A Machine Learning Artificial Neural Network Calibration of the Strong-Line Oxygen Abundance

Abstract

The HII region oxygen abundance is a key observable for studying chemical properties of galaxies. Deriving oxygen abundances using optical spectra often relies on empirical strong-line calibrations calibrated to the direct method. Existing calibrations usually adopt linear or polynomial functions to describe the non-linear relationships between strong line ratios and Te oxygen abundances. Here, I explore the possibility of using an artificial neural network model to construct a non-linear strong-line calibration. Using about 950 literature HII region spectra with auroral line detections, I build multi-layer perceptron models under the machine learning framework of training and testing. I show that complex models, like the neural network, are preferred at the current sample size and can better predict oxygen abundance than simple linear models. I demonstrate that the new calibration can reproduce metallicity gradients in nearby galaxies and the mass-metallicity relationship. Finally, I discuss the prospects of developing new neural network calibrations using forthcoming large samples of HII region and also the challenges faced.

12 pages, 15 figures. Accepted to MNRAS

References in corpus (13)

• The Host Galaxies and Classification of Active Galactic Nuclei
• The SAMI Galaxy Survey: instrument specification and target selection
• New fully empirical calibrations of strong-line metallicity indicators in star forming galaxies
• Metallicity gradients in local field star-forming galaxies: Insights on inflows, outflows, and the coevolution of gas, stars and metals
• Keck HIRES Spectroscopy of Extragalactic HII Regions: C and O Abundances from Recombination Lines
• CHAOS I: Direct Chemical Abundances for HII Regions in NGC 628
• On the abundance discrepancy problem in HII regions
• Nebular spectroscopy: A guide on H II regions and planetary nebulae
• SDSS-V: Pioneering Panoptic Spectroscopy
• Young stars and ionized nebulae in M83: comparing chemical abundances at high metallicity
• Separate ways: The Mass-Metallicity Relation does not strongly correlate with Star Formation Rate in SDSS-IV MaNGA galaxies
• Red Supergiants as Cosmic Abundance Probes: massive star clusters in M83, and the mass-metallicity relation of nearby galaxies
• Galaxy emission line classification using 3D line ratio diagrams

Cited by in corpus (13)

• Mapping metallicity variations across nearby galaxy disks
• Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
• Density biases and temperature relations for DESIRED HII regions
• A re-assessment of strong line metallicity conversions in the machine learning era
• HII regions in CALIFA survey: II. The relation between their physical properties and galaxy evolution
• A physically motivated `charge-exchange method' for measuring electron temperatures within HII regions
• How do supernova remnants cool? -- I. Morphology, optical emission lines, and shocks
• An Empirical Calibration of the Helium Abundance in HII Regions based in Literature and CALIFA Survey data
• Temperature based radial metallicity gradients in nearby galaxies
• How do supernova remnants cool? II. Machine learning analysis of supernova remnant simulations
• The impact of bars on the properties of HII regions in the TIMER survey
• [$\alpha$/Fe] traced by H ii regions from the CALIFA survey: The connection between morphology and chemical abundance patterns
• Analysis of the Effect of Bars on Environmental Dependence of Disc Galaxies with MaNGA Survey Data