VIPERS DR2 line fits readme
==================
Source
--------
DR2 data
http://vipers.inaf.it/rel-pdr2.html
DR2 paper
https://arxiv.org/pdf/1611.07048v1.pdf
In the catalogs, we have
----------------------------
selection All (W1 / W4 )
photometry 752,582 (475,726 / 276,856 )
spectroscopy 91,507 (60,528 / 30,979 )
Match (with column "num") in spectro and photo
90,994 (60,259 / 30,735 )
Only in spectro 513 (269 / 244 )
Line fitting procedure
------------------------
Outputs in : VIPERS_W14_summary_v2.1.linesFitted.fits
* compute the fraction observed in I and in R (foI, foR) by integrating the spectra in the r and i broad band filters from the CFHT and compare to the observed magnitude. Because VIPERS si flux calibrated, one should not use these quantities.
* fit Gaussian line profile models to the spectra at the positions where lines are expected. For VIPERS, we only consider four lines: O2_3728, O3_5007, H1_4862, H1_6564. These are the linenames reported in the header.
HEADER
======
The header reads {lineName + quantity}, with quantity taking the following values :
_a0 : position of the line in A ,
_flux : fitted total flux in erg/cm2/s,
_fluxErr : error on the fitted flux in erg/cm2/s,
_sigma : width of the Gaussian in A ,
_sigmaErr : error on the width of the Gaussian in A ,
_continu : continuum estimated with a median filter on the flux array around the line erg/cm2/s/A,
_continuErr : continuum error estimated with a median filter on the flus error array around the line erg/cm2/s/A,
_EW : equivalent width estimation = _flux / _continu,
_fd_a0_l : flux observed in the first pixel left of the line position _a0 in erg/cm2/s/A,
_fd_a0_r : flux observed in the first pixel right of the line position _a0 in erg/cm2/s/A,
_chi2 : chi square of the fit,
_ndof : number of degrees of freedom of the fit = number of data point used in the fit. You need to subtract the number of parameters to computed the reduced chi2.
Select and emission line sample
====================
First remove galaxies where the fit did not converge :
did_not_converge : O2_3728_flux_1 == 8.0E-17 && O2_3728_fluxErr_1 == Infinity
reliable_flux : O2_3728_flux_1 > 1e-19 && O2_3728_flux_1 < 1e-14 && O2_3728_fluxErr_1 > 1e-20 && O2_3728_fluxErr_1 < 1e-14 && O2_3728_flux_1 > O2_3728_fluxErr_1 * 5
ELG_SAMPLE : NOT(did_not_converge) && reliable_flux
One can use 7 instead of 5 to have a stricter sample.
Note: flux calibration errors could amount to an additional 20% uncertainty on
the line flux measurement.