[alma-config] telecon this Wednesday

[Min Yun]

> > Morita-san,
> > 
> > Looking at your latest report, I am struck by the fact that the
> > sidelobe behavior of the spiral configuration is not as good as it 
> > can be (John must be working on optimizing this right now).  
> > If one of your conclusions is indeed a possible correlation 
> > between the sidelobe levels and imaging quality, then 
> > there are even more reason to include
> > the donut strawperson configurations to your tests.  
> 
> Remember that the donut arrays are the results of optimizing the
> sidelobes for a SNAPSHOT, and that John convincingly demonstrated
> that the nearin sidelobes don't average down with long integrations.
> The distinction between nearin and farout sidelobes is somewhat unclear,
> but image dynamic range (measured by looking far off the source) should
> be related more to far out sidelobes.  In Morita's test image, composed
> of small Gaussians (which does not require mosaicing), the on-source
> errors may be more dominated by the near in sidelobes.

Mark,  I am not disagreeing with the fact that earth rotation does little
to solve the near-in sidelobe problem.  It certainly suppresses the 
far-sidelobes.  Deep imaging studies with the VLA certainly suggests
far-sidelobes cause serious imaging problems (ask Frazer).  
My point is that the strawperson donut arrays have SMALLER near-in 
sidelobes than the spiral array Morita-san used for his study -- because
donut arrays also have tapered uv coverage and the particular sprial
array used was not optimized -- and it will make an interesting
comparison.  I would like to shatter this incorrect notion some people
have that the donut arrays have a much worse near-in sidelobe problem.
The near-in sidelobe problem is a general one and is not inherent to 
any particular style of array.

> Perhaps the direction we should be going in is: over what range of
> hour angles is it relevant to try to optimize the sidelobes over?
> The compact configuration, which will do many snapshot mosaics, should
> have the lowest sidelobe levels possible for snapshots.  The larger
> configurations might have hour angle ranges for sidelobe optimization
> which is given by some criteria similar to what I did in MMA Memo 201.

No one has yet to offer a concrete reason why Leonia's snapshot 
optimization is flawed.  While earth rotation clearly has an impact
on the imaging quality due to changes in the far-sidelobe levels,
I don't know of any telescope TAC that has yet to grant extra observing
time so the imaging quantity can be improved by a factor of 2.  
In other words, earth rotation synthesis is a relevant imaging issue
(and we should warn the users about this as much as other
critical issues such as the thermal and atmospheric noise), but it
is also a science issue that is outside our considerations.  We are 
not going to impose a 4 hr earth rotation synthesis on any project
that does not require it (from the raw sensitivity requirement)
only because the configuration designer
determined this mode produces the best imaging characteristics.
Personally I like what you have done in the project book, namely
offering the amount of HA coverage that is required to sample
some fraction of the uv cells.  No matter what the configuration scheme
we adopt in the end, we should provide this information (i.e.
what HA coverage is needed to sample 40 or 50% of uv cells for
"good" imaging), just the same way we we inform people of the 
sensitivity and resolution achievable.

> > As you stated
> > at the beginning, comparing the two extreme cases should give us
> > the ranges of outcomes we may expect from these imaging simulations,
> > but it is unavoidable that you should also compare the spiral and
> > donut arrays directly.  Even your possible conclusion that the
> > two produce nearly identical results should be a useful and tangible
> > work in making further decisions down the line.
> > 
> 
> True.
> I was somewhat disturbed though, that the simulation plans are ahead of
> discussion of the pointing error results in the agenda.  It is
> exactly backwards: the whole point of the pointing error simulations
> was to say that the original simulation plans were flawed.
> Would you consider reordering the agenda?

Since we apparently we both missed the telecon, you will have to make
your point at our next telecon.  What Morita-san has shown is that
point error can severely impact imaging.  The same is true for a number
of other things such as atmospheric phase noise and gain fluctuations.
Perhaps you or some other brilliant person will someday
produce a software solution so that pointing errors are properly
accounted for and thus does not limiting the imaging.




					-- Min