Monday, August 11, 2008

Mosaicing: AIPS/Miriad/whatever works!

Is anyone familiar with mosacing techniques in AIPS, Miriad, or both? I have 21cm VLA spectral line data, taken mosaic-style, and am having a heck of a time trying to stitch the pointings together. Here's what I have done so far:

Reduced in the standard manner with AIPS up until the 'SPLIT' step
Wrote out each pointing with FITTP
Loaded into Miriad
In Miriad, 'invert', 'mossdi', and 'restor', mostly with the defaults for parameters

The best way I can describe the results is like a bad patchwork quilt...does anyone know of a good cookbook or guide for putting together pointings into a mosaic? I know AIPS has some tasks for this (VTESS/UTESS/LTESS) but I haven't had success with them. Thanks!

Katie

7 comments:

amanda said...

I think Miriad is supposed to have better mosaicing stuff than AIPS. I've always used Miriad when mosaicing. A couple of things to check: 1) make sure you've applied your calibration when splitting, 2) are you using option=mosaic in invert, and 3) are you boxing your source in mossdi?

If you can, it would be really helpful to post the inputs you're using and a picture of the output.

katie mae said...

Yes to 1) and 2), and for 3) I am looking for extended emission across the whole mosaic so I'd like to leave the region at the entire image. What sort of picture would be helpful? Here are my inputs (only the non-blank fields):

invert:
vis = *.uvlin
map = uvlin_map
beam = uvlin_beam
options = mosaic

mossdi:
map = uvlin_map
beam = uvlin_beam
out = uvlin_model
niters = 5000
options = positive

restor:
model = uvlin_model
beam = uvlin_beam
map = uvlin_map
out = mosaic_uvlin

emily said...

i have done this so many times in miriad with vla data ... so many years ago. :) (people say that AIPS can make nice mosaics these days but i haven't had a chance to try it yet.)
when you are running invert make sure you have options=mosaic,double .

amanda said...

There might be a limitation on cleaning the entire image if you don't use options=double as Emily suggests. I think otherwise it can only safely clean the inner quarter. Check the task documentation and see.

katie mae said...

ok, i am re-doing "invert" with option=mosaic, double...and it is taking forever, i'll let you know the result! thanks for your help!

Laura said...

I've always just used FLATN in AIPS, which I think just basically averages images together, but is reasonably intelligent in that it can do the right primary beam corrections (so no need to run PBCOR). However, you are not making use of all the information in the UV plane, so in that respect FLATN is less than ideal. But it does work, if you'd like to use it as a check. You just image each pointing individually, as deeply as you can (and self cal each pointing, if you like). The images of all the pointings should have the same names, but each pointing should have a different sequence number. Then set NMAPS in FLATN to the number of pointings you have. FLATN even works if each individual pointing has been imaged with multiple facets! In that case, set NFIELD to the number of facets each pointing has (or the max number, if the number of facets varies).

DonkeyPuncher said...
This comment has been removed by a blog administrator.