Important notice to Argo users (pressure offset errors, 2nd notice):
This is to provide an update to the earlier notice (see http://www-argo.ucsd.edu/Acpres_offset.html),
with regard to progress in correcting pressure offset errors.
WMO ID numbers for the affected instruments in each of the groups discussed above are posted below.
- Pressure offset errors in WHOI/FSI Argo float profiles include a subset (a) that can be corrected exactly
using automated procedures and a subset (b) that requires expert examination to produce an approximate correction.
- Profiles in subset (a) have now been corrected, and replacement of GDAC files for these profiles will be
completed by 20 October. Procedures for real-time correction of these profiles are in place as of 10
October. These floats have been removed from the greylist as of 10 October, meaning the profiles will be
available on the GTS.
- Profiles in subset (b) are undergoing expert examination and the files are being replaced on the GDAC as
they are completed. Errors are noted in the files corresponding to the uncertainty in the pressure
correction. Since these profiles cannot be corrected automatically, the corresponding instruments
continue to be greylisted (i.e. profiles are not on the GTS).
- While studying the pressure offset errors, a related problem was discovered in a group of WHOI/SBE
profiles. Reported pressures from these instruments corresponded to the bottom pressure of bins rather
than to the mid-bin pressure. This ½ bin pressure offset error is generally less than for the profiles
noted in (1) above. For the affected WHOI/SBE instruments, all profiles have now been corrected and are
available on the GDACS. The real-time data stream for these instruments has been
corrected (as of 14 September). These instruments are not greylisted.
The Argo project will continue to improve procedures for detection and correction of data quality problems, and
Argo endeavors to provide data of highest possible quality. Users should note that near real-time data are subject
only to automated quality checking, and the best quality data for climate research applications are available
only in delayed-mode (“D” files). Users can assist by reporting any data quality problems to
to the AST co-chairs if the problems persist.
Dean Roemmich and Howard Freeland, on behalf of the international Argo Steering Team