I have written to you regarding few problems regarding "splot" but no answer. We have another problem regarding extracting spectra. We can take our spectra with the arc on the same image. In this case we can extract the object spectra and use the arc to calibrate it because both object and arc are in the same frame. But recently we took the object spectra and arc separately. Now our object and the arc are in two frames. But we can't take the object and arc in the same area of the frame. I mean the arc is always appeared beside the spectrum. See the attachment. If we extract the arc using "apall" by referring the object spectra is it correct?
I have attached
1. Object spectra
2. When Arc arc is added to object
3. Arc only
Post generated using Mail2Forum (http://www.mail2forum.com)
From: Frank Valdes <firstname.lastname@example.org>
Date: March 23, 2006 8:59:32 AM MST
To: Janaka Adassuriya <email@example.com>
Subject: Re: spectra
I'm sorry for not replying sooner on you first message.
> I reduced some emission line spectra to find out the FWHM. In =20
> "splot" task there are two ways.
> 1. using "h" key and type a/b/c
> 2. using a fit. i.e. type k and select the profile g-for gaussian
> both methods give FWHM but slightly different values. What is the =20
> different of these two and where we use "h" and "k". Please let me =20
> know it.
> Your help is highly appreciated
There is no real answer to this. It is a scientific decision based =20
on the data and the project. I would say that the two methods most =20
used would be profile fitting ('k') and numerical integration ('e').
> We have another problem regarding extracting spectra. We can take =20
> our spectra with the arc on the same image. In this case we can =20
> extract the object spectra and use the arc to calibrate it because =20
> both object and arc are in the same frame. But recently we took the =20=
> object spectra and arc separately. Now our object and the arc are =20
> in two frames. But we can't take the object and arc in the same =20
> area of the frame. I mean the arc is always appeared beside the =20
> spectrum. See the attachment. If we extract the arc using "apall" =20
> by referring the object spectra is it correct?
You would use the object aperture as the reference. But then you =20
enter the editing and shift it to where the
arc is located. You need to use the object because, normally, you =20
can't trace an arc since there is no signal between the emission lines.
> We have no idea what is happening in the task "apall". Specially =20
> what is happening in the task called "tracing".
> could you please give me an idea about this.
Tracing is the process of centroiding a profile repeatedly at a =20
number of places along the spectrum and then fitting a curve to those =20=
centroids. The number of places is set by a t_step parameter while =20
the number of rows or columns averaged before centroiding is given by =20=
t_nsum. For the starting point in the centroiding it used the =20
position found in the previous step. It is required that the next =20
center be within radius of the previous value to avoid jumping to a =20
neighboring profile or noise. If the centroid is not found in one =20
step it tries going on to the next step to see if it can "recover" =20
the profile. The number of times it does this before deciding it =20
cannot find the profile is set by t_nlost.
The curve fitting is used to minimize the noise of the individual =20
centroids and because profiles should generally be a smooth =20
continuous path in the image set by camera and detector distortions =20
and alignment. Once the curve is found the software can then =20
evaluate where the profile is at every point along the spectrum.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum