylight limits non-radar satellites to day 33-268 and in summer clouds can be present for weeks on end. Oddly coverage at Zachariae/Niog at 79ºN is far less frequent.

Didn't Schoof or someone have a Glen's law exponent of 4 way back in 2007?

sidd, I edited cites and verbiage from all the AGU abstracts; these provide more Glen flow history. It takes a very specialized type of laboratory to make physically relevant experimental measurements. The 'n' is discussed fairly frequently, the poorly characterized 'A' and its temperature dependence are usually passed over. The sensitivity of model outcomes to their variation is occasionally explored.

The interesting part for me is where Chao Qi concludes Glen flow is a superpositioning of two creep regimes: dislocation creep proper with n=4 and dislocation-accommodated grain boundary sliding with n about 2. Otherwise "it fails to accurately describe the rheological behavior of ice over the broad range of strain rates in natural ice bodies."

that white line sticking out of the latest image of the south branch -- it's just a contrail, isn't it?

Good suggestion. The white line can also be seen in the second image though it's less noticeable (dissipated?). It seems out of kilter with other cloud forms in the image. We've also seen elongated smoke plumes at this latitude. Given the low resolution of these images, further analysis seems infeasible.

image comparison slider tool is so effective.

Most be stock code, I see it on other sites so wonder if WordPress could support it. In fact one of the bloggers here had it going on a personal site, am recalling artische penguin. I'm skeptical though that it is as transportable as an animated gif. It may just be an active alpha transparency channel over a two image stack.

Actually, what's even more effective is to load the images (as many as you want) into an ImageJ "stack". This has a very very rapid slider action over the entire image.

This is great to see a NASA science writer digging into the forum rather than just news page. They seem to have some secret sauce that produces far more realistic color than anything served over at Earth Explorer. Something more than just pan-sharpening with band 8 or 731 color; we do that already.

These NASA Earth Observatory images were made by Joshua Stevens from the USGS Landsats but it doesn't say how. The originals are 17481 × 17581, meaning only 1/34th was processed. I've written him to see if the method is something we could do ourselves.

Their processing is really fantastic at distinguishing clear water, till water, muddy water, algal lakes, berg-filled meltwater and wet areas from nunataks and similar. The third image highlights melt within the south branch elbow via color inversion.

For convenient reference, those 3000x3000 pixel 15 m images without the caption overlays are stored at:

http://eoimages.gsfc.nasa.gov/images/imagerecords/86000/86436/jakobshavn_oli_2015228_lrg.jpg
http://eoimages.gsfc.nasa.gov/images/imagerecords/86000/86436/jakobshavn_oli_2015212_lrg.jpg
http://earthobservatory.nasa.gov/IOTD/view.php?id=86436&eocn=home&eoci=iotd_readmore (story)