[alma-config] CLEAN and difference results for B arrays

[Bryan Butler]

mark,

yes, the artifacts in the mars images are exactly what you
expect if you start out with no initial model.

i would suggest trying it with an initial model.

in fact, i've been thinking that there is no reason to not
have an initial model for *all* of the images.  just do a
fit to some small (10ish?) number of gaussians (OMFIT in
AIPS is good for this, maybe there is a similar thing in
AIPS++?) and start with that model.

i have developed over the years most of the tools needed
for this for the case of the planets.  OMFIT does the fit
for the total flux density and limb-darkening parameter
from the visibilities.  then i have a private version of
CCMOD that calculates the clean components that make up the
initial model (standard AIPS CCMOD only does uniform disks,
not limb-darkened ones).  i didn't do much trickery on the
edge pixels, but the presumption is that CLEAN fixes that
up (though now that you bring it up, i'm thinking that
maybe i should fix that part of my version of CCMOD)...

as far as masks for the other images, doesn't that make some
assumption about you knowing where the flux density is in
the image (which we don't necessarily)?

steven, are you BOXing the CLEANs at all?

	-bryan