Spectral (line or continuum) data

     

 

If you decide to retain the spectral information, whether you are interested in continuum or spectral line data, then the editing procedure is more time consuming. This is obviously because of the additional dimension and no-one wants to fossick through every channel, one by one. There are two approaches to editing the spectral data base.

1. The first method is to form the channel averaged data base (channel 0, see § 6) anyway, and flag it with QUACK, IBLED, TVFLG, and UVFLG just as described in the previous section. You can copy the FG table back to the spectral data base, and apply the flags to all channels for the flagged time ranges and baselines. The rationale is that many problems occur at a certain time in all channels rather than in individual channels. This is probably a good way to proceed at 3 and 6 cm where channel-specific interference is rare (except for the single channel birdies at odd integral multiples of 128 MHz). At 13 and 20 cm I would advise against this approach.
2. Similar to TVFLG is the task SPFLG. This task does the same job as TVFLG except that the spectral dimension is displayed in the x-direction instead of the baseline. In this way, you can edit fairly fast, but you must display each baseline separately. It also removes the useful feature of flagging all baselines simultaneously at a common time stamp. IBLED suffers from the same defects. SPFLG is absolutely vital in most spectral line experiments as, up until the time of writing, interference has been a severe problem, especially in the 13 and 20 cm band. It is the only way to edit such severely corrupted data properly. If you just worked on a channel 0 data base, you would probably find that you flagged everything.

   See the discussion on TVFLG in § 7.1 for some help on how to use SPFLG.

   docat=1 Catalogue master work file

   in2seq=0 Create new master work file or

   in2seq=2 Select old master work file with sequence number 2

   sources=' ' Select all sources

   timerang=0 Select all times

   stokes='I' Display Stokes I or Stokes

   stokes='RR' RR or LL if polarization

   stokes='LL' trickery (really XX,YY) invoked

   freqid=1 Select desired frequency

   bchan=1 Select all channels echan=0

   antennas=0 Select all baselines for

   baseline=0 sequential display

   docalib=-1 No calibration

   flagver=0 Write to highest version of FG table already present or make a new one

   doband=-1 No bandpass to apply dparm=0

   dparm(6)=10 Set minimum integration time in seconds

3. Another useful way to look at the spectral data is with POSSM. This task can display spectra from selected baselines and also average them in time. If you average data in time, be sure to think carefully about whether you want amp-scalar or vector averaging of the visiblities; see the discussion in § 6.2 on POSSM.

Last update : 03/03/94

1