Exploring Biases in Derived Stellar Parameters and the Ionizing Photon Production Efficiency

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Constraining the timescale and manner in which the Epoch of Reionization (EoR) occurred is a major JWST science goal. However, any constraints on the stellar or ionizing parameters ( ξ ion ) of galaxies in the EoR must contend with biases introduced by both the data and the models used. We explore three techniques that use spectroscopic and photometric data as well as three different spectral energy distribution (SED) fitting codes, each comprising multiple star formation history, stellar population synthesis, dust, and photoionization prescriptions to determine their relative influence on stellar parameters and ξ ion . We use z ∼ 3 EoR analog galaxies due to their reliable photometric coverage (improved physical constraints) in comparison to direct EoR sources and potential for research into direct Lyman continuum escape. For this population the median stellar mass can vary by over 0.6 dex and the star formation rate by more than 0.9 dex. Further, ξ ion can vary by over 1.1 dex for individual sources when comparing spectroscopic and photometric derivations, or by more than 0.5 dex when fitting SEDs with different models. As such, the choice of methodology can have significant consequences for the derived ξ ion and the subsequent sources of reionization. We find that the presence of a redshift evolution for ξ ion is dependent on the method adopted for its derivation, where a consistent method yields an evolutionary trend with redshift in extreme emitters while an indiscriminant selection of studies does not. The model, method, and data dependence of the ξ ion parameter are undeniable even for a homogeneous population.
All Related Versions
Funding Information
  • Australian Research Council (CE170100013)

This publication has 121 references indexed in Scilit: