[alma-config] parameters for IMAGR, and integration times

[Bryan Butler]

> 
> >> Steven wrote:
> >> The A array simulations have been redone with an integration time of 60
> >> seconds rather than the 180 seconds used to date, and for the other
> arrays,
> >> as this was felt to be too long for the longer baselines. The redone
> results
> >> have been posted to the site, accessible from
> >> http://www.stevenheddle.co.uk/ALMA/ALMA_IND.HTM
> >>
> >> Cheers,
> >>     Steven
> >>
> >>
> >
> >Hi Steven,
> >
> >I do not believe this is correct. The Nyquist sampling for
> >a 10 km baseline is 8 seconds... It is 27 seconds for a 3 km baseline.
> >So your UV data are fairly heavily undersampled at their longest baselines.
> >This will add significant sidelobes, and perhaps explain why you get
> >large negative sidelobes than theory says.
> >
> >Stephane

it depends a bit on where the source is (how fast it moves through the
u-v cells), how far you are from the phase center, and how much error you
are willing to accept, at least as i understand it.  stephane has probably 
made some sensible assumptions here and just calculated a rule of thumb.

here is probably something similar to what he did...

take for illustration purposes an east-west baseline, and a source passing
overhead at zenith.  then the rate of change of u is:

  du    B
  -- ~ --- w
  dt    l

where B is the baseline length, l is the wavelength, and w is the earth
rotation rate.  now, convert the u-coordinate back to physical distance
(instead of wavelengths), and you have:

  du ~ B w dt

so, for a 10 km baseline, a 1 second integration time gives du ~ 0.7 m.
now, we'll want u-v cells of order an antenna size divided by 2 (at the
*very* least), or about 6 m, so the max integration time is of order 8 sec...


> 
> My problem is mainly the practical one of dealing with the UV data files
> this implies- at 60 seconds integration time the files are already 48MB, so
> at 8 seconds they will balloon to approximately 360 MB per file. As it
> stands (60 seconds) I reckoned I could shoehorn the A array simulations
> (including CLEANing)  into 6GB of AIPS disk, but am now faced with the
> prospect of requiring 45GB for only the A arrays! 

yes, this is a serious problem...  why are we concentrating on the A array
first, just out of my own curiosity?

> Also I was not aware that
> there was a problem with large negative sidleobes, but if there is (or
> indeed any other problems) please tell me as soon as possible.

well, if you don't sample your visibilities rapidly enough, then you can
certainly get artifacts.  it's not clear to me that larger sidelobes would
be a result, but i imagine it could be...

> Are these the integration times I should be using? Are there other issues
> such as noise which work against such short integration times? 

the noise of each visibility goes up, but the overall noise in the final
image should be the same (barring no systematics) because you're 
effectively vector adding the visibilities together to make the image.

> What
> integration times are commonly used on the VLA with its >20 km baselines?

at short wavelengths, 3.33 sec is what i would consider "standard".  but
this is as much to combat atmosphere as to combat time-average smearing.
even at the longer wavelengths (6 cm, e.g.), i don't think anybody would go to
integration times longer than 10 sec, in practice.


	-bryan