The SAMI Galaxy Survey: reconciling strong emission line metallicity diagnostics using metallicity gradients

• Published in: Monthly Notices of the Royal Astronomical Society
• Main Authors: Poetrodjojo, Henry, Groves, Brent, Kewley, Lisa J, Sweet, Sarah M, Sanchez, Sebastian F, Medling, Anne M, López-Sánchez, Ángel R, Brough, Sarah, Cortese, Luca, van de Sande, Jesse, Vaughan, Sam, Richards, Samuel N, Bryant, Julia J, Croom, Scott M, Bland-Hawthorn, Joss, Goodwin, Michael, Lawrence, Jon S, Owers, Matt S, Scott, Nicholas
• Other Authors: Australian Research Council
• Format: Article in Journal/Newspaper
• Language: English
• Published: Oxford University Press (OUP) 2021
• Subjects: sami
• Online Access: http://dx.doi.org/10.1093/mnras/stab205; http://academic.oup.com/mnras/advance-article-pdf/doi/10.1093/mnras/stab205/36150192/stab205.pdf; http://academic.oup.com/mnras/article-pdf/502/3/3357/36296989/stab205.pdf

ABSTRACT We measure the gas-phase metallicity gradients of 248 galaxies selected from Data Release 2 of the SAMI Galaxy Survey. We demonstrate that there are large systematic discrepancies between the metallicity gradients derived using common strong emission line metallicity diagnostics. We determine which pairs of diagnostics have Spearman’s rank coefficients greater than 0.6 and provide linear conversions to allow the accurate comparison of metallicity gradients derived using different strong emission line diagnostics. For galaxies within the mass range 8.5 < log (M/M⊙) < 11.0, we find discrepancies of up to 0.11 dex/Re between seven popular diagnostics in the metallicity gradient–mass relation. We find a suggestion of a break in the metallicity gradient–mass relation, where the slope shifts from negative to positive, occurs between 9.5 < log (M/M⊙) < 10.5 for the seven chosen diagnostics. Applying our conversions to the metallicity gradient–mass relation, we reduce the maximum dispersion from 0.11 dex/Re to 0.02 dex/Re. These conversions provide the most accurate method of converting metallicity gradients when key emission lines are unavailable. We find that diagnostics that share common sets of emission line ratios agree best, and that diagnostics calibrated through the electron temperature provide more consistent results compared to those calibrated through photoionization models.