There are no requirements on the astrometric accuracy for the 3-Band Cryo data release. The pipeline reduction of the WISE 3-Band Cryo data, however is similar to that of WISE All-Sky Data Release processing. Since the astrometric solutions derive primarly from W1 data, which has nearly identical performance to the fully cryogenic survey, the 3-Band astrometric positions should be of comparable quality to the All-Sky Products (described in Section VI.4). This section characterizes the 3-Band Cryo astrometric results i) relative to the 2MASS reference frame as was done for the All-Sky Release and ii) relative to the All-Sky Release itself providing an evaluation of internal WISE vs. WISE astrometric consistency and thus limits on internal measurement of proper motions with WISE data alone.
The All-Sky Release Level 1 specifications require that WISE astrometric positions be consistent with the 2MASS reference frame to <0.5 arcsec for sources with SNR > 20. Section VI.4 shows that the All-Sky Release products meet this requirement. Bright sources in the All-Sky Release Catalog, with position uncertainties dominated by systematics rather than noise, have RMS distribution of < 0.2 arcsec. The 3-Band Cryo data products share this level of precision with astrometric behavior essentially identical to the All-Sky Release (figures 1 and 2). This similarity between the All-Sky and 3-Band Cryo astrometry is not surprising given that the astrometric solutions derive from the shortest WISE wavelengths whose performance is largely unaffected by the exhaustion of the outer cryogen tank (or even the complete exhausion of WISE cryogens).
 |
 |
| Figure 1 - Positional difference in Right Ascension relative to 2MASS catalog position for bright (12.5 < W1<13.0 mag) WISE 3-Band Cryo WDB sources. The best fit Gaussian distribution has sigma=0.12 arcsec while the distribution itself has RMS=0.27 arcsec. The tails of the the actual distribution that extend beyond the Gaussian are most likely due to proper motion between the 2MASS and WISE epochs. The widths of the distribution above slightly exaggerate the WISE contribution as 2MASS positional uncertainty for these sources is ~0.10 arcsec | Figure 2 - Positional difference in Right Ascension relative to 2MASS catalog position for WISE 3-Band Cryo WDB sources with SNR~20 (15.5<W1<16.0 mag). The best fit Gaussian distribution has sigma=0.26 arcsec while the distribution itself has RMS=0.43 arcsec. The tails of the the actual distribution that extend beyond the Gaussian are most likely due to proper motion between the 2MASS and WISE epochs. |
The All-Sky Release Catalog analysis also included a Tile-by-Tile evaluation of the mean position offset of all of the brighter 2MASS stars in a given tile, mapping out these offsets over the entire sky. In the All-Sky analysis small spatially systematic deviations of up to 30 milliarcseconds were evident, attributable to the accumulated proper motion of the astrometric reference stars between the 2MASS and WISE epochs. Figures 3 through 6 compare these All-Sky offsets with those obtained from the 3-Band Cryo data. As expected the same trends are evident in the 3-Band results and overall mean offsets from the 2MASS reference frame are similar in amplitude and spatial bias.
 |
 |
| Figure 3 - Mean R.A. difference, Tile-by-Tile, between WISE All-Sky Release Catalog positions and 2MASS positions for all sources on a given Tile with W1 SNR > 20 and w1rchi2 < 2. The projection is in galactic coordinates and the color coding shows position offset in milliarcseconds. | Figure 4 - Same as Figure 3, but for the WISE 3-Band Cryo Release. Sky coverage is thus only partial. The absolute color scaling for Figures 3 and 4 are different, but the overall sense and amplitude of the offsets (which are typically smaller than 30 milliarcseconds) is consistent between the two figures. |
 |
 |
| Figure 5 - All-Sky Catalog Release Tile-by-Tile position offsets relative to 2MASS as described in Figure 3, but here in the Declination direction. | Figure 6 - 3-Band Cryo Release Tile-by-Tile position offsets relative to 2MASS. Once again the magnitude and sense of the small deviations is the same in the All-Sky and 3-Band Cryo data products. |
The WISE All-Sky Data Release itself provides an astrometric reference separated in epoch by only 6-months from the 3-Band Cryo release. The comparison between WISE epochs reveals positions that are significantly more consistent than the 2MASS/WISE comparison with the RMS for bright sources being less than 50 millarcseconds rms. The better astrometric consistency is attributable to the much smaller epoch difference between the two astrometric comparison samples as well as the common mode rejection of any systematics in the WISE astrometric pipeline (although it is important to note that the scan direction for the 3-Band Cryo data is opposite that for the All-Sky data). Figures 7 and 8 show position differences between the All-Sky and 3-Band Cryo epoch both as a function of magnitude (Figure 7) for sources with w1snr > 20 and as a scatter diagram in R.A. and Dec (Figure 8) for sources with w1snr > 40.
 |
|
| Figure 7 - Declination difference between stars common to the All-Sky and 3-Band Cryo data releases as a function of W1 magnitude selected to have w1snr > 20. The astrometric scatter is smaller than that resulting from the WISE-2MASS position comparison likely due to both the small epoch difference (6 months) between the All-Sky and 3-Band Cryo observations (and thus the suppression of significant proper motion) as well as comparing similar data reduced with a common pipeline. | Figure 8 - A scatter diagram of R.A. and Declination position differences for brighter sources (w1snr > 40) demonstrating the limiting positional precision for WISE extraction between the two WISE epochs. |
As the focal plane temperature changed after the exhaustion of the outer cryogen tank, and thus during the acquisition of the 3-Band Cryo data, measureable distortion was induced in the images in W3. For this band the rising focal plane temperature led to increased background and required shorter integration times than standard in order to avoid saturation. Shorter integration times, in turn, correspond to usings a smaller portion of the complete secondary mirror tilt cycle. Since this secondary mirror tilt induces distortion into the images distortion changed as a function of time The pipeline processing attempted to track this changing distortion, but there were intervals of time (Figures 7 and 8) when the software could not compensate for the rapidly changing distortion. This effect is small relative to the 0.5 arcsec goal for WISE astrometric precision and since W1, and to a lesser extent W2, dominate the multiband astrometric solution, this effect only manifests itself for the reddest sources in the multiband data, and in the Single-exposure W3 images.
 |
 |
| Figure 9 - Cross-scan position difference relative to 2MASS as a function of time during the 3-Band Cryo observations. For brief periods uncompensated distortion produces position offsets as large as 0.2". | Figure 10 - As in Figure 9, but comparing position difference in the in-scan direction. |
Last update: 2012 July 26