pronouncements
I'm equally baffled by other comments above. If folks are not looking at incoming satellite data, what are they using -- an undescribed private model that works off a dodgy long-term weather prediction? If just ideology + hunches, this might not be the appropriate forum.
Back at the Beaufort, floe blueing is progressing but contrary to statements above, not one floe has been identified that changes one measurable iota in size as of 28 Jun 16. Without any physical data on current floe thickness, internal temperature profile or under-ice conditions, how are folks able to determine time-to-melt? They aren't.
To compare the Beaufort on same dates in 2012-2016, please consider 5 clicks of the mouse at Worldview
prior to posting.
To compute volume by pixel color counting ratios, recall NSIDC products including MASIE are in polar stereographic which is conformal, not equal area. You would have to weight each successive latitudinal ring separately yet rectangular image pixels don't sit comfortably there. The error is then very unevenly distributed making year-on-year (or even day to day) comparisons problematic. You would have to go back to netCDF raw data because the resolution provided does not support map re-projection.
The new MASAM2 product noticed in #2642 and #2671 is extremely interesting. We should probably start a separate forum for it and add it to the daily graph section as there is published evidence it improves the ACNFS product line.
It is a blend of AMSR2 10 km with multi-source expert-annotated MASIE 4 km. The data back to 2012 is bundled into monthly .nc file sets but offered more conveniently as a very satisfactory daily.png of 2100 x 2550 pixels at 4 km ground resolution.
The Arctic Ocean requires 785 x 815 pixels; the image needs a 45º CW rotation (aka Greenland down) to overlay a WorldView. For animations we would like to get it down below 700 without resizing.
Astonishingly, the palette colors correspond exactly to map colors, there being compression blunders and processing disconnects in 99% of cryosphere maps we see.
The test is to apply a stringent color picker to a palette square and re-color the ice pink (or whatever). After picking through the entire palette, the entire map should be colored pink with no ice left over. If you counted pixels as you went, you get a first approximation of sea ice concentration areas in each bin (but see above).
However as you can see from the image below, there is one shade of blue left-over, ie they mistakenly left off one of the ice concentration classes, the one for 65% to non-zero below. (Note also the second palette block is a peculiar size.)
https://nsidc.org/data/docs/noaa/g10005-masam2/ product description
ftp://sidads.colorado.edu/pub/DATASETS/NOAA/G10005/ data archive 10 May 16 to 28 Jun 16
Improving Arctic sea ice edge forecasts by assimilating high horizontal resolution sea ice concentration data into the US Navy's ice forecast systems.
PGE Posey et al Cryosphere 9: 1735-1745 doi:10.5194/tc-9-1735-2015.
http://www.the-cryosphere.net/9/1735/2015/tc-9-1735-2015.html free full, 3 on-topic cites