New ice thickness map of the Arctic unveiled

From the ESA webpage:

The first map of sea-ice thickness from ESA’s CryoSat mission was revealed today at the Paris Air and Space Show. This new information is set to change our understanding of the complex relationship between ice and climate.
 
From an altitude of just over 700 km and reaching unprecedented latitudes of 88º, CryoSat has spent the last seven months delivering precise measurements to study changes in the thickness of Earth’s ice.

Check out their first map of sea ice thickness data (for Jan/Feb 2011) that they made to show what will eventually be the deal:

I want to get this out quickly, so I haven't checked yet to compare this image with the ice thickness models. But I will at some later point.

It's important to stress that however exciting all of this is, we're still some way off of 100% accurate data. Calibration will be an ongoing issue and as CryoSat-2 has now gone into its second year of data delivery and a substantial time series is being built up, its true value will start to emerge more and more.

Hopefully one organisation or other will then start to provide graphs and maps for us lay people to drool over. But we have to remain patient, because it will take some more time. The good news is that the satellite is operating so efficiently that the mission will probably be prolonged until 2017!

I was really looking forward to watching the ESA webcast today, but unfortunately I slept through it because I had to work all night to make a deadline. The video of the presentation can be watched here. I highly recommend watching it. It has also been uploaded to Youtube:

 

Jonathan Amos from the BBC was present (article) and he interviewed CPOM researcher Dr Katharine Giles:

Comments

[cynicus]

Very exiting news indeed. On first glance I notice thinner ice near de East-Greenland coast compared to more offshore. Is that to be expected?

[Ennis George]

I am surprised how "thin" the sea ice is. Most of the Arctic is less than 3 1/2 meters thick. I would have thought it would be more but I confess I have no expertise or particular knowledge about the sea ice. However, I suspect that such ice will not stand well particularly if it is discovered the ice is melting from below.

[Lord Soth]

Im getting the impression that the numbers are high.

Didn't a recent survey give results of 1.4 meters on average for the beaufort sea. From the cryostat numbers, it shows readings from 2 to 3 meters for the same area.

I guess their is need of some ground truthing of the cryostat numbers to improve accuracy, but the ice map is an excelent start.

[Janne Tuukkanen]

Wasn't it the aerial survey by Alfred Wegener Institute early May? Melting season was on its way by then, and the satellite image is from Jan/Feb. I would be surprised if there were huge discrepancies, because those measurements were the very ones used for Cryosat-2 calibration.

[Daniel Bailey]

Don't make any judgements based on this particular graphic. The granularity is simply too high, especially for a product with such high resolution.

This is simply a demo/beta to show what the platform can deliver. Expect future product rollouts with greater resolution and accuracy, once all of the bugs and kinks are ironed out.

The bottom melt accounts for about half of all sea ice melt.

[toby]

Compared with Neven's analysis is March, this map does show thicker ice in comparable places - but this map is earlier, for Jan-Feb.

http://neven1.typepad.com/blog/2011/04/ice-thickness-models.html

However, I take Daniel's point that it is early days. Great to see this product .. I have been looking forward to this for a year. Looks like a winner!

[Peter Ellis]

I'm very surprised that the scale only goes down to 2 metres, even at the edge of the ice pack - that seems a little un-physical. Maybe the marginal zone at the edge of the pack is too narrow to show up on this scale?

Alternative explanation is that they've just set a cutoff to only display ice over 2m (maybe they're still calibrating the data for thinner areas). That would explain the white margin along the Beaufort sea coast. Remember that the 1.2m figure from the Alfred Wegener institute specifically referred to the first-year ice in this area (i.e the fringe along the coast). Most of the Beaufort is a mix of older ice with first-year stuff glueing the floes together - presumably Cryosat data will average these together, at least on the scale shown above.

[Daniel Bailey]

Consider one potential, unforeseen, issue with Cryosat-2: It's too precise for a highly mobile ice pack. The design tolerances give a very precise reading, but stitching together the flight "footprints" of an icepack that is continuously shifting about and moving and melting and refreezing is probably playing havoc with their calibration efforts.

