Data missing two 128-MHz-wide spectral windows at the high frequency end

[xinglunju]

Several observations do not have the two 128-MHz-wide spectral windows at the high frequency end, inside the red circle in the snapshot below:

For comparision, this is what the SPWs should have been:

List of affected SBs

• https://github.com/ACES-CMZ/reduction_JACKS/issues/3
• https://github.com/ACES-CMZ/reduction_JACKS/issues/5
• more cases? please report here

Answer from NRAO staff

After some deep investigation we came to the conclusion that during the time of the observation some of the Correlator/Baseline boards were missing (Hardware issue). Therefore the data is unrecoverable.

[xinglunju]

This could worsen the continuum sensitivity (sqrt(14/16), i.e., ~7% increase in rms), and also affect the estimate the in-band spectral index (although in general in-band spectral indices have large uncertainties).