Parameter Recovery Study on IZI—a Bayesian Analysis Tool for Emission Lines from H ii Regions and Star-forming Galaxies 2026 Affiliation: Korea Astronomy and Space Science Institute, 776 Daeduk-daero, Yuseong-gu, Daejeon, 34055, the Republic of Korea Parameter Recovery Study on IZI—a Bayesian Analys
Jong-Ho Shinn, Rory Smith, Kyuseok Oh · 2025
Read Parameter Recovery Study on IZI—a Bayesian Analysis Tool for Emission Lines from H ii Regions and Star-forming Galaxies 2026 Affiliation: Korea Astronomy and Space Science Institute, 776 Daeduk-daero, Yuseong-gu, Daejeon, 34055, the Republic of Korea Parameter Recovery Study on IZI—a Bayesian Analys by Jong-Ho Shinn, Rory Smith, Kyuseok Oh (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-13124.txt.
Introduction
Jong-Ho Shinn, Rory Smith, and Kyuseok Oh stress-test an astronomy analysis tool in this two thousand twenty-five paper. IZI is a Bayesian program that reads emission lines from ionized gas regions and star-forming galaxies and estimates metallicity, ionization parameter, and dust reddening. The authors created mock datasets with IZI itself, in idealized and realistic variants, then ran the estimation back on the mocks to see whether the true parameters were recovered.
Recovery was biased: estimates missed the truth by about one standard deviation or more once model error was included, even using every available emission line. The cause is substantial model error inside the likelihood that varies with the parameters. The authors strongly recommend users run their own recovery tests before trusting IZI output, and they share a script for doing so. The warning extends to similar tools such as NebulaBayes.
Catalog record
| CREATOR | Jong-Ho Shinn, Rory Smith, Kyuseok Oh |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-13124
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Parameter Recovery Study on IZI—a Bayesian Analysis Tool for Emission Lines from H ii Regions and Star-forming Galaxies 2026 Affiliation: Korea Astronomy and Space Science Institute, 776 Daeduk-daero, Yuseong-gu, Daejeon, 34055, the Republic of Korea Parameter Recovery Study on IZI—a Bayesian Analysis Tool for Emission Lines from H ii Regions and Star-forming Galaxies – References Affiliation: Korea Astronomy and Space Science Institute, 776 Daeduk-daero, Yuseong-gu, Daejeon, 34055, the Republic of Korea Jong-Ho Shinn, Rory Smith, Kyuseok Oh Abstract We present a series of parameter recovery test results of the Bayesian analysis tool IZI, which analyses emission lines from H ii regions and star-forming galaxies and returns the estimates of the gas metallicity 12 + log (O/H) 12+\text{log\,(O/H)} , ionisation parameter log q \text{log}\,q , and nebular emission-line colour excess E ( B − V ) E(B-V) . We created several mock datasets using IZI to represent a few different ideal or realistic datasets and performed parameter estimation on the mock data with IZI. We found that IZI underestimated or overestimated the parameters by approximately 1- σ \sigma or greater when the model error was included, even when using all emission lines available in the model grids. We strongly recommend that IZI users run parameter recovery tests adjusted for their data before interpreting the IZI estimates. To encourage the appropriate use of IZI, we also share a script for parameter recovery tests. The cause of IZI’s biased estimation is the substantial model error in the likelihood term, which varies with the model parameters. We thus note that any parameter estimation with a substantial, varying model error in the likelihood term could return biased estimates for the model parameters, such as in the case of NebulaBayes, another Bayesian analysis tool for photoionisation emission lines. We also note two issues relevant to setting the log q \text{log}\,q prior using the observed line ratio [S III] λ λ 9068 , 9532 \lambda\lambda 9068,9532 / [S II] λ 6717 , 6731 \lambda 6717,6731 (the mismatch of line flux terms and violation of Bayes’ theorem) and propose a way to avoid the issues. 