Talk:Atmospheric parameters of VSOP F-G-K stars

Overview
Team: Ivo, Thomas (lead)

Goal: obtain Teff, logg, [Fe/H], plus radial velocity and vsini for all F-K stars to be treated in VSOP-1. Obtain abundance estimates of other elements in case of non-solar metallicity.

Methods:
 * For low vsini stars, use DAOSPEC (e.g. Dall et al. 2005, A&A 444, 573) with atmospheric models (ATLAS9) to find Teff, log g and abundances. In case of high vsini, use MOOG to calculate synthetic spectra and do a chi^2 minimization to fit for Teff, logg and [Fe/H].
 * For binaries, the spectral subtraction technique of Montes at al. (2000; A&AS 146, 103 + 1995; A&A 294, 165) may be tried in order to simultaneously fit and separate the spectra of the two components. Alternatively, smaller regions may be fit in order to estimate Teff, logg of the components.

Deadlines:


 * 1) Procedures for low-vsini analysis defined and models ready
 * 2) Low-vsini analysis done
 * 3) Define regions to fit for high-vsini stars
 * 4) Procedures for high-vsini iterative analysis defined
 * 5) High-vsini analysis done
 * 6) Binaries procedures defined...

Remember that the team leader should organize the work, and make sure dates are assigned to the deadlines. These should be published on this Team-page before May 24 (for Paper1)

Categories:
 * F stars
 * G stars
 * K stars

Discussion of procedures
To start the definition of our procedures: I've setup the full analysis 'pipeline' on my laptop, ready to run. For the low vsini stars, the process is mas o menos like this: For the high vsini stars, we need to rely on matching a few prominent features with synthetic spectra calculated using MOOG and atmospheric models. We need to iterate/think a bit about this...
 * 1) create directory HDxxxxxx with the spectrum (HDxxxxxx.fits). Example: /home/tdall/VSOP/HD003456/HD003456.fits
 * 2) * The 'HD' prefix is necessary for abund_gui to recognize star-directories.
 * 3) * The name must contain exactly 6 digits after the 'HD'!!!
 * 4) start abund_gui from one directory higher! Example: in /home/tdall/VSOP issue the command abund_gui.pl
 * 5) choose the star in the 'Calculate' pull-down menu. Example: choose HD003456 from the menu.
 * 6) in the main window of abund_gui, click the 'Get EWs' button, and fill in the fields.
 * 7) * with IRAF/MIDAS measure an approximate FWHM in pixels of an isolated spectral line.
 * 8) * estimate the RV, e.g. from the prominent Ca I line at 6162.17 AA.
 * 9) * set 'Scale FWHM...' to 1, and 'Fix FWHM' to 0. Remaining parameters are good defaults for HARPS spectra.
 * 10) click 'Run DAOSPEC' and wait for it to finish. Check you get some reasonable number for RV, within the limits you specified.
 * 11) * Note that this can take a loooong time when the linelist is long, which is the default...
 * 12) in the main window, click the 'Linelist' button and select the lines. This is a long and tedious process, but need only be done once for any spectral type. More info to come! See perldoc abund_gui.pl for details (when the new update is available). (Trick: copy the file vald.kanonlist from somewhere else and be sure to call it something that ends in .lst before you begin)
 * 13) Select a model from the pull-down menu (or make a new one), make sure reference is set to HD000000 (the Sun) and click 'Get new abund.' Based on the plots, adjust the microturbulence and/or make new models until there are no slopes in the plots and FeI/FeII abundances agree.

As for the regions to fit, one way is to look at the solar spectrum, finding regions that contains some reasonably isolated and not too strong FeI and FeII lines. I am trying to compile a preliminary list of lines we should check, using the solar spectrum, and checking what other people have done. I will post it here later.