II. User's Guide to the WISE Preliminary Data Release

2. Source Catalog

A summary of the properties of the Source Catalog and a detailed Catalog format description are presented below. More detailed descriptions of photometric and astrometric characteristics of the Catalog are presented in VI. Descriptions of processing algorithms used to generate the the Source Catalog can be found in IV.4.

Users are strongly encouraged to read the Cautionary Notes before using the Catalog.

a. Format and Column Descriptions

b. Source Selection Criteria

The WISE Preliminary Release Source Catalog contains positions and four-band photometry for 257,310,278 objects extracted from 10,464 Tiles covering approximately 57% of the sky.

Sources in the Catalog were drawn from a Working Database (WDB) of all detections made on the coadded Atlas Images that was generated during Multiframe pipeline processing. The WDB contains 352,839,115 entries, significantly more than the Source Catalog. Many WDB entries are spurious detections of noise and image artifacts. The reliability of the WISE Source Catalog was achieved by selecting entries in the WDB that satisfy the criteria that are described in V.3 and summarized in Table 1.

c. General Properties

• Total Number of Objects in the Catalog:     257,310,278


• Number of Objects With >2σ Detections in Each Band:



Table 2

Band	Number of Objects	Percentage of Total
W1	257,194,955	>99.9
W2	244,285,878	94.9
W3	73,187,473	28.4
W4	18,854,790	7.3

• Band-Detection Combinations (>2σ per Band)

Table 3

Band-Combination	det_bit	Number	Percentage of Total
W1-W2-W3-W4	15	13,357,219	5.2
W2-W3-W4	14	16,205	<0.1
W1-W3-W4	13	95,715	<0.1
W3-W4	12	65,143	<0.1
W1-W2-W4	11	5,020,804	2.0
W2-W4	10	835	<0.1
W1-W4	9	296,798	0.1
W4	8	2,071	<0.1
W1-W2-W3	7	58,555,788	22.8
W2-W3	6	5,538	<0.1
W1-W3	5	1,071,550	0.4
W3	4	20,315	<0.1
W1-W2	3	167,324,273	65.0
W2	2	5,216	<0.1
W1	1	11,472,808	4.5



• Sky Coverage

  The approximate sky coverage of the Source Catalog is 23,460 deg², after taking into account coverage loss due to poor frame data quality, and frame and pixel outlier rejection (VI.2). The Catalog has an additional loss of effective coverage because Sources were required to fall >50" from Tile edge during Catalog source selection (V.3). At the outer boundaries of the Release area where there are no overlapping Tiles, Catalog sources will not extend all the way to the unbounded Tile edges.



• Distribution of Catalog Sources with Galactic Latitude

The cumulative number of Preliminary Release Source Catalog entries and the approximate coverage area as a function of galactic latitude are listed in Table 5. However, the distribution of sources with latitude is a complex function of brightness in each band. Figures 1-4 are animated GIF images that show the surface density of sources in 0.5 magnitude bins in progressively fainter magnitude slices. Completeness in the Galactic Plane begins to fall off quickly in the short wavelength bands because of confusion in high source density regions. The distribution of sources at higher Galactic latitudes remains relative uniform until W1<16.5, W2>16.0, W3>12.5 and W4>9.0 mag. For fainter magnitudes, the non-uniform depth-of-coverage in the Survey Strategy begins to dominate the source density distributions.


Table 5 - Cumulative Source Counts

|b| < (deg)	Number	Percentage of Total	Approx. Area (deg2)
2°	14,473,367	5.6	1053.3
5°	36,315,936	14.1	2660.8
10°	72,766,813	28.3	5356.8
15°	108,158,687	42.0	7969.9
20°	139,438,621	54.2	10530.0
30°	190,530,346	74.0	15398.4
40°	226,439,071	88.0	19403.8
50°	246,197,694	95.7	21873.0
60°	254,652,111	99.0	23048.9
70°	257,302,403	>99.9	23457.2
80°	257,310,278	100.0	23457.2
90°	257,310,278	100.0	23457.2





Figure 1 - W1: Range = 0-200	Figure 2 - W2: Range = 0-190
Figure 3 - W3: Range = 0-130	Figure 4 - W4: Range = 0-300
Click on the thumbnails to view animated GIF images that show Source Catalog surface density maps for each band in 0.5 mag intervals at progressively fainter levels. The maps are Hammer projections in galactic coordinates with spatial bin sizes of 0.2°x0.2° (0.5°x0.5° in W4). The center of the maps corresponds to galactic coordinates 0°,0°. Galactic north is towards the top of the maps and galactic longitude increases to the right.

• Differential Source Counts

The WISE Source Catalog is >95% complete down to W1=16.6. W2=16.0, W3=10.8 and W4=6.7 mag in unconfused regions of the sky with ≥12 frame depth-of-coverage. Completeness decreases for fainter sources, in lower coverage regions, and in regions of high source density and/or complex backgrounds.

The top panels of Figures 5, 6 and 7 show examples of source count log histograms in the WISE bands for three representative regions in the Preliminary Release Source Catalog. Figures 5 and 6 show mid-galactic latitude fields with 16 and 24 average frame depth-of-coverage. Figure 7 shows the counts for a field in the Galactic plane with very high source density. The spatial distribution and variation of the magnitude at which the peaks of the Catalog source count histograms occur are shown in Figures 8-11. The impact on the effective depth of the Catalog by survey depth-of-coverage is easily seen in all bands by the gradient in depth that increases from the ecliptic plane to the poles. The loss of depth due to confusion in the Galactic Plane is most prominent in W1 and W2.