I anticipate future deliveries to "tighten up" the resolutions shown as the smoothing algorithms are tuned to eliminate the "blur" of the flight overlap zones (which will be substantial, given the high latitudes).

Shorter answer: Cryosat-2 is near-sighted and needs an eye adjustment from the eye doctor.

[dorlomin]

2m ice in the Baltic? Hmmm thats interesting.

[cynicus]

dolormin, where do you see the Baltic sea on the Cryosat image?

[fredt34]

Neven, you're supposed to know this: vacationers coming back hit deadlines. Another reason not to leave...

Man, you're amazing, I had got no information about this presentation.

It's nice to see that Cryosat actually works!!

[toby]

The map does not show the Baltic - I think that is Novaya Zemlya where the ice is at 2m thickness. You can just see the top of Scandanavia at the bottom of the map, above & to the left of the letters CPOM etc.

I think I also boobed above ... of course the ice should be thicker in March because that is the maximum.

[idunno]

Hi all,

Well it seems to show a similar pattern to the final maps I could generate on TOPAZ version 3, for 24 Feb, which is still available online @

http://topaz.nersc.no/

... with the important difference that the Cryosat peak thicknesses are at least 1 metre thicker, maybe 2 metres thicker.

This is a big surprise, particularly after the air flights were reporting 1.4 m thickness in Beaufort in May. And I think that ice thickness usually grows between January and May. The PIOMAS volume averages sinewave certainly peaks in April...

[Janne Tuukkanen]

Data from the AWI expedition is here:

http://www.awi.de/en/research/research_divisions/climate_science/sea_ice_physics/expeditions/pamarcmip_2011/daily_data/calendar_week_14/

[Enno Zinngrebe]

A) Janne - these AWI expedition weekly data do not seem to contain thickness data or I cant see them. Even though they said that was the focus of their effort. But I recall that the AWI post expedition press releases did talk about 1.3-1.7 m thick first year ice. Whereas the Cryosat map shows 2.x m for the same area. But somewhere in the Cryosat press conference video they say explicitly that their data was checked against an airborne measurement of the same area (Beaufort Sea) and that that airborne measurement was 2.7 m thus right agreeing with the satellite data. What the reason for the discrepancy is I have no clue but maybe what AWI talked about was first year ice only selected in an area and what Cryosat measured was all the ice on average in an area? Just a guess.

B) Hidden in the video in Neven´s post is actually an interesting part that directly answers a discussion that several people had some days back on these pages - namely the relation between sea level rise from melting and counteracting isostasy. They explain that indeed, isostasy acts immediately, but that more importantly, a large mass of on-land ice attracts water to it, and that the melting of the ice sheet would mean that this gravitational attraction towards Greenland stops, letting the water flow away, lowering the local sea level. They show a map saying that in fact this gravitational release effect more than counterbalances melt-related SLR in the local Greenland area, leading to a net *sinking* of the sea level around Greenland due to its ice sheet melting!

I thought this should be interesting to some people here ...

[dorlomin]

dolormin, where do you see the Baltic sea on the Cryosat image?

Posted by: cynicus

I didnt I mistook an island for the Scandanavian Peninsula

[Werther]

Great to see reporting out of Cryosat-2. But the actual representation for jan/feb 2011 doesn’t seem to fit to any model, assumption or on-site measurement (at first glance). The team may say it is calibrated well, but they will have to come up with much more resolution in the representations.
For my counting area north of Greenland, I’d have to go for a metre more (3,5 instead of 2,5) on the floes, and 2,1 m more for the leads (there's no difference..). Of course, there is bottom/surface melt since day 150. But it seems impossible to melt down a metre in twenty days.
As for now, I agree with Daniel’s remarks. I think the Cryosat-2 team is going through a daunting task calibrating an ever moving pack. The sat is pushing down a multitude of readings, hardly allowing time for working out any representation whatsoever. Just look at the leads in my research area; it isn’t a coherent layer, in place nor in time. It moved 35 km south in the last 9 days. It’s gaps, leads and polynia’s changed 2% in area (larger..) in the same time. The rubble in the leads consists of millions of shattered, tumbling, stacking bits and pieces. I wouldn’t be surprised if the Cryosat-2 mission has it’s best shot at the glaciers and icecaps...

