Tuesday 19 April 2016

Problems with DMSP-F17 SSMIS data

So far this month NSIDC have suspended all processing of the Sea Ice Index (G02135). This is explained on the NSIDC Sea Ice Index page as follows.
Notice (04/05/2016): Due to issues with the DMSP-F17 SSMIS data used to create the Sea Ice Index, we are suspending processing and investigating the problem with our data providers. Learn More. For questions, contact NSIDC User Services.


Note that this is from 5 April 2016, the date order is peculiar to the US.

The issue is as follows. The presence of ice is measured by microwave emissions from the ice, in this case the DMSP-F17 SSMIS satellite platform has what is essentially a sensitive radio receiver operating at 37GHz (37000MHz). To get a handle on that frequency it may help to consider that the medium wave band is around 1MHz, the broadcast FM band is around 100MHz, and Wifi and microwave ovens operate at 2.4GHz.

The satellite's 37GHz receiver measures the emissions of microwave energy from the ice, and the vertically polarised receiver has developed a fault. The effect of this fault is shown below in images from the NSIDC (source).

Working correctly.



Faulty.


My immediate feeling is that if NSIDC consider the matter serious enough to suspend processing I should suspend usage of sea ice data that might use the affected sensor. Wipneus's data uses NSIDC 0081 gridded concentration, which uses the instrument that has gone faulty, and that is a crucial data set in the regional based analyses I have done over the last few years.

This problem might also affect PIOMAS, which assimilates sea ice concentration data to 'trim' the model to actual ice conditions. NISE V4 is the source of the assimilation data, and that uses DSMP-F17 SSMIS, the system exhibiting faulty data at present. However Dr Zhang has stated in an email to me that he expects PIOMAS data will be released in May.

I think probably the best thing is to not post until I post on April PIOMAS in early May, and then decide what to do for the rest of the season. NSIDC state they are considering using a different satellite but that calibration between the new and old data may take some time. How long I do not know.

2 comments:

plg said...

To mirror the conspiracy-mindedness of Wats, L Smih, Inhofe et al, we should of course infer that somebody wrecked the sensor just when things were getting not only intensely interesting if not earth shattering... ;-)

Chris Reynolds said...

Hi Plg,

Ah yes, the orbiting Iluminati at it again, from their base in the orbiting tea-pot. :)