[daip] making models

[Eric Greisen]

Ian McHardy wrote:
> Hello,
> 
> I want to do a few simulations to find out what sort of a map I would have
> made, with a variety of different VLA observational set-ups (eg taken from
> given real observations) of some model radio source.
> 
> So my first task, which I'd like to ask about here, is to make the model
> radio source in such a form that it can be easily inserted into a variety
> of datasets. Here I might be thinking of a few small scale components near
> a nucleus and some larger components further out.
> 
> I think I know how to insert clean components from a previous real
> observation into a uv dataset with uvsub. However if I want to put in a
> second set of ccpts, I have to run uvsub again, and if I want to put in
> some model that I have made up - which I do - then I can only see a way of
> including one model component at a time with smodel. Clearly this process
> is not feasible if I want to repeat it many times.
> 
> Maybe I can write a complex run-file which will run uvsub many times but
> is there is a simpler way of including not just one, but a whole load of
> man-made model components at one time?
> 
> many thanks,
> Ian McHardy

I would think that a CC file containing all of the various Gaussian 
components would allow a DFT (not gridded unless all are same Gaussian 
shape) UVSUB or whatever.  There are verbs to put values in tables so 
use any old CC file that has all the columns (IMFIT will make one or 
IMAGR multi-scale) and fill it as needed and TACOP it where needed.
Note that attaching a CC file to the UV data set will now work and 
always forces DFT.

Eric Greisen