[Gas Glo]

The area shown on that map plus Canadian archipelago must have area of at least 10 million km^2 (we currently have some coverage of Hudson but that must be made up for by having less extent in Chukchi, Barents, Kara ...)

The green and above colours seem to have larger area than purples so average thickness on that map must be at least 3m. So if the map is taken as accurate then the ice volume must be at least 30K Km^3.

PIOMAS has volume at end of February as volume at 19.9K Km^3 (average of Jan/Feb is rather less).

A difference of over 10,000 Km^3 seems quite significant.

[Anu]

Excellent - good start.

Hopefully the Centre for Polar Observation and Modelling will get a nice publicly accessible interface up and running before things get really interesting this September.

Here's the same map with the "extent of multi-year ice" outlined in black.

I'd like to see daily total Arctic ice volume values in this interface too ( with the polar projection handled correctly, unlike at WUWT :-) )

[Rob Dekker]

Great that Crysat is finally producing data on ice thickness. But I share the opinion here that this map seems to show much thicker ice than what in-situ measurements have shown.

Neven even did a post on these in-situ measurements :
http://neven1.typepad.com/blog/2011/05/ice-thickness-models-2.html#tp

Where the Cryosat validation team reports : April 14, 2011, 85.6 N 69 W, 1.8 meter ice.

This location (just north of Greenland) fall right into the green color on the map showing 4 meter ice.

Any way, unless somebody can explain how Arctic sea ice can melt a good 2 meters from January to April (while it still freezes crazy), the Cryosat team / CPOM have some explaining to do.

Until then, I think I put more trust in PIOMAS, which at least seems to match better with in-situ measurements.

[Rob Dekker]

Neven, once again, my deep respect for your awesome reporting on developments in the cryosphere. We have all been look forward to this data that Cryosat 2 would finally release ice thickness data.

I'm not sure about you, but I'm really baffled by this Cryosat map, and its inconsistency with PIOMAS as well as the in-situ measurements done by the Cryosat validation team and the Alfred Wegener Institute.

I recall your post on the Cryosat validation measurements :
http://neven1.typepad.com/blog/2011/05/ice-thickness-models-2.html#tp

There, you mentioned that you talked to CryoSat validation manager Malcom Davidson.

Could you possibly send him another email asking for clarification of the presented map ? I would suggest sending him something like this :

---
First of all, it is exciting that the Cryosat team finally presents sea ice thickness numbers from Cryosat's data and validation results.

However, there seems to be a substantial difference between the validation numbers and this sea ice thickness map. For example :

http://blogs.esa.int/cryosat-ice-blog/2011/04/15/getting-off-the-ground/

April 14, location 85.6° N 69.8° W, ice thickness measured : 1.8 meter.

This same location (just north of Greenland) shows 3-4 meters of ice in Jan/Feb.

Does the Cryosat team and/or the validation team care to explain how the ice could melt in half from Jan to April ?

Or is there another explanation for this significant inconsistency between in-situ measurements and the presented thickness map ?
---

If you agree with the text above, could you possibly sent this to Malcom Davidson, and post his reply ?

Because as it stands, this map makes no sense at all.

[Neven]

Rob, I've been discussing this with a few people through mail. Everyone says the same (and it took me a while to get it): More data is needed for a longer time series, calibration is an ongoing process. So we need to be patient for a while longer.

But I agree with you that there are some inconsistencies on this map for Jan/Feb 2011. We'll see how it pans out. They obviously wanted to show something at the high profile Paris Air and Space Show. Which is perfectly understandable, of course.

However the definite thickness maps turn out (either confirming or disproving models, or something in between), I'll be extremely happy to have more certainty when it comes to thickness and volume.