Documentation

Changes between DR2 and DR3

The stellar population measurements are unchanged from SAMI DR2 to DR3, other than the addition of extra galaxies and changes to the spectra themselves, based on updated data reduction.

Stellar population measurements

Stellar population parameters are measured from the aperture spectra only, using the method described in Scott et al. 2017, which we briefly summarise here. We note that the measurements presented as part of this release are not identical to those used in Scott et al. 2017 as i) the data version has changed and ii) the aperture definitions are somewhat different, though the agreement between the two sets of measurements is in general excellent.

We begin by measuring Lick absorption line strength indices for all spectra (Faber et al.1973, Worthey et al. 1994). For each spectrum, we correct for gas emission using an iterative template cleaning approach. We then broaden each spectrum to the wavelength-dependent Lick spectral resolution. We measure absorption line strengths for a set of 20 Lick indices on the emission-corrected, broadened spectra following the index definitions of Trager et al. 1998. For spectra where the effective resolution is already broader than the Lick resolution we correct for the effect of intrinsic broadening following Schiavon 2007. Uncertainties on all indices are determined by a Monte Carlo reallocation of the residuals and repeating the measurements on 100 realisations of the spectra.

The observed Lick index measurements are converted to Single Stellar Population (SSP) equivalent age, metallicity, [Z/H], and alpha-abundance, [$\alpha$/Fe]. We do this by comparing the observed absorption line strengths to the predictions of the stellar population synthesis models of Schiavon 2007 and Thomas et al. 2010. We use a $\chi^2$ minimisation approach with an iterative rejection of discrepant indices to determine the best-matching SSP parameters. For the Thomas et al. 2010 models we use all 20 measured indices. For the Schiavon 2007 models only 16 of the measured indices are predicted by the stellar population synthesis models. For the reasons outlined in Scott et al. 2017 we use SSP equivalent ages from Schiavon 2007 models and SSP equivalent [Z/H] and [$\alpha$/Fe] from Thomas et al. 2010 models.

In this release we make available all 20 Lick index measurements and the three SSP equivalent stellar population measurements for all 6 apertures. For the Lick index measurements we provide a flag for each aperture to indicate where the measurements may be unreliable due to unresolved issues in the measurement process. For the SSP equivalent measurements, we provide age and [Z/H] only where the S/N of the given aperture spectrum is greater than 10. We provide [$\alpha$/Fe] only where the S/N of the aperture spectrum is greater than 20. Missing SSP measurements are indicated by the value $-99$.

The data are provided in two tables, SSPApertures and IndexApertures. For each galaxy, each table contains the SAMI CATID, CUBEID and CUBEIDPUB, then six sets of columns containing the SSP or Lick index measurements and their uncertainties for each of the six apertures. Column names have the general form {Index or SSP}_{Aperture}{_err}, for example Age_RE or Mgb_4_ARCSECOND_err. The IndexApertures table contains six additional columns, FLAG_{Aperture}, where a '0' indicates measurements for that aperture are definitely reliable and a '1' indicates that care should be taken with using measurements for that aperture. The syntax for the six different apertures are: '3KPC_ROUND', 'RE', '1.4_ARCSECOND', '2_ARCSECOND', '3_ARCSECOND' and '4_ARCSECOND'.