Setting the Flux Density Scale

       

When you set up your observation you should have used CACAL (or DELCOR in olden times) to make the initial on-line calibration. This generally involves setting the flux density scale via 1934-638 In the off-line calibration procedure, we essentially redo that step, although you could proceed with out it if you are confident you got it right on line.

We continue by establishing the flux density scale of the primary calibrator. Generally, this calibrator will be 1934-638, although possibly it might be one of the northern sources 3C286 and 3C48. You need to enter its flux density at all observed frequencies into the source (SU) table with SETJY (you can print the contents of the SU table with PRTAB). Later it will be used to determine the secondary calibrator flux densities, which will in turn be used to calibrate the program sources. There are very few sources with constant flux density and this is the reason for the leap-frogging technique.

You can either enter the correct flux density by hand, or allow SETJY\ to work it out for any of the above sources. First, I will discuss manual setting of the primary flux density.

1. • Make sure optype is blank.
   • Fill in the source name of the primary calibrator.
   • Assign the flux density of the primary to the first element of the zerosp adverb with something like zerosp=10.0,0 (this is also the correct thing to if you didn't convert to Stokes and have tricked AIPS into thinking you have circular polarizations). zerosp has room for only one source at a time, so don't give SETJY a list of sources. zerosp allows you to specify a model of the polarized flux density in a source as well, but we won't use this capability.
   • Here also is your chance to assign a calcode adverb for the primary calibrator source (see Appendix B). You can make this anything you like for easy selection of your calibrators. For example, you might use calcode='P' for primary calibrators, calcode='S' for secondary calibrators (although you will leave their flux density at zero for the moment), and calcode='BP' for bandpass calibrators (you can choose any character code you like, they don't have to be these ones). Later on, you can select all calibrator sources which have non-blank calcodes by putting calcode='*'. Don't set this for your program sources (i.e., any non-calibrator sources). Similarly, you can select all the calibrators that have a common calcode by setting, say, calcode='S' for the secondary calibrators. It is just a way to avoid typing in the a long list of source names over and over. See HELP calcode for more information. I recommend making use of this adverb in this way.
   • You will see that freqid is an input to SETJY. However, it does not behave as you might expect. freqid does not have any effect on source selection in SETJY, and you should ignore it for this particular application of SETJY (I will describe what it does do later). There is no space in the SU table for the freqid. Thus, when a source has the same name, but a different freqid, you would be unable to set the source flux differently for the two freqids. This is a design limitation.
   • To deal with the above problem, ATLOD uses the qual source selection adverb, and gives it the freqid as its value. As double insurance, ATLOD also appends the freqid to the source name if it detects any name/frequency ambiguity. The value of qual for different sources can be seen with the LISTR scan summary output discussed above, or by printing the SU table with PRTAB. If you only have one qualifier, or no source name ambiguity, set qual=-1, which means any qualifier. In addition, you may have appended strings such as `_A' and `_B' to the source names for different frequencies to deal with this ambiguity.
   • Somewhat different to freqid is the IF axis. If you have more than one simultaneous frequency (i.e., the IF axis is of length greater than 1), you should select the IF with bif and eif, as the flux density of the primary is likely to vary significantly between the frequencies designated by the IF axis. The combination of freqid (which specifies a group of simultaneous frequencies) and the location on the IF axis (which specifies which of the group you are interested in) fully specifies the frequency.

   sources='1934-638',' ' Select primary calibrator

   zerosp=6.33,0 Set flux density

   optype=' ' Use specified source parameters

   calcode='p' Make 'p' the primary calibrator code

   qual=-1 Means any qualifier or

   qual=2 do each qualifier separately

   bif=1 Select range of IFs to set eif=1

   freqid=0 Not relevant here

2. If you would like SETJY to enter the correct flux density into the SU table for you automatically, put optype='calc'. Fill in the source name as before (note that as long as 1934 is the first 4 characters of the name SETJY will recognize it as 1934-638). You must specify the freqid correctly for your source because this is the only way SETJY can know what frequency to use when calculating the flux density. As before, freqid does not interact with the source selection criteria.

   Recently, the spectrum of our primary calibrator has been more accurately determined (J. Reynolds, private communication). Previously, we relied on a very old Parkes spectrum augmented by a MOST point. The new spectrum differs from the old by some 5% at cm wavelengths. If you would like to use the most recent determination of 1934-638 then leave aparm(2)=0. If you are combining new data with older previously calibrated data (using a flux density scale set with the old spectrum), then set aparm(2)=1.

   sources='1934-638',' ' Select primary calibrator

   zerosp=0 Allow program to work

   optype='calc' out flux density calcode=' '

   qual=-1 Means any qualifier or

   qual=2 do each qualifier separately

   bif=1 Select range of IFs to set eif=1

   freqid=2 Must specify

   aparm(3)=0 New polynomial coefficients

3. To set the calcode (see above) you will need to make a separate run of SETJY with optype=' '. For secondary calibrators, you should not enter any flux densities, but you might like to give them calibrator codes such as calc='s' and so on.
4. SETJY can also reset information in the SU table. For example, if you put optype='rese' then the flux densities for the selected sources are zeroed. You might want to do this if the boot-strapping procedure that evaluates the secondary calibrator fluxes goes wrong. See the HELP file for more details.
5. After finishing with SETJY you may wish to re-summarize the observation (§ 5.2) to check the new source information. Alternatively, you could print the SU table directly with the task PRTAB.