Escaping ionizing photons from massive spiral galaxies at $z\sim 1$

TL;DR

This study targets the escape of ionizing photons from massive spiral galaxies at $z\sim1$ by identifying LyC leakers in the AUDFs/UVIT and MUSE HUDF datasets. Focusing on MUSE ID 16, a likely star-forming, non-AGN spiral, the authors combine multiwavelength imaging and spectroscopy to measure ionizing-photon production ($\xi_{\rm ion}$), nebular and stellar dust attenuation, and LyC escape fractions ($f_{\rm esc}$), finding $f_{\rm esc}$ values of roughly $0.28$–$0.39$ and a substantial contribution from a vertically oriented escape channel. They report tentative evidence that face-on disk orientations enhance LyC detectability, supporting anisotropic escape in stratified disks. The results highlight the importance of geometry and feedback in LyC leakage and provide a new observational window into escape mechanisms in disk galaxies at intermediate redshift. These findings have implications for models of cosmic reionization that must account for anisotropic LyC escape and orientation effects in disk galaxies.

Abstract

We report the detection of Lyman continuum (LyC) photons from three massive ($\text{M}_{*}>10^{10}\:\text{M}_{\odot}$) spiral galaxies at a redshift of nearly 1 in the AstroSat UV Deep Field South. Notably, all three systems are viewed at low inclination (i.e., nearly face-on), prompting an investigation into the role of galaxy orientation in the detectability of LyC emission from disk systems. Two of the three galaxies, however, host active galactic nuclei (AGNs), adding complexity to the interpretation of the LyC signal. We present a detailed analysis of the likely star-forming case, and report tentative evidence that a face-on viewing angle may enhance the likelihood of LyC detection in disk galaxies. This represents the first detection of LyC emission from well-characterized spiral galaxies at high redshift, offering a new window into LyC escape mechanisms in such systems. Our findings highlight the need to consider geometric factors and anisotropic escape pathways facilitated by feedback processes alongside more traditional density-bounded scenarios that imply isotropic escape.

Figures (6)

• Figure 1: Multiband imaging of MUSE ID 16. The RGB image ($5"\times5"$ size) is composed using the HST F336W (blue), JWST F115W (green), and JWST F444W (red) band images, which are also shown individually in the bottom panel in a zoomed-in view ($2"\times2"$ size). The yellow aperture in the upper panel is centered on the F444W band image centroid and is of radius 1.2$^{"}$. The cyan arrows in the RGB image highlight the blue star-forming regions identified within the galaxy. The first three panels from the left in the bottom row show MUSE ID 16 in the bluest available HST bands, covering a rest-frame wavelength range of 677–1489 Å, while the last two panels display the galaxy in JWST/NIRCam F115W and F444W bands, probing optical to near-infrared rest-frame wavelengths. The red aperture of radius 0.15$"$ marks the emission complex observed in the HST F150LP band, which is likely associated with the galaxy. The signal within this aperture is detected at a significance of $\sim2.8\sigma$ (see Section \ref{['sec:morph']} for details). The blue dashed contour marks the 15$\sigma$ isophote in the F444W image and is shown for visual reference. Note that the UVIT F154W and the HST F150LP band images displayed have been smoothed using a Gaussian kernel of FWHM 3 pixels for visual clarity.
• Figure 2: Panel a) shows the MUSE 1D spectrum of MUSE ID 16 in the rest-frame, highlighting wavelengths redward of the [O II] doublet. The lines which we have used for analysis in this work are indicated with the blue dashed lines. The grey spectrum denotes the error spectrum. Panel b) presents the H$\alpha$ emission line map derived from HST G141 grism data. Panel c) displays the 2D JWST/NIRCam grism spectrum, zoomed in on the Pa$\alpha$ emission complex. White arrows indicate the spectral (increasing wavelength) and spatial directions.
• Figure 3: The best-fit SED of MUSE ID 16, showing the observed model (black dashed), the intrinsic unattenuated stellar component (green dashed), and the dust-attenuated stellar component (cyan), overlaid with the measured photometric data points (blue circles). The error in the x-axis denotes the effective width of the photometric broad bands. The blue shaded region represent the $16^{\text{th}}-84^{\text{th}}$ percentiles around the best-fit SED. The reduced $\chi^2$ of the best-fit is 0.72.
• Figure 4: Panel a) displays the star-formation rate surface density, ($\log(\Sigma_{\rm{SFR}}/\rm{M}_{\odot}\:\rm{yr}^{-1}\:\rm{kpc}^{-2})$, versus the UV half-light radius, ($\log(r_{50}/\rm{kpc})$), for MUSE ID 16 (red markers), compared to the low-redshift Lyman-continuum leaker pulus sample (LzLCS+) from Flury22a (blue and grey markers), with updated structural measurements adopted from LeReste25. Panel b) shows the distribution of LyC detections (blue, red, and magenta markers) and non-detections (inverted grey triangles) from our sample (Section \ref{['sec:selection']}) in the inclination–Sérsic index plane. The axis ratios and Sérsic indices are taken from vanderWel12. Galaxies are classified as near face-on or near edge-on disks according to the criteria described in Section \ref{['sec:inclination']}.
• Figure 5: Simplified cartoon schematic illustrating the escape of ionizing photons from a star-forming spiral disk galaxy viewed nearly face-on, representing the plausible scenario for MUSE ID 16. LyC photons are depicted emerging in a conical fashion from a density-bounded, oblate HII region. In contrast, ionization-bounded star-forming regions remain embedded within dust and neutral gas. Panel b) highlights the blueshifted Fe II$\lambda$2383 absorption relative to the systemic redshift determined from the [O II] doublet, indicative of cool galactic winds and outflowing, entrained gas along the line of sight. Note that the CGM and molecular clouds are not included in this schematic.