As a tightly wound cyclone drifted over the Polarstern this morning, near-surface conditions went from 1 m/s under the eye to 16 m/s in single hour accompanied by a 180º shift in wind direction, then reached an expedition peak of sustained 19-20 m/s winds. (In other units, 20 m/s is 77 km/hr, 38.9 knots, 44.7 m/h).
Note the extraordinarily warm near-surface air temperatures in the early am: -11.5ºC. GFS has this at 11.0ºC. FoMo goes on with
crazy talk about this warmth coming up from Svalbard despite the wind there blowing straight down the Fram. FoMo ignores the storm raging outside to share more
trivia on microscopic plankton and cod minnows.
Better to have saved the life boat story for today? Some 90% of ship emergencies are due to fires on board. The captain has discussed the many preparations for that. Despite off-loading gear to the ice, many tons of untested electrical devices from 19 countries remain aboard.
Should the ship list seriously to port because of impinging starboard pressure ridge ice (on the Fram, these got higher than the mast), it won't be possible to launch half the lifeboats using eight guys tugging on a rope around a dislodged mooring I-beam. The other lifeboats might launch adrift in a corresponding port lead and so be inaccessible. Would it make sense to store some well-stocked lifeboats on the ice the whole time?
Without the Pistenbullys, an emergency airstrip could not be groomed. Indeed, we've not seen the promised follow-up on the trial airstrip of Jan 8th. It must have become totally inoperative within days from the daily ice pack grinding seen on bow radar; it would take an amphibious 4WD ATV to go down the runway now. This scenario could leave 150 people doubled up on half rations inside the functioning lifeboats for a month or more.
2020 AWI is recruiting 17 PhD candidates and 12 Postdocs to strengthen its new research program "Changing Earth – Sustaining our Future". Applications ar now open in the following research areas: atmosphere, ocean, ice and land; Arctic ecosystems and permafrost, marine and polar life marine bio-economy. Over €1 000 000 000 has been allocated to build Polarstern II.
How about €50/day to pay a science grad to oversee daily FoMo blurbing? Climate change --> no fisheries bio-economy in the Arctic or anywhere else after the ice melts, why throw resources at it.
The sharp curvature and high (and rapidly changing) gradient of the wind stress field will couple strongly to edges of pressure ridge jumbles and giant sails like the Polarstern side. The ship's orientation (bearing) is not disclosed so we don't know when the winds hit broadside or astern; there's no S1AB coverage today.
Pitch and roll are not disclosed either, though even the min-sub has a nice heads-up display of them. We can see from swings in the bow radar contrast and orientation that the ship is not really frozen-in but merely moored on the starboard side.
This is by design as the hull must lift up (rather than collapse) if pressured by ice on the side. The draft as currently loaded is ~4.0m so most of submerged parts (eg keel and propeller) lie in water too warm too freeze. An early FoMo photo out from the moon pool confirms this.
These high wind speeds will put record stress on the icepack. However today's bow radar will only cover the fracturing until midnight of Jan 31st; the event is continuing through 1500Z today. The drift is strongly to the Siberian south and slightly west ('to the right' of GFS wind as Nansen noted).
In the awiMET table, when their one decimal latitude and longitude "hold steady" despite sustained strong winds from 330º, that means the Polarstern is rapidly moving due south towards the New Siberian Islands. This will eventually show up as 87.3 latitude. (The Polarstern drifts about 7 km/day, 0.29 km/hr on Uniq buoy scale, under local winds speed averaging 6 m/s or 22.6 km/hr.)
Only minor plate readjustments on the port side have occurred so far, raising the question of time lag between applied stress and brittle failure. However a 4.2º CCW rotation of the bow took place in the 12 hours before midnight on Jan 31.
87.4 95.9 20-02-01 18:00 08 330 -38.2 1000.7 third radar frame tomorrow
87.4 95.8 20-02-01 17:00 09 330 -37.8 1000.1
87.4 95.8 20-02-01 16:00 11 330 -37.6 998.9
87.4 95.7 20-02-01 15:00 11 320 -37.1 998.1
87.4 95.7 20-02-01 14:00 11 320 -36.2 997.1
87.4 95.7 20-02-01 13:00 12 330 -35.4 995.6
87.4 95.7 20-02-01 12:00 13 330 -34.7 994.4 second radar frame tomorrow
87.4 95.7 20-02-01 11:00 15 330 -33.9 992.6
87.4 95.8 20-02-01 10:00 16 330 -33.2 990.9
87.4 95.9 20-02-01 09:00 16 340 -32.8 989.1
87.4 96.0 20-02-01 08:00 18 340 -31.6 987.3
87.4 96.1 20-02-01 07:00 19 350 -30.9 985.0
87.5 96.2 20-02-01 06:00 19 350 -29.0 982.7 first bow radar frame tomorrow
87.5 96.2 20-02-01 05:00
20 360 -27.6 979.8
87.5 96.3 20-02-01 04:00 19 360 -25.7 977.3
87.5 96.4 20-02-01 03:00 16 010 -21.6 975.2
87.5 96.4 20-02-01 02:00 01 300
-11.5 974.6 gap of one hour
87.5 96.4 20-02-01 00:00 07 190 -12.0 976.4 last bow radar frame today
Very nice amsr2 arctic lead product.