1 Introduction Galaxies evolve in diverse manners since their formation over the cosmic time. They merge ( Conselice et al. 2003 ; Tacconi et al. 2008 ; Lotz et al. 2011 ) , change their morphologies ( Brinchmann et al. 1998 ; Lotz et al. 2008 ; Conselice 2014 ) and colours ( Bell et al. 2004 ; Whitaker et al. 2011 ) , accrete nearby gas ( Block et al. 2002 ; Sancisi et al. 2008 ; Papovich et al. 2011 ) , form new stars of diverse masses ( Kennicutt 1998 ; Madau & Dickinson 2014 ) , and enrich themselves with newly synthesised elements from supernovae ( Nomoto et al. 2013 ) . Among these processes, star formation is of special interest, because it changes the properties of galaxies in various ways. Star formation consumes gas and dust, which are the raw materials for new stars ( McKee & Ostriker 2007 ; Zinnecker & Yorke 2007 ; Klessen & Glover 2023 ) ; adds new stars, which changes the galaxy’s spectrum ( Conroy 2013 ) ; and generates ultraviolet radiation and stellar winds, which ionise and transform surrounding environments ( Yorke 1986 ; Shields 1990 ; Puls et al. 2008 ; Eldridge & Stanway 2022 ) . Recently formed massive stars ( \gtrsim 8\,M_{\odot} , Zinnecker & Yorke 2007 ) create ionised regions around themselves— H ii Regions—where diverse emission lines emanate ( Osterbrock & Ferland 2005 ) . These emission lines enable astronomers to extract the host galaxy’s properties, such as the gas metallicity, and multiple methods have been developed for gas metallicity estimation. These methods use the electron temperature, metal recombination lines, photoionisation model, and strong emission lines ( Peimbert et al. 2017 ; Maiolino & Mannucci 2019 ; Aller 1954 ; Peimbert 1967 ; Kewley & Dopita 2002 ) . Among such methods, photoionisation models have several advantages in gas metallicity estimations ( Maiolino & Mannucci 2019 ) . It has no limits on the parameter range that can be explored in principle and provides information about the ionisation property in addition to the metallicity. It enables calibrating strong emission lines, where the direct electron-temperature method is unfavourable. IZI, coined from ‘I’nferring the gas metallicity (‘Z’) and ‘I’onization parameter, is an analysis tool that uses photoionisation model grids ( Blanc et al. 2015 ; Mingozzi et al. 2020 ) . IZI analyses the emission line ratios and returns Bayesian estimates of the gas metallicity, ionisation parameter, and nebular emission-line colour excess. Numerous studies have used IZI for their analyses, and their topics are diverse—for example, the impact of diffuse ionised gas on the emission-line ratios from star-forming galaxies ( Zhang et al. 2017 ) , evolution of the dust-to-metal ratios in galaxies ( De Vis et al. 2019 ) , metallicity variations across galactic disks ( Kreckel et al. 2019 ) , host galaxy of a fast radio burst source ( Xu et al. 2022 ) , galactic azimuthal variations of oxygen abundance ( Ho et al. 2017 ) , mass-metallicity relation in galaxies ( Barrera-Ballesteros et al. 2017 ) , value-added catalogue of galactic spectral energy distributions ( Chilingarian et al. 2017 ) , and Ly \alpha emitter at the epoch of reionisation ( Jung et al. 2024 ) . Meanwhile, Shinn 2020 reported a possible overestimation of model parameters in studying the gas metallicity estimation of star-forming galaxies. Shinn 2020 modelled emission line fluxes based on the relationships between the metallicity and line-flux ratios, which is used in the strong line method described above ( Peimbert et al. 2017 ; Maiolino & Mannucci 2019 ) . Shinn 2020 found that the nebula emission-line colour excess and intrinsic emission-line fluxes can be overestimated by up to a factor of five and showed that the overestimation stems from the model error in the likelihood term, which varies with model parameter. Considering this, we noticed that IZI includes the same type of model error in its likelihood term. This naturally led us to investigate IZI further, since the model error might cause an underestimation or overestimation of the model parameters, in which case we aim to guide the community on the proper use of IZI. Here, we present a series of results on a parameter recovery test of IZI. We created several mock datasets using IZI that represent different ideal or realistic cases and performed parameter estimation on the mock data using IZI. IZI should recover the values of the input model parameter used to create the mock data if it does not suffer from any estimation bias; however, we found that one parameter was underestimated or overestimated by more than 1- \sigma for the mock datasets mimicking realistic observations when the model error was included in the likelihood term. 2 IZI: Bayesian Analysis Tool for Emission Lines from H II Regions and Star-forming Galaxies IZI is a Bayesian analysis tool for emission lines from H ii regions and star-forming galaxies. It was originally written in IDL by Blanc et al. 2015 and designed to return the estimates of gas metallicity and