JWST Advanced Deep Extragalactic Survey (JADES) Data Release 5: Photometric Catalog

TL;DR

The paper presents JADES DR5, a comprehensive photometric catalog product built from 35 space-based mosaics, integrating JWST NIRCam/MIRI and HST ACS/WFC3 imaging in GOODS-N/S. It introduces a full pipeline: detection on long-wavelength stacks, deblending with a short-wavelength deblending image, thorough source-cataloging, and advanced photometry including forced circular and Kron measurements, all tied together with a new pixel-level uncertainty model and common-PSF mosaics. A key novelty is the Gaussian-regression approach for fast, robust estimation of source sizes and Kron apertures, coupled with an uncertainty framework that accounts for correlated noise across heterogeneous mosaics. The DR5 release provides detailed quality flags, provenance hashes, and curve-of-growth measurements, and delivers photometric redshifts with EAZY, enabling broad, reliable studies of galaxy formation and evolution across cosmic time.

Abstract

JADES Data Release 5 (DR5) photometric catalogs and describes the methodologies used for source detection, deblending, photometry, uncertainty estimation, and catalog curation. The catalogs are constructed from 35 space-based imaging mosaics obtained with JWST/NIRCam, JWST/MIRI, HST/ACS, and HST/WFC3, combining approximately 1250 hours of JADES imaging with extensive additional public JWST and HST observations in the GOODS fields. Sources are identified using custom signal-to-noise-based detection and deblending algorithms optimized for the depth, resolution, and complex point-spread-function structure of JWST imaging. Source centroids, shapes, and photometric apertures are determined using a new fast two-dimensional Gaussian regression method applied to detection-image profiles. We provide forced circular-aperture photometry, ellipsoidal Kron photometry, and curve-of-growth measurements for every source in every band. We introduce a new pixel-level regression framework to model photometric uncertainties as a function of aperture size and local mosaic properties, accounting for correlated noise in heterogeneous JWST mosaics. Photometric redshifts are computed using template-based fitting applied to both small-aperture photometry on unconvolved images and Kron photometry on common-PSF mosaics. The JADES DR5 catalogs supersede previous JADES photometric releases, and are publicly released through the Mikulski Archive for Space Telescopes and an interactive web interface.

Figures (7)

• Figure 1: Overview of the detection and deblending process. The multi-band NIRCam imaging from J26 forms the core data used to detect and isolate sources, shown as an RGB composite of the JADES Origins Field region eisenstein2025a of GOODS-S (upper left). The long-wavelength NIRCam images are stacked to form a deep multi-band detection image that provides the pixel-level signal-to-noise ratio of sources (upper right; logarithmic projection), while a multi-band stack of short-wavelength NIRCam filters near $\lambda\sim2\mu$m are used to deblend sources. Using the detection and deblending methods described in Section \ref{['sec:detection']}, a segmentation map of pixel assignments to individual sources is created (lower left). The properties of these pixels are used in a new Gaussian regression model to determine source sizes and kron1980a photometry apertures for each source, shown as ellipsoidal overlays on a harsh stretch of the detection image in the lower right panel and described in Section \ref{['sec:source-catalog']}.
• Figure 2: Illustration of the interactive catalog curation tool in FitsMap hausen2022b applied to amend the JADES DR5 photometric catalog. Using the source catalog overlay (green ellipses) on the JWST mosaic (RGB), clearly spurious sources in need of curation, such as shredded diffraction spikes, can be identified visually (far left panel). Using the interactive selection tool in FitsMap (left center panel), sources within a polygonal bounding box (yellow line) can be selected (red ellipses) for further processing. Once selected, the object properties can be recorded and the sources marked (yellow ellipses in center right panel)) for curation (e.g., merging or deletion. In this case, the diffraction spike shreds are re-merged into the central star (blue ellipse in far right panel).
• Figure 3: Example of Gaussian regression modeling, shown for object ID=615 from the JADES DR5 GOODS-S region. The pixel SNR data from the detection image (center left panel), masked by the object segmentation. The regression method described in Section \ref{['sec:sb-model']} is used to fit a two-dimensional Gaussian profile to the source, as shown in the center right panel. The semimajor and semiminor axes of the best-fit are indicated in pixels, and the 1$\sigma$ isodensity contour shown as an overlaid white ellipse in the center left, center right, and far right panels. The goodness of fit can be judged from the residual of the SNR image less the model, as shown in the far right panel, and by over plotting the circularized SNR pixel data (blue points) and model (orange curve) as shown in the far left panel.
• Figure 4: Histograms of the source counts for 5$\sigma$-significant objects measured in $r=0.1$" circular apertures, for the 35 HST/ACS, HST/WFC3, JWST/NIRCam, and JWST/MIRI filters used in the JADES DR5 photometric catalogs. Shown in each panel are the source counts in the GOODS-N (blue) and GOODS-S (red) fields in bins of width $\Delta M=0.1$ magnitudes, and the filter mosaic measured in each panel is indicated in the upper left corner. All detected sources are considered for inclusion in each panel, but only sources with SNR$>5$CIRC1 photometry in a given filter are included in the histograms for that band. The source counts between GOODS-S and GOODS-N are consistent once differences in areal coverage are considered.
• Figure 5: Comparison between the JADES DR3 vs. DR5 photometry. Shown are the fractional flux differences in F090W, F115W, F150W, F200W, F277W, F335M, F356W, F410M, and F444W between the DR3 and DR5 $r=0.1$" circular aperture photometry (CIRC1), measured relative to the DR5 flux, for sources with DR5 SNR$>5$ in each filter, as a function of the CIRC1 AB magnitude in the filter. The distribution of flux differences are shown as normalized two-dimensional histograms (shaded regions). Running windowed medians of the histogram values are computed and reported as a blue line, and the median of the running medians is reported as an annotation in each panel. The typical flux differences between DR3 and DR5 for well-matched sources is better than 1%.