[RADS] GFO flag bit 7
remko at altimetrics.com
Fri May 9 22:45:07 CEST 2008
Dear RADS users,
Josef Cherniowsky reported a problem that led to the identification of
a few problems with GFO flag bit number 7.
1) The bit was reportedly "not used" according to the getraw_g1.nml
file. That was incorrect. The bit was actually used and was supposed
to contain a flag identifying possible sea ice, similar to flag bit 7
in Envisat, for example.
The bit on the GFO data is actually related to a quality flag on VATT,
a voltage related to attitude. This voltage itself is averaged over a
significant length of time (1 minute) and the quality flag is raised
when there is too much variation in it. John Lillibridge once compared
this flag with the possible sea ice locations and found a high
correlation. Hence the dubbing of this quality flag as a sea ice
2) Because of the lengthy period of averaging, the quality flag is set
to bad (sea ice) for long stretches from coast lines, effectively
wiping away all coastal regions (for hundreds of kilometers) and even
entire land-locked seas like the Baltic. Hence, this flag removes too
much data and should not be used, unless you are not concerned about
coasts at all.
As a consequence, I've now removed the checking of bit 7 from the
defaults. If you want to check for it nonetheless, you should add it
in your local settings (by adding 2^7 = 128 to the editing limit).
John and I will be working on a more distinctive sea ice flag, based
on the radiometer and altimeter data, rather than VATT. When that is
done, flag bit 7 will probably return to the active state.
Note that this change requires only affects the file $RADSROOT/nml/
getraw_g1.nml. None of the data has changed.
More information about the RADS