Unveiling the Ionized and Neutral ISM at z > 10 : The Origin of [O III] /[C II] Ratios from a Sub-parsec Resolution Radiative Transfer Simulation (2505.12397v1)
Abstract: Recent multi-wavelength observations by JWST and ALMA are unveiling both ionized and neutral ISM components in high-redshift ($z>6$) galaxies. In this work, we investigate the origin of rest-frame far-infrared [OIII]88 $\mu$m and [CII]158 $\mu$m emission by performing zoom-in cosmological simulations of dwarf-galaxy progenitors at $z=9-13$. Our simulations incorporate on-the-fly radiative transfer at sub-pc ($\sim$ 0.1 pc) resolution, allowing us to resolve the multi-phase ISM. We compute emission lines on a cell-by-cell basis, taking into account local temperature, density, metallicity, radiation field strength, column density, and spectral hardness of radiation bins. We find that [OIII] predominantly arises from centrally located ionizing bubbles with temperatures of $\sim (1-5)\times 104\,\mathrm{K}$ and high ionization parameters of $\log U_{\mathrm{ion}} \simeq -1.5$. In contrast, [CII] is produced in the surrounding dense neutral regions at $\sim 5\times 103\,\mathrm{K}$, which are heated by strong FUV radiation from the central stellar clusters. This spatial arrangement leads to large local variations in [OIII]/[CII], ranging from $\sim$ 100 to 0.01. Our galaxy reproduces the global ratio [OIII]/[CII]$\sim5-30$, consistent with recent ALMA detections at $z>6$ without invoking enhanced O/C abundance ratios. We further derive that [OIII]/[CII] linearly scales with the mass and density ratios of ionized to neutral gas, $M_{\rm HII}/M_{\rm HI}$ and $n_{\rm HII}/n_{\rm HI}$ and show that the [OIII]/[CII] ratio typically changes from 5.7 to 0.3 from high-z to low-z. For future synergies of JWST and ALMA, we derived $M_{\rm HII}/M_{\rm HI}$ for observed $z >6$ galaxies using ${\rm H}\beta$ and [CII] and show the validity of our scaling relations.
Collections
Sign up for free to add this paper to one or more collections.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.