[daip] AIPS task UVMOD

[Eric Greisen]

On 04/14/2016 05:32 AM, lisakov wrote:
> Dear Eric,
>
> I've faced an insufficient(for me) explain section of the AIPS task
> UVMOD. Namely a FLUX adverb.
>
> Help says it is:
> FLUX.......Noise level to be added (in Jy. per Weight).
>
> I've tried to use a UV-plane RMS of "data minus model", which, I
> believe, describes the noise level. But it results in noise level of
> simulated data 5-10 times less than in original ones.
>
> So I would ask your assistance in understanding what does FLUX really
> mean in UVMOD task.
>
>
> Thank you in advance.
>
> Regards,
> Mikhail Lisakov.
>


I agree that it is confusing.  The assumption is that the weights are 
properly calibrated in 1/Jy^2.  The noise that is then added is

      Flux / sqrt (weight)

time a random number from -1 to 1.

Eric Greisen