I asked Dr. P Chang of NOAA today why so many swaths on ASCAT have been blacked out for 4-5 days (similar to but different than AMSR2).
He said there were no problems with the satellite nor its instrument but that "it’s a network connectivity issue between eumetsat and NOAA...instrument is still healthy....
If we have the data we can recreate the images...I asked my folks to do that".
The Eumetsat home page that operates this satellite is serving a truly crappy dumbed-down version of the data that cannot serve as replacement for NOAA's secondary version.
How long did Eumetsat plan to let the glitch run before fixing the network connectivity issue? Does Eumetsat plan to repair the damage by resending the data? How long has NOAA been aware of this problem prior to my email?
Meanwhile, all these satellite time series have glitches from time to time. I've explored various technical fixes in the past, such as layering up the defective image over the previous day, deleting the bad swaths to transparency using a color picker, capturing show-through along with the good parts of the bad image, and replacing it.
There's a cross-platform variation on this that stubs in a grayscale from another satellite for the missing pieces, say Jaxa, which so far has been unaffected. That is doable but much more complex because of scaling, swath timing and correlative tonal matching. It does put in real contemporaneous data unlike putting the previous day but then again it's not scatterometer data.
*/*/*/
Update: NOAA was able to repair days 111 and 112 but not the three previous. Day 113 is not available yet, about 6 hours late. [Update: it has appeared, about 18 hours late. Update: day 114 arrived on schedule, minor problems with it indicating the "network connectivity issue between eumetsat and NOAA" is still not repaired.]
Actually very little has happened in terms of ice movement.
Scrolling out five days in nullschool GFS pressure and wind, a slight low cuts itself off from a larger Siberian feature and drifts aimlessly across the central Arctic. Winds will be incoherent at a reglional level except westward along the Alaskan coast.
OSI-SAF has not been able to repair their 3 blanked-out days, 10-12 April 2018. This may have a different explanation.
http://osisaf.met.no/p/osisaf_hlprod_qlook.php?year=2018&month=04&day=09&action=d%2B&prod=LR-Drift&area=NH&size=100%25Checking UH AMSR2 to see if they have significant glitches, the answer is no for the month of April. Black-out wedges do occur from time to time; the archive is never repaired by management. These are largely peripheral and not as damaging as changes in pixel dimensions.
My theory is no one but us ever looks at the data as it comes in. It is all robotic pipeline from satellite to web archive. Once in a great while, maybe a researcher goes back and uses the archived imagery for a paper. Only then are gaps noticed. By that time, repairs won't be possible any longer.
There's too much reliance on robotics; it makes no economic sense given the expense of the satellite, the cheapness of an intern or elderly volunteer, and the value of a long time series.