There are two kinds of unWISE catalogs: individual-image catalogs and band-merged catalogs. The individual catalogs are the outputs of crowdsource
on the single-band unWISE tiles. The band-merged catalogs are derived from the individual-image catalogs, merged so that different detections of the same source in different bands are collected together in the same catalog entry. Most users will want the band-merged catalogs.
All catalogs are available for download.
Download
- astrometry atlas
The astrometry atlas contains the names and astrometry for each coadd composing the unWISE imaging. This is useful to see exactly which images contribute to different parts of the footprint.
- individual image catalogs
The individual image catalogs are available for every unWISE coadd (36480 files; 18240 tiles and two bands). The tabulated fields are described below. Sizes range from 3 to 20 MB, totaling 373 GB.
The entire individual image catalog can be downloaded via
wget -I unwise/release/cat/ --reject "index.html*" -nH --recursive http://faun.rc.fas.harvard.edu/unwise/release/cat/
. - band-merged catalogs
The band-merged catalogs are very similar to the individual-image catalogs, except most fields have been transformed into two-element vectors, corresponding to the WISE 3.4 and 4.6 micron bands. There is one file for each of the 18240 tiles. The fields are described below. Sizes range from 20 to 60 MB, totaling 541 GB.
The entire object catalog can be downloaded via
wget -I unwise/release/band-merged/ --reject "index.html*" -nH --recursive http://faun.rc.fas.harvard.edu/unwise/release/band-merged/
.
Individual-image Catalogs
crowdsource
analyzes each single-band unWISE tile individually to determine the position and flux of all sources in the image. The resulting output catalog has the following fields:
x | f8 | x coordinate (pix) |
y | f8 | y coordinate (pix) |
flux | f4 | flux (Vega nMgy) |
dx | f4 | uncertainty in x |
dy | f4 | uncertainty in y |
dflux | f4 | uncertainty in flux (statistical only) |
qf | f4 | "quality factor" |
rchi2 | f4 | average \(\chi^2\) per pixel, weighted by PSF |
fracflux | f4 | fraction of flux in this object's PSF that comes from this object |
fluxlbs | f4 | local-background-subtracted flux (Vega nMgy) |
dfluxlbs | f4 | uncertainty in local-background-subtracted flux |
fwhm | f4 | full-width at half-maximum of PSF (pixels) |
spread_model | f4 | SExtractor spread_model parameter |
dspread_model | f4 | uncertainty in spread_model |
flags_unwise | i4 | unWISE Coadd Flags flags at central pixel |
flags_info | i4 | additional informational flags at central pixel |
sky | f4 | sky (Vega nMgy) |
ra | f8 | right ascension (degree) |
dec | f8 | declination (degree) |
unwise_detid | i8 | unique ID |
nm | i4 | number of single-exposure images of this part of sky in coadd |
primary | i2 | the center of this source is in the primary region of its coadd |
The "quality factor" is a concept inherited from PS1. It tabulates the PSF-weighted fraction of pixels contributing to this detection. Good detections will have quality factors close to one. Heavily saturated detections will have quality factors close to zero.
The x and y coordinates are defined so that in Python, image[x, y]
gives the value of the pixel (x, y). This means that in terms of layout in memory, the y index increases faster than the x index. This is in contrast to the usual convention, so caveat emptor.
Band-merged Catalogs
Most columns in the unWISE band-merged catalogs are identical to the individual-image catalogs. In these cases, the columns are simply two-element vectors corresponding to the W1 and W2 measurements of a source. These columns are not documented further. A few columns are added, however, and are documented below.
Field Name | Data Type | Description |
---|---|---|
ra12, dec12 | f8 | Positions from individual-image catalogs |
ra, dec | f8 | W1 position, if available; otherwise W2 position |
primary12 | i2 | 'primary' status from individual-image catalogs |
primary | i2 | W1 primary status, if available; otherwise W2 primary status. |
unwise_objid | a16 | unique object id |
Flux Scale
All fluxes are in "Vega nMgy", so that the Vega magnitude of a source is given by \(22.5-2.5\log(\mathrm{flux})\). The absolute calibration is ultimately inherited from AllWISE through the calibration of Meisner et al. (2017). This inheritance depends on details of the PSF normalization at large radii, which is uncertain. To improve the agreement between unWISE and AllWISE fluxes, we recommend subtracting 4 mmag from unWISE W1 and 32 mmag from unWISE W2 fluxes.
Corresponding AB magnitudes are given by the following equations:
- \(m_{\mathrm{W1}, \mathrm{AB}} = m_{\mathrm{W1}, \mathrm{Vega}} + 2.699\)
- \(m_{\mathrm{W2}, \mathrm{AB}} = m_{\mathrm{W2}, \mathrm{Vega}} + 3.339\)
unWISE Flags
The unWISE flags simply give the value of the unWISE mask image for the central pixel of the unWISE catalog source, after "collapsing" the time-resolved masks into a smaller number of bits appropriate for static sources. All bits mark features of very bright stars. These flags are described in more detail here.
Bit | Description |
---|---|
0 | In core or wings |
1 | In diffraction spike |
2 | In ghost |
3 | In first latent |
4 | In second latetnt |
5 | In circular halo |
6 | Saturated |
7 | In geometric diffraction spike |
Additional Flags
The unWISE info flags contain additional fields related to the crowdsource
processing or unWISE coadd bright star mask. These flags are described in more detail here.
Bit | Description |
---|---|
0 | In PSF of bright star falling off coadd |
1 | In HyperLeda large galaxy |
2 | In "big object" (e.g., a Magellanic cloud) |
3 | Pixel may contain the centroid of a very bright star |
4 | crowdsource considers this pixel potentially affected by saturation |
5 | Pixel may contain nebulosity |
6 | Sources in this pixel will not be aggressively deblended |
7 | Candidate sources in this pixel must be "sharp" to be optimized |