[daip] Regarding VLA " First sky survey" catalogue with 946432 sources.

[Eric Greisen]

On 05/21/2019 02:44 AM, Prabhakar wrote:
> Dear sir,
> 
> I am Msc-PhD student at IIT Kanpur(Indian Institute of Technology, Kanpur)
> (INDIA), in the Department of Physics. I am doing some project work using
> VLA, First sky survey" catalogue with 946432 sources. I have some queries
> regarding this catalogue.Please let me know the reason behind the quires
> that I am mentioning below:
> 
> 1. I saw around 170000 sources with both Bmaj and Bmin zero and around
> 3lakhs sources with Bmaj non zero but still Bmin zero. What it implies!
> 
> 2. Through my work I have noticed when I plotted a histogram of position
> angle(PA), it appeared most of the sources with PA 90 deg. and even more
> sources with PA 0 and 180 deg. What it implies!
> 
> 3.** As for extended sources ratio of peak_flux/flux_inte.  would be less
> than 1 or close to 0.9. But most of the sources with its ratio greater
> than 1. How is it possible? If yes, then What it implies!
> 
> 4** I want very random data, with PA  random distribute. Is there any bias
> or limitations on these data, let me know, please.
> I will be very grateful for your response to these issues.
> Thank you
> 
> Regards
> Prabhakar
> Msc-phd student
> IIT kanpur(INDIA)
> 
> _______________________________________________
> Daip mailing list
> Daip at listmgr.nrao.edu
> https://listmgr.nrao.edu/mailman/listinfo/daip
> 

My previous reply was based on the assumption that the authors of FIRST 
used AIPS for the fitting.  It is quite likely that they did not.
Nonetheless, the issue of deconvolution is always present.  If the fit 
values do not resemble the Clean beam due to noise etc then the 
deconvolution will produce improper solutions.  The authors may also 
have chosen to change insignificant values of Bmaj and Bmin to 0.

The observations in FIRST were short snapshots with the VLA.  As such, 
the center part of the dirty beam is always elliptical with the long 
axis NS for southerly sources.  This may well bias the fit PAs toward EW 
(i.e. 90).  AIPS does not give the answers you see in question 3 which 
also suggests that the authors used their own software for the fitting.

If you restrict your study to sources of higher S/N (i.e. >=10) you may 
find a more random distribution of things.

These questions should really be directed not to NRAO but to the primary 
author Bob Becker.  I believe he has retired but is still publishing.

Eric Greisen