[alma-config] simulation

[Mark Holdaway]

> Dear Morita-san,
> 
> I have read your short memo about simulations including antenna pointing.
> I'd like to make two comments:
> 
> 1. I suppose that your simulation does not include mosaic. If it does then 
>    skip this comment.
>    You consider the case when the image(model) size is bigger than the primary 
>    beam. Of cource the big effect of the antenna pointing error can be expected 
>    at this case. 
>    On the other hand if the image(model) size is bigger than the primary 
>    beam than mosaic should be used. So this case is not practical for 
>    the single pointing observation.
> 

I believe the source structure was a field full of compact sources, much
like a survey.  In this case, you need mosaicing to get uniform
sensitivity over the field, but you do not need mosaicing to correctly
reconstruct each compact source, though the flux will be modulated by the
primary beam.

> 2. The synthesyzed beam given at figure 5 shows negative side lobes as big 
>    as ~10%. This looks strange for me. I shown in my paper that negative side 
>    lobes can not be more than 1/(N-1) for snapshot observation and for 
>    natural weighting, where N is number of antennas. I think they should be 
>    even less for non snapshot observation. 
>    May be you do not use natural weighting?
> 

Uniform weighting should give lower RMS sidelobe level; the theory is that
in the presence of pointing errors, image quality degrades proportional to
the RMS sidelobe level and proportional to the pointing error.
Interestingly enough, Morita-san's results show the image quality
degrading with the pointing error, but not with the rms sidelobe level.
IE, two configurations with very different sidelobe characteristics gave
similar results when the pointing errors began to dominate.

	-Mark