Figure 5 - (top) Differential WISE Source Catalog counts in a 116 deg2 near l,b=225°,-55° with average frame depth-of-coverage of ~16. (bottom) Average photometric uncertainty as a function of source brightness.	Figure 6 - (top) Differential WISE Source Catalog counts in a 125 deg2 near l,b=303°,-27° with average frame depth-of-coverage of ~24. (bottom) Average photometric uncertainty as a function of source brightness.	Figure 7 - (top) Differential WISE Source Catalog counts in a 116 deg2 near l,b=338°,-1° with average frame depth-of-coverage of ~16. (bottom) Average photometric uncertainty as a function of source brightness.




Figure 8 - W1	Figure 9 - W2
Figure 10 - W3	Figure 11 - W4
False-color maps showing the spatial distribution and variation of the peak of the differential source count histograms computed in 0.2x0.2 deg spatial bins. The color scale in the images gives the peak magnitude color encoding for each band. The Maps are Hammer projections in ecliptic coordinates. The centers correspond to ecliptic coordinate 180°,0°. Ecliptic north is towards the top, and ecliptic longitude increases towards the right.

• Saturation and Bright Source Detection Limits

Listed in Table 6 are the approximate magnitudes for which point sources began to saturate the WISE detectors. Photometry is performed for sources brighter than the saturation limits by PSF fitting to the non-saturated wings of the source profiles. The quality of the photometry for saturated sources is progressively degraded. Brighter than the extraction limit levels given in Table 6, saturation sources could no longer be reliably extracted and may be missing from the Catalog.

Table 6 - Point Source Saturation and Bright Source Extraction Limits

Band	Saturation Limit1 (mag)	Extraction Limit2 (mag)
W1	8.0	1.0
W2	6.7	0.0
W3	3.8	-2.0
W4	-0.4	-6.0
Notes: 1    Approximate values. Onset of saturation can vary by ~1mag depending on BG and sub-pixel location (see VI.4.d) 2    Approximate values. Accuracy of photometry for heavily saturated sources may be poor (see VI.4.c.i)

d. Photometric Properties

• Photometric Accuracy and Sensitivity

Photometry of bright, non-saturated sources in the Preliminary Release Source Catalog has an accuracy of ~2% in W1, W2 and W3, and ~3% in W4 (VI.4.b). The photometric SNR >5 for sources with W1<17.0, W2< 15.6, W3<11.5 and W4<7.9 mag in the unconfused regions of the sky with at least 8 independent coverages. The characteristic sensitivity of the Source Catalog varies significantly around the sky because of the variable depth-of-coverage, background emission levels and source confusion.

The bottom panels of Figures 5, 6 and 7 show the average quoted W1, W2, W3 and W4 profile-fitting photometric uncertainties in 0.2 magnitude bins for three representative regions in the Catalog. The vertical bars show the RMS dispersion in the mean uncertainties in each brightness. Photometric uncertainties increase at the bright end due to the onset of saturation. Uncertainties also increase towards fainter flux levels as photon noise increasingly dominates the measurements.

Figures 12-15 show the spatial distribution of the average magnitudes within each Atlas Tile for which the mean measurement uncertainty is 0.155 mag, corresponding to 7σ (see VI.4.a).

Figure 12 - W1	Figure 13 - W2
Figure 14 - W3	Figure 15 - W4
Equatorial projection sky maps showing the spatial variation of average magnitude at which the average source SNR=7 in the Preliminary Release Source Catalog (see VI.4.a)

• Band-Band Photometric Performance

W1-W2-W3 (3.4-4.6-12 μm) color-color diagrams drawn from high and low galactic latitude regions in the Preliminary Data Release Source Catalog are shown in Figures 16 and 17, respectively. The locations in color-space for various classes of astrophysical objects are shown in the annotated version of Figure 16.

Figure 16 - WISE 3.4-4.6-12 μm color-color diagram for a 116 deg2 region at l,b=225°,-55°. Green contours trace the density of sources in color bins. Click here for an annotated version of the color-color diagram showing the location of different classes of objects.	Figure 17 - WISE 3.4-4.6-12 μm color-color diagram for a 116 deg2 region at l,b=338°,-1°. Green contours trace the density of sources in color bins.

e. Astrometric Properties

WISE Source Catalog positions are reconstructed with respect to the 2MASS Point Source Catalog reference frame. The accuracy of the astrometric solution for the sources brighter than W1~13.0 mag is better than is 0.2 arcsec (see Figure 1 in VI.5).

Approximately 25% of the source fainter than W1~14.5 mag have declinations coordinates that can be systematically offset from their true positions by 0.2" to 1." This bias can be seen in a decreasing fraction of sources as bright as 13.0 mag, and the sign of the offset alternates with scan direction and equatorial. The bias was caused by an identified source extraction software error that failed to apply a 0.5 Atlas Image pixel declination offset to some sources when performing simultaneous PSF-fitting with other detections in close proximity (i.e. passive deblending). The error does not affect right ascension measurements. See VI.5 for illustrations and further descriptions of the impact of this error.

There was insufficient time to correct this bias prior to the Preliminary Data Release. Therefore, the quoted declination uncertainties in the Source Catalog, sigdec, have been inflated to reflect the presence of the bias. The uncertainties were adjusted by adding 0.5" in quadrature to the statistical declination measurement uncertainty (see VI.5.b). This yields a maximum absolute deviation for the distribution of uncertainties equal to the what would be produced if the declination error distribution was Gaussian.

Last Updated: 2011 December 28