EW Computation (further discussion opened by Marco)

[Francois Teyssier]

Further discussion opened by Marco on spectro-l list.

I used the spectrum sent by Marco in my excel spreadsheet.

The results are :

Direct computation ( simple integration) EW = 16.03



Gaussian fit : EW = 15.93 and center = 6562.81 (excellent wavelenght calibration, Marco !)
(The Gaussian fit is a simple gaussian curve, with one compound).


Less than 1% diffrence between the two measurements.

Determining the edges of the line does not pose problems
Here's the integration range is [6553,6573]
If I adopt the range used by Marco [6548,6578], the EW becomes 15.96 : only 0.7% difference.


The real difficulty is : where is the continuum ?

François Teyssier

[Marco Leonardi]

François,

I agree that this forum should be the first point of reference for discussion like this. Thanks to open this topic in ARAS and to spend some of your time on my kap Dra spectra.

I take the opportunity to correct the software table comparison I posted yesterday, and after Benjamin MAUCLAIRE today post in spectro-l the new table is the following:

1. SpcAud’ACE (V2.10) command spc_ew kapdra 6548 6578 6 n 0 = -16.28
2. VSpec (V4.0.0) console command: Comp l 6548 6578 gives EW = -15.9
3. VSpec (V4.0.0) option EW lin fit = -15.905
4. VSpec (V4.0.0) option EW cst fit = -15.908
5. VSpec (V4.0.0) option EW cst 1 = -16.285
6. ISIS (V3.3.1) EW = -16.28
7. SPLAT (V3.9-6) Quick fit EW = -16.3

So all softwares seems to be in-line and your further confirmation is another important cross-check.

Another point is a link just posted in spectro-l by Stan Gorodensky related with EW computation. I did not read the pdf but I think it should be very interesting and probably can help to understand more about the question: where is the continuum ?

http://www.socastrosci.org/images/SAS_Vol_9_2.pdf

Regards,

Marco Leonardi

[Benjamin Mauclaire]

Hi,

I've made a tutorial about how to compute EW with SpcAudace in few different situation depending on the continuum:

http://bmauclaire.free.fr/spcaudace/docs/ew

Assumed that main factor that impact on EW has been set to continuum moldelisation as describe in OHP2010 workshop:

http://bmauclaire.free.fr/astronomie/sp ... iences/ew/

Benji