As many readers have noted, one of the Arctic sea ice extent plots on our WUWT sea ice page took a Serreze style nosedive today:
According to DMI (Danish Meteorological Institute), this is the source of the data:
The ice extent values are calculated from the ice type data from the Ocean and Sea Ice, Satellite Application Facility (OSISAF), where areas with ice concentration higher than 30% are classified as ice.
Data used: SSM/I (DMSP F15), ECMWF forecast for atmospheric correction
The glitch is reminiscent of the Feb 2009 failure of an SSMI sensor used by NSIDC.
Click for larger image
A couple of days later they were forced by the failure of the sensor to take their data offline, so apparently it was worth blogging about after all.
They wrote in the press release at the time:
Last year, F13 started showing large amounts of missing data. The sensor was almost 13 years old, and no longer provided complete daily data to allow us to track total daily sea ice extent. As a result, we switched to the DMSP F15 sensor for our near-real-time analysis.
And as noted above, DMI uses SSM/I (DMSP F15), the same as NSIDC. Is this glitch worth blogging about? I think so since NSIDC was unaware last time that a problem had developed until we pointed it out for them.
This looks like the beginning of the problem on August 6th, as seen at the OSI SAF page:
The day before on August 5th:
It may be related to the three Coronal Mass Ejections, (CME) that hit Earth about that time. From Spaceweather.com
Earth’s magnetic field is still reverberating from a CME strike on August 5th that sparked one of the strongest geomagnetic storms in years. Registering 8 on the 0 to 9 “K-index” scale of magnetic disturbances, the storm at maximum sparked auroras across Europe and in many northern-tier US states.
It is possible the satellite operator shut down the bird for protection, but nobody got the memo. There’s no mention of data outages on NSIDC’s page or at CT or other ice product websites that I’ve found. Or, the sensor data might be so corrupt as to be unusable, or the sensor has been fried by the CME.
So, like before, I’ll send NSIDC’s Dr. Walt Meier a courtesy note on this one and see what he says. NSIDC’s plot averages over 5 days, IIRC, so it won’t show up for a few days and they have time to correct it if in fact it is the satellite sensor data again.
This may be a sensor issue, or it may be an algorithm issue. Since other plots aren’t showing it, we know it doesn’t represent a real loss of ice, just loss of data.
Curiously though, I’ve noted another glitch half a world away:
Which looks to be unrelated, since it is the AMSR-E sensor on a different satellite.
Must be the day for glitches in sea ice.
Must be a sea ice glitch of a different kind.
Dr.Walt Meier of NSIDC responds:
This is quite clearly a data issue. We don’t work with the F15 satellite
anymore – we’ve been using the sensor on the newer F17 satellite, so I
can’t say if it is a a sensor problem or a processing issue at DMI. I
could be the CME, though it doesn’t seem to have affected the F17
sensor. From the image, it looks to be a missing swath of data, perhaps
from CME, perhaps from some other issue. A missing swath is not
particularly unexpected. Sometimes the data can be recovered later and
added in, sometimes not. The AMSR-E issue in the Antarctic also appears
to be due to one or more missing swaths of data on Aug. 5:
In our images, as you point out, we do a 5-day averaging to remove the
noise, often errors due to ephemeral weather effects, from the
timeseries. This avoids the day-to-day ups and downs that can be
misleading and provides a more representative overall trajectory (though
we do get occasional wiggles from the preliminary data used in the 5-day
data that is later replaced).
For the timeseries plot, we also interpolate over missing data (such as
a missing swath) using data for that region from the day before and
(when it becomes available) the day after. However, there doesn’t appear
to be any missing swaths in our F17 data over the last several days.
Info on the sensor we use and the interpolation are explained on our
You’re welcome to print the above, though if you do, I would appreciate
if you would also add the following links, where we addressed the sensor
issue and made corrections to the near-real-time data.
And also here, where I discussed some the issues dealing with
near-real-time data from satellite sensors:
These may be useful for new readers or to refresh other readers’
memories, such as some of the readers who posted in the comments section.
Walt Meier, Research Scientist
National Snow and Ice Data Center
University of Colorado
UCB 449, Boulder, CO 80309