Page 1 of 1

cclean, pmosmem

PostPosted: Wed Oct 25, 2017 8:19 pm
by ncy
I am planning a polarization observation of an extended source. I prefer the Maximum Entropy Method (MEM) since it usually gives me better results for extended source. Is there anything similar to cclean, but with MEM?

As an alternative, can I use pmosmem on single pointing observation? Or I should do the observation with mosaic, even the source can be covered by a single pointing.

Re: cclean, pmosmem

PostPosted: Mon Oct 30, 2017 10:28 am
by Mark.Wieringa
Hi Stephen,

I think your suggestion of pmosmem is good - I just tried it on a single field and it works fine. No need to use multiple pointings if you don't need them to cover the object.

The other thing you can try is a new feature in cclean (borrowed from wsclean) to do a second clean to a deeper level, but only for pixels that have already been found in the first clean. Just specify a second cutoff level - you could use 5 times the noise for the first level and 1/2 to 1 times the noise for the second level.



Re: cclean, pmosmem

PostPosted: Mon Oct 30, 2017 6:12 pm
by ncy
Thanks Mark,

Good to know that, I will also try PMOSMEM + CCLEAN as suggested.