Date: Thu, 6 Sep 2012 12:12:14 -0400 (EDT)
From: Jun-Hui Zhao 
To: Christopher Faesi 
Subject: Re: Question about continuum subtraction



On Wed, 5 Sep 2012, Christopher Faesi wrote:

> Hi Jun-Hui,
>
> Thanks for the quick response. In addition to the syntax I showed below, I
> also tried using channels from neighboring spectral chunks to fit the
> continuum, but then noticed that it fits each chunk separately, so it
> wasn't contributing to the part of the spectrum having the line. The
> strange thing is that the uv spectrum I'm displaying using smauvspec seems
> to have a different continuum amplitude than the images I'm making, which
> appear to be pretty close to 0 on average. Does this have something to do
> with the fact that the uv data are complex (and have a real and im part),
> while the imaged data are by definition just real? Is there some conversion
> that takes place behind the scenes?

Hi Chris,
    For a point source at the phase center, the amplitude of uv spectrum
would be the same as that from a map. You appear to have an extended weak
source with a peak emission that is off from the phase center. The
line emission is partly cancelled in the vector average of the uv data.
The amplitude of your uv spectrum reflects pretty much the noise level.
Your channel image looks good indicating that you have significantly
dectected a narrow line. Also, by eyeballs, the noise level from the
channel map is consistent with that from the uv spectrum.
    BTW, since the lines are week, you can use all the channels in
the continuum fitting with UVLIN to make a continuum-free uv data.
      Jun-Hui



>
> I'm working with my data in my group's area on the CF machines, which I
> don't think I have permission to make accessible; however, I've moved a few
> plots over to /home/cfaesi/smashare/, including the UV spec covering
> several chunks around the CO 2-1 line (in chunk 18), the "final" image,
> averaging for ~8 km/s around the line center, and an array of channel maps.
>
> I'd also be grateful to speak in person if that would be easier.
>
> Thanks again,
> Chris
>
> On Wed, Sep 5, 2012 at 5:22 PM, Jun-Hui Zhao  wrote:
>
> >
> >
> > On Wed, 5 Sep 2012, Christopher Faesi wrote:
> >
> > > Hello,
> > >
> > > I'm a Harvard graduate student and a fairly new MIRIAD user, and I'm
> > trying
....