SDMEdit does not turn off data for NSST

[ADCollard]

Buoy 4602630 off of Baja Caifornia is reading anomalously low SST values. This is dragging the local SST in the analysis to very low values.

We tried to remove this observation with SDMEdit but to no avail. There are two issues:

1. The SDM edit file does not contain an entry for SST. So we cannot modify the QMST descriptor in the NSSTBUFR file (it is set to missing)
2. read_nsstbufr.f90 does not have logic to read in the QC Marks from the BUFR file.

At present no solution to turn off this buoy upstream of GSI exists. So we need to be ready to make an emergency code change.

It increases risk to try to change SDM edit without extensive testing, so we will attempt to key off the QMAT flag in read_nsstbufr.f90 as a temporary solution.

[ADCollard]

Development is in https://github.com/ADCollard/GSI/tree/release/gfsda.v16

[ADCollard]

The buoy appears to be at least temporarily recovered, giving us more time to find the best solution to this issue.

[RussTreadon-NOAA]

@ADCollard : should we keep this issue open or close it?

[RussTreadon-NOAA]

@ADCollard , what is the status of this issue? Do we need any changes for GFS v17?