Search

In my previous reanalysis of the local star-forming galaxies observed in the Hobby-Eberly Telescope Dark Energy Experiment (HETDEX) pilot survey, I reported that the overestimation of E (B − V), hence the star formation rate (SFR), undermined the claim of new galaxy population discovery in the original study. Herein, I re-examine whether the E (B − V) overestimation problem can be alleviated in the Bayesian parameter estimation framework by adopting scientifically motivated new priors. I modelled the emission-line fluxes of galaxies using the strong-line method and four model parameters – metallicity |$12 + \text{log {}(O/H)}$|⁠ , nebula emission-line colour excess E (B − V), intrinsic [O iii ] λ5007 line flux, and intrinsic [N ii ] λ6584 line flux. Based on mock data tests, I found that all input values can be recovered within and around the 1σ credible interval by adopting suitable priors for the intrinsic [O iii ] λ5007 and [N ii ] λ6584 line fluxes: the inverse gamma distribution reflecting the logical constraint that an intrinsic emission-line flux must exceed the observed (reddened) emission-line flux. The mock data tests were performed for two metallicity calibrations, three colour excess input values [ E (B − V) = 0.1, 0.3, and 0.5] and two metallicity input values [ |$12 + \text{log {}(O/H)}$| = 8.0 and 8.5]. I also found that the new prior can diminish the SFR overestimation eightfold. This study demonstrates how the Bayesian parameter estimation can achieve more accurate estimates with no further observations when the likelihood does not constrain the model parameters correctly. [ABSTRACT FROM AUTHOR]

I have reanalysed the data obtained for local (z < 0.15) star-forming galaxies during the pilot survey for the Hobby–Eberly Telescope Dark Energy Experiment (HETDEX) – called the HETDEX Pilot Survey (HPS) – which uses an integral-field-unit spectrograph and covers ∼3500−5800 Å at ∼5-Å resolution. I have newly determined the gas metallicities, |$12 + \text{log (O/H)}$|⁠ , following the Bayesian analysis scheme of the previous study, but dealing carefully with the uncertainty of strong-line calibration, performing reproducibility tests with mock data, and monitoring the convergence of the Markov chain Monte Carlo (MCMC) sampling. From the mock-data tests, I found that the nebular emission-line colour excess E (B − V) can be overestimated by as much as 2 σ or more, although the metallicity can recover the input value to within 1 σ. The new metallicity estimates on the HPS data are from well-converged MCMC samples (effective sample sizes > 2000), and they are higher than the previous estimates by ∼2 σ. Using the HPS data, I also showed that the MCMC sampling can have the statistical accuracy as poor as the one near the iteration start if done without convergence monitoring. The overestimation of E (B − V) indicates the overestimation of the star formation rates (SFRs) in the previous study, which can be as much as a factor of 5. This finding undermines the previous suggestion of a hitherto-unknown galaxy population based on the locations of galaxies in the mass–SFR plane. I found that the independent determination of E (B − V) using either H  β -H γ or H  α -H  β line pair is ideal for the analysis of forthcoming HETDEX data, but it requires additional cost. [ABSTRACT FROM AUTHOR]