WISE photometry for 400 million SDSS sources

TL;DR

This paper introduces forced photometry of WISE data at SDSS source positions, leveraging The Tractor and unWISE coadds to extract fluxes for hundreds of millions of SDSS sources. By fixing calibrations and modeling sources with SDSS-informed profiles, it achieves consistent cross-survey flux measurements and recovers many sub-detection-threshold WISE fluxes, expanding the useful catalog beyond the standard AllWISE entries. The approach improves cross-match completeness, enables stacking analyses, and provides insights into blending and PSF-related artifacts, with results closely aligning with AllWISE for isolated point sources while highlighting calibration-like tilts between releases. The method offers a scalable, information-rich way to fuse optical and infrared surveys, with potential for future joint-fitting improvements.

Abstract

We present photometry of images from the Wide-Field Infrared Survey Explorer (WISE; Wright et al. 2010) of over 400 million sources detected by the Sloan Digital Sky Survey (SDSS; York et al. 2000). We use a "forced photometry" technique, using measured SDSS source positions, star-galaxy separation and galaxy profiles to define the sources whose fluxes are to be measured in the WISE images. We perform photometry with The Tractor image modeling code, working on our "unWISE" coaddds and taking account of the WISE point-spread function and a noise model. The result is a measurement of the flux of each SDSS source in each WISE band. Many sources have little flux in the WISE bands, so often the measurements we report are consistent with zero. However, for many sources we get three- or four-sigma measurements; these sources would not be reported by the WISE pipeline and will not appear in the WISE catalog, yet they can be highly informative for some scientific questions. In addition, these small-signal measurements can be used in stacking analyses at catalog level. The forced photometry approach has the advantage that we measure a consistent set of sources between SDSS and WISE, taking advantage of the resolution and depth of the SDSS images to interpret the WISE images; objects that are resolved in SDSS but blended together in WISE still have accurate measurements in our photometry. Our results, and the code used to produce them, are publicly available at http://unwise.me.

Figures (10)

• Figure 1: PSF models from the AllWISE Release, from meisner, and our three-component, zero-mean, isotropic, concentric Gaussian fits. Note the log scale. The AllWISE and meisner plots are horizontal (solid) and vertical (dashed) slices through the PSF center. Our mixture-of-Gaussian models capture the PSF cores---roughly three orders of magnitude---quite effectively, but degrade somewhat at large radii and low flux levels.
• Figure 2: Example forced photometry results. Top-left: SDSS $r$-band image, with point sources (small red) and galaxies (large orange) marked. (Some of these sources will have been detected in the other SDSS bands.) These are the sources that will be photometered, along with sources from the WISE catalog that are detected only in WISE. Top-middle: WISE W1 image, smoothed by resampling. Top-right: WISE W1 image at original (native) resolution. Bottom-left: WISE W1 image, with WISE catalog detections (green cross) and SDSS sources marked. Notice that the central source is detected as a single source in WISE, but resolved into two sources by SDSS. Also notice a number of WISE-only and SDSS-only detections. Bottom-middle: Forced photometry model image. This is a weighted sum of WISE PSF models (convolved by the galaxy profile for the one galaxy in this field) at the positions of SDSS and WISE-only sources, with weights chosen to minimize the chi-squared residuals from the WISE W1 image. Bottom-right: The model image at native resolution, plus per-pixel noise equal to that in the real image. The model is a good approximation to the real image, given the observational noise.
• Figure 3:
• Figure 4: Comparison of our forced photometry magnitudes and WISE AllWISE Release catalog magnitudes, in W1, for matched sources within 4 arcseconds, in $\sim 100$ square degrees of sky around RA,Dec = $(180, 45)$. We show only unique (one-to-one) matches, since otherwise the Tractor photometry resolves sources that are blended in WISE. Left: All sources; Middle: Sources identified in the SDSS imaging as galaxies, and not treated as point sources in our photometry; Right: sources identified by SDSS as point-like, plus nominally extended sources treated as point sources in our photometry. The error bars shown are the median WISE catalog error bars per magnitude bin. The WISE catalog magnitude entry we are plotting is w1mpro, a point-source measurement; this explains why the Tractor measurements of galaxies tend to be brighter: we measure all the galaxy's flux, while the WISE catalog only measures the fraction in the point-like core.
• Figure 5: Comparison of our forced photometry magnitudes and WISE AllWISE Release catalog magnitudes, for sources treated as point sources in our photometry. These are sources in the same $\sim 100$ square degrees in \ref{['fig:comp']}. The slight tilts seen are comparable in magnitude to the differences between the All-Sky and AllWISE releases, and may be due to differences in photometric calibration between the releases.