Investigating the distribution of metals in star-forming galaxies 
: (Alternative Format Thesis)

  • Bethan Easeman

Student thesis: Doctoral ThesisPhD


During the life cycles of stars, metals are formed and subsequently dispersed into the surrounding interstellar medium through a number of processes. The metallicity of this interstellar medium therefore contains a wealth of information about the galaxy's formation and evolution, showing signatures of the star formation history, the impact of outflows from the galaxy, and events such as mergers. However, measuring this metallicity is no simple task - a number of methods have been developed, using the relative strengths of various emission lines, but large discrepancies exist in the results from different methods.

As well as known systematic offsets between the diagnostics, variations in the shapes of radial metallicity profiles in galaxies have also been observed when different methods are used, with suggestions of some diagnostics becoming unreliable under certain physical conditions. We have investigated these observed differences in two contexts. First, we explored the impact of using different diagnostics on the shapes of the measured radial profiles, and searched for any links to global properties of the galaxies which may shed light on the origin of various features. To do this, we used a sample of 758 galaxies from the MaNGA survey, with the resolved 2D maps allowing for variations in metallicity to be studied across the whole galaxy. We find the number of central dips observed to vary by almost a factor of three when different diagnostics are used, and that there is surprisingly little agreement on the shape classification of the profiles between the different diagnostics. When the O3N2 diagnostic is used, there appears to be a link between the prevalence of dips and certain global properties of the galaxy, such as the specific star formation rate, and the global stellar mass. Conversely, no clear link between the galaxies showing dips and these global properties were found when the N2S2Hα diagnostic was used. This implies that environmental conditions have a strong impact on the shapes of the profiles returned by certain diagnostics.

We then investigated the reliability of the strong line diagnostics which can be applied to MUSE data, using a smaller sample of 36 galaxies observed as part of the MUSE Atlas of Disks survey, at much higher spatial resolution than the MaNGA data. The improved resolution allowed for the dependence of various diagnostics on properties such as the ionisation parameter to be studied in greater detail, and the increased sensitivity meant that auroral lines could be detected in a greater number of sources. Therefore, to assess the reliability of the diagnostics, we compared results from the strong line diagnostics to results from a more reliable and accurate electron temperature (Te) diagnostic based on the faint auroral lines. For the identified HII regions, we found the S-calibration, O3N2, and N2 diagnostics to have significant metallicity-dependent offsets below the Te-based values. For the O3N2 and N2 diagnostics, these discrepancies additionally showed a strong dependence on the ionisation parameter, with increased discrepancies at high ionisation parameters. The N2S2Hα diagnostic, due to the line ratios used, is independent of log(U). When comparing to the Te-based results, the N2S2Hα results had a constant offset, with no clear dependence of the offset on either metallicity or ionisation parameter, suggesting this diagnostic to return more reliable results across a range of physical properties.

Utilising the high spatial resolution of MUSE, we then investigated the trends observed on sub-HII region scales. We found suggestions of the O3N2 and N2 diagnostics additionally being unreliable on sub-HII region scales, returning metallicity profiles with decreasing metallicity towards the centres of regions, which cannot be physically motivated. These diagnostics rely on an anti-correlation between metallicity and ionisation parameter, yet results when using the seemingly more reliable N2S2Hα diagnostic suggested instead a positive relationship, especially clearly at high metallicities.

This work has further highlighted the complexity of measuring metallicity, and the caution one must take when interpreting results. Further work is needed to conclusively define the conditions under which the strong line diagnostics can be reliably applied, especially with data from JWST allowing for metallicity measurements to be made out to higher redshifts, where the environmental conditions of galaxies are known to vary significantly from what is observed in the local universe.
Date of Award28 Jun 2023
Original languageEnglish
Awarding Institution
  • University of Bath
SupervisorPatricia Schady (Supervisor) & Stijn Wuyts (Supervisor)

Cite this