Faint galaxies in the Zone of Avoidance revealed by JWST/NIRCam

TL;DR

This study demonstrates that JWST/NIRCam can directly detect and morphologically classify background galaxies through heavily obscured regions of the Milky Way, addressing the Zone of Avoidance gap in large-scale structure mapping. Using deep F090W, F200W, and F444W imaging of NGC 3324 and a tailored SExtractor-based pipeline, the authors build a robust photometric catalog and validate it against DAOPHOT PSF photometry. They identify 102 background galaxies behind the molecular cloud, including a potential compact galaxy association at $z sim 0.3$--$0.4$ and transnebular galaxies seen through dense extinction. The results establish a path to systematic ZoA mapping and highlight the planned synergy with Roman for wide-area, high-resolution coverage.

Abstract

The Zone of Avoidance (ZoA) remains one of the last frontiers in constructing a comprehensive three-dimensional map of the Universe. Galactic extinction, stellar crowding, and confusion noise have historically limited the detection of background galaxies in these regions, with implications for large-scale structure and cosmological measurements. We assess the capability of the James Webb Space Telescope (JWST) Near Infrared Camera (NIRCam) to detect extragalactic sources in a heavily contaminated region of the Milky Way. We analyzed JWST/NIRCam wide-filter images of NGC 3324 with a customized implementation of SExtractor v2.28. Sources were detected in the F444W band, cross-matched with F090W and F200W, and validated against recent DAOPHOT point spread function (PSF) photometry. A refined sample was obtained through full width at half maximum (FWHM) - signal-to-noise ratio (SNR) criteria and visual inspection. We identified 102 galaxies across the JWST/NIRCam field of view. The magnitude (F444W) distribution is bimodal, with about 10% brighter than m_F444W < 15 mag and about 60% in the range 17 < m_F444W < 19 mag. Typical sizes are FWHM ~6.5 arcsec, from compact to extended systems with isophotal areas up to ~2000 pixels (~7.9 arcsec^2). Morphologies span from compact to spiral and lenticular systems, including a compact group at the eastern edge of the field. We also report the detection of "transnebular galaxies", visible through the most opaque regions of the molecular cloud. These results demonstrate the potential of JWST/NIRCam to probe extragalactic sources through highly obscured Galactic regions, opening new avenues for mapping large-scale structures across the ZoA.

Figures (13)

• Figure 1: Photometric calibration for the F090W (top), F200W (middle), and F444W (bottom) bands. Matched sources are shown as faint red dots. The subset of point-like sources (eccentricity $< 0.01$) used for calibration is indicated by red circles. Blue circles with error bars represent bin-averaged values computed via the Binscatter method. The lower inset zooms into the region used for regression, delimited by dashed lines. Horizontal dashed lines indicate the linear fits used to compute the final magnitude offsets.
• Figure 2: Completeness as a function of apparent magnitude in the F444W band. The blue curve shows the recovery fraction of 5000 injected synthetic galaxies, with shaded regions indicating the 95% confidence interval derived from binomial statistics. Horizontal dashed lines mark the 80% and 50% completeness thresholds. The inset panel shows the magnitude distribution of the injected sources, sampled from a beta distribution to mimic the shape of the observed luminosity function.
• Figure 3: The parameter space used to distinguish point-like from extended sources is shown. Blue dots represent all detections in the field, while red circles indicate sources matched with Gaia stars. Vertical and horizontal lines mark the third quartile of the distributions in FWHM and SNR_WIN, respectively, providing empirical thresholds for source separation.
• Figure 4: Mutual information matrix among morphometric parameters. Color intensity reflects the degree of statistical dependence, from 0 (independent) to 1 (fully dependent).
• Figure 5: Boxplot summary of CLASS_STAR, ELLIPTICITY, ECCENTRICITY, ISOAREA_IMAGE across the four sectors. Orange notches mark median values. Outliers are shown as open circles. Color codes correspond to sector classification as described in the text.