Page 1 of 2

ISIS V5.5.3

Posted: Thu Sep 29, 2016 9:44 pm
by Christian Buil
ISIS V5.5.3 add the possibility to edit CUNIT2 and CTYPE2 keyword of spectra FITS file (optional, for example
for indicate absolute flux unit i.e. erg/cm^2/s/A):

http://www.astrosurf.com/buil/isis/isis_en.htm

Christian

Re: ISIS V5.5.3

Posted: Fri Sep 30, 2016 7:04 am
by Olivier GARDE
Christian,

Est ce que cette version tient compte du traitement des ordres d'un spectro eshel >59 ? (par rapport à la 5.5.2a)

Olivier

Re: ISIS V5.5.3

Posted: Fri Sep 30, 2016 9:01 am
by Christian Buil
Normalement oui Olivier, elle en est issue (j'ai oublier de préciser cette évolution dans les news).

J'ai changé le mode d'installation, qui est maintenant plus manuel, désolé.

Pour une installation pratique, fait toi un répertoire ISIS, ISIS2, ... par exemples sur ton disque,
décompresse le zip et clique sur l'icône ISIS (va falloir que tu re-remplisse les boites setup du eshel,
à faire une seule fois).

Christian

Re: ISIS V5.5.3

Posted: Fri Sep 30, 2016 4:18 pm
by Robin Leadbeater
Hello Christian,

Thank you for the update. It is good to have you back. I missed you and of course I have some questions :D

Is there any way to use spectra of stars for instrument response correction which are not in the ISIS database ?
http://www.spectro-aras.com/forum/viewt ... 1414#p6424

Also the ELODIE H alpha spectrum of Vega in the ISIS database does not look to have a good continuum calibration. The ones in the ELODIE 3.1 library are better.
http://www.spectro-aras.com/forum/viewt ... f=8&t=1369

Cheers
Robin

Re: ISIS V5.5.3

Posted: Sat Oct 01, 2016 9:13 am
by Christian Buil
I have a potentiel problem with Elodie 3.1: the flux calibration seem correct interstellar extinction.
If it is exact the base is not well appropriate for correct (on some stars) the instrumental spectral
response of your instruments because the Elodie 3.1 is not the apparent profile (out of terrestrial atmosphere). You confirm ? It is unclear.

Christian

Re: ISIS V5.5.3

Posted: Sat Oct 01, 2016 11:13 am
by Francois Teyssier
Thank you Christian

Next step : data about the reference star

Cheers,

François

Re: ISIS V5.5.3

Posted: Sat Oct 01, 2016 11:41 am
by Robin Leadbeater
Hello Christian,

I believe the LL data set downloadable here
http://www.obs.u-bordeaux1.fr/m2a/soubi ... nload.html
is calibrated in physical flux above atmosphere, uncorrected for galactic extinction but I will check with the authors to make sure.

It also states on that page :-
"It is also possible to retrieve individual spectra from HyperLeda - Spectrophotometric resources, where it is possible to choose the flux calibration (physical flux above atmosphere, continuum normalized, flux corrected for Galactic extinction)."
but I cannot get that HyperLeda to work.

I do not think though that this is the cause of the problem with the ELODIE Vega profile in ISIS. The agreement between ELODIE 3.1 and Calspec is very good.
(Before ISIS, when I used ELODIE spectra of Vega directly from the ELODIE archive, I noticed differences in continuum between the different spectra)

Wishing you good health
Robin

Re: ISIS V5.5.3

Posted: Mon Oct 03, 2016 12:54 pm
by Christian Buil
Hello,
(Robin, health is very much better! Les affaires reprennent :-))

An update of ISIS (V5.6) is just available. Integration of notion of personal spectral database (very rudimental). See explanation here which the example of ELODIE V3.1 integration:

http://www.astrosurf.com/buil/isis/new/release.html

Note also the adoption of BUNIT and BTYPE in place of CUNIT2 and CTYPE2 for define flux unit (more standard and more logical).

About ELODIE V3.1: you can use absolute flux version of course, but improper for instrumental response calculation (it is necessary to normalize the profile near unity before).
Also, the ELODIE 3.1 continuum shape for wavelength shorter then 4000 A is almost uncertain. To be careful.

François, the archive of calibration files info in the FITS header is not an easy task but an important subject (it is also a question philosophical question and modification of ISIS can be large - for the future...).

Christian

Re: ISIS V5.5.3

Posted: Mon Oct 03, 2016 4:02 pm
by Robin Leadbeater
Thanks Christian!

I agree about the poor flux calibration below 4000A in ELODIE 3.1. The original ELODIE 1.0 covered 4100 to 6800A and was normalised to 1 at 5550A.
arXiv:astro-ph/0101378v1 22 Jan 2001

Philippe Prugniel has confirmed that the LL spectra are uncorrected for galactic extinction.

He also has some warnings about using library spectra from any source. If we continue to use these spectra (eg MILES etc) as secondary relative flux standards, we need to take care that they are reliable. It is also important to carefully document the reference star and the source of the reference spectrum used for flux calibration so that if they are later found to be inaccurate we can at least correct the spectra. (This is similar to what BAA do in their photometry database (but AAVSO do not).

Here is the full reply from Philippe Prugniel:

The spectra having the keyword BTYPE=FLUX-PHY are reduced above the Earth atmosphere. They are not corrected for Galactic extinction (all the info needed to make this correction is in the header).

Note however that ELODIE, nor most of the libraries you may find, are not suited for the purpose of flux-calibrating spectra. Beside the issue of Galactic correction that you are pointing, their photometric precision is not large, and some observations may even happen to be quite poorly calibrated. The best solution is probably to use medium or high resolution observations of flux standards to study and derive the instrumental response, and separately correct the atmospheric response with the help of broad band photometry or low resolution (photoelectric) spectro-photometry.

Sincerely.
Philippe Prugniel.

Re: ISIS V5.5.3

Posted: Tue Oct 04, 2016 9:44 pm
by Paolo Berardi
Thank you very much super Christian!

About reference stars I just opened a thread with some questions:
http://www.spectro-aras.com/forum/viewt ... f=8&t=1582

All the best!
Paolo