Very nice indeed! Kaleschke has been publishing on AMSR-E (passive 18.7, 89 GHz, since 2002) for 13 years including lead detection so this is not out of the blue -- but what sudden inspiration struck him there on the wharf? Hopefully updates will keep coming while at sea; posting time changes suggest a pipeline is not yet fully set up.
This will get us into
Hough transforms, a familiar technique from image analysis used to detect sub-pixel geometric features such as linear leads. ImageJ plugins are totally on top of this.
Hough was automating bubble chamber decay analysis in the 1950's. In math, these have been known as
2D radon transforms from 1917 on. The inverse is used in medical scans.
In the case of leads, a time-dependent
2D+T Hough transform might sharpen the product. That is, OsiSaf describes basin-wide how ice is moving, meaning the day n+1 AMSRE can be brought back to augment the search of leads on day n. (For all we know, this could have been Kaleschke's processing breakthrough.)
Regionally, ice motion is available from consecutive S1AB pair overlaps which generally exist for the Polarstern area. Locally, the buoy swarm gives ice motion very accurately at 30 minute intervals; leads can be extracted from dispersion in the delaunay diagram (the GPS is precise, ice does not stretch). However the scale of buoy deployment is well below that of AMSRE.
Pressure ridges don't show up on that satellite but can be inferred to exist when a lead closes, provided that can be distinguished from freezing over. Tracking these features might identify multiple openings over time of the same weakness.
The Feb 1st lead frame is below, slightly modified. Note how the view is cut off well east of Banks Island and much of the Beaufort/Chukchi; the detectable leads detected are almost entirely in MYI.
The base map is the same projection but different scale as forum AMRS2. However that can provide convenient land and open water masks and a graticule at 124.98% enlargement. Note the pole hole is slightly larger in AMSRE-K but still well inside 89º. Ascat then needs to be resized by 2.2827 * 1.2498 = 2.8592 to take a dark or colored overlay from what we think is AMSRE-Kaleschke.
However even if the latter is cropped to the 'good stuff', it is still 770 pxls wide, whereas the forum limit is 700 so the above enlargements need to be dialed back by 10%. Assuming only MYI is left at the time of mooring on Oct 4th, it may make sense to use the AMSR2 open water to mask AMSRE-Kaleschke on all the latter dates since the leads (and Polarstern) are in the MYI and drift within the frame.
On the bright side, the 90º Eurasian quadrant of the Polarstern provides 350x350 pixels of AMSRE-K lead resolution, meaning it probably can withstand 2x enlargement or double up as is with 3 other resources as a 700x700 (alt 4:3 or 16:9 aspect) mp4.
Care must be taken with repeatability of crop boundaries, enlargements and contrast enhancements so that incoming Feb updates to the leads can be easily concatenated onto the current product.
Sea ice leads in the Arctic Ocean: Model assessment, interannual variability and trends
Q Wang, S Danilov, T Jung, L Kaleschke… - Geophysical …, 2016
Wernecke, A.; Kaleschke, L. . “Lead detection in Arctic sea ice from CryoSat-2: quality assessment, lead area fraction and width distribution”. The Cryosphere 9. (2015): S. 1955-1968.doi:10.5194/tc-9-1955-2015
Röhrs, J.; Kaleschke, L.. “An algorithm to detect sea ice leads by using AMSR-E passive microwave imagery”. The Cryosphere 6 (2). (2012): S. 343-352.doi:10.5194/tc-6-343-2012
Röhrs, J.; Kaleschke, L.; Bröhan, D.; Siligam, P.K.. “Corrigendum to "An algorithm to detect sea ice leads by using AMSR-E passive microwave imagery" published in The Cryosphere, 6, 343–352, 2012 ”. The Cryosphere 6 (2). (2012): S. 365-365.doi:10.5194/tc-6-365-2012
Gunnar Spreen, Lars Kaleschke, Georg Heygster, "Sea ice remote sensing using AMSR-E 89-GHz channels", Journal of Geophysical Research, vol. 113, 2008.