I. Introduction

6. Cautionary Notes

a. General

i. Preliminary Nature of the Incremental Release Products

The 2MASS Incremental Data Releases are the result of the first processing of data from the Survey performed while Survey observations are ongoing. Rigorous quality assurance procedures used during Survey data processing indicate that the vast bulk of the released 2MASS data products meet or exceed the high scientific standards of the Survey. However, validation of a dataset as large as that being produced by 2MASS presents a great challenge, and it is expected that there will be problems both known and unforeseen that persist into the release data products. These data do not yet benefit from all experiences that will be gained over the full Survey, in particular the analyses of the full-sky dataset, nor have they undergone all the rigorous analyses that accompany data releases at the end of mission. However, the benefits of releasing data to the community now exceed potential risks, and the feedback from the community on the data products and documentation will ultimately contribute to a better final product. The knowledge gained as the survey continues, and from the feedback received from users will be incorporated when the entire 2MASS dataset is reprocessed at the completion of the Survey observations.

NB: Users are strongly recommended to review the caveats listed below.

ii. Sky Coverage and the Survey Tiles

Figure 1, a composite map of the J, H and Ks surface density of objects in the 2MASS Second Incremental Release Point Source Catalog, illustrates that there are many gaps in the sky coverage for this release. The uneven sky coverage results from the spatial and scheduling strategy used to carry out the Survey. The absence of an object of interest from the Incremental Release Catalogs does not necessarily mean that the object was not detected in the Survey, but rather it may be in a region not included in this Release.

The 2MASS telescopes scan the sky in units of Survey Tiles which are one camera frame (8.5´) wide in the Right Ascension direction and 6° long in declination. On average, each telescope scans approximately 60 tiles each clear night. Because the RA coverage rate is slower than the sidereal rate, several contiguous, but not necessarily adjacent, blocks of Tiles are observed each night. Therefore, adjacent Tiles may be observed at very different times during the Survey. In addition, non-photometric or other sky conditions can force the rejection of the observations of some of each night's Tiles. Those Tiles are reobserved at a later date which further adds to the time discontinuity in sky coverage.

The regions included in the Incremental Releases can be examined using the Survey Visualizer, or queried with the Coordinate Search Tool. The Scan Information Table that is accessible via IRSA, ftp, or DVD-ROM, provides the observation date, reconstructed position (J2000 RA and DEC of the four scan corners) and various observational and calibration characteristics for each Tile contained in the Release.

The effective sky coverage of 2MASS is also diminished because areas around bright stars are masked to remove the large number of spurious detections that are trigerred by diffraction spikes, dichroic glints and latent images. In addition, several larger regions have been masked out the Extended Source Catalog. These areas include the Galactic Center, where the source densities are so high that virtually all XSC sources are multiple stars and not truly extended, and segments of a few tiles in which the atmospheric seeing was not tracked properly, leading to inaccurate extended classification for sources.
Figure 1

iii. Catalog Reliability

In unconfused regions, the target reliability for sources in the 2MASS Catalogs is 99.95% (PSC) and 99% (|b|>20o and 80% (10o < |b| < 20o) for the XSC. Analyses of the Incremental Release Catalogs indicate that a relatively small fraction of sources that technically meet the Catalog source selection criteria are known to have photometry and/or classification corrupted by nearby sources, or are outright spurious detections of image artifacts. Extensive efforts have been made to identify and flag such objects that remain in the Catalogs. The PSC and XSC source records include a contamination and confusion flag (cc_flg) that identifies sources that should be considered with extreme caution.

User's who wish to draw samples of sources from the PSC or XSC that maximize reliability and photometric accuracy should select sources that have cc_flg="000" in the case of the PSC, and cc_flg="0" for the XSC.

iv. Photometric System

Magnitudes reported in the Point and Extended Source Catalogs are in the natural 2MASS photometric systems. Sections III.2.c and IV.8 describe the Survey calibration procedure. Transformation equations to other photometric systems are not yet available. However, 2MASS colors for normal stars lie very close to those reported by Koorneef (1983, As.Ap. 129, 84) and Bessell and Brett (1988, PASP, 100, 1134).

v. Global Photometric Accuracy

The global photometric uniformity of the 2MASS Catalogs is enforced by nightly photometric calibration to an extensive set of standard star fields. However, it will be difficult to assess the overall photometric uniformity of the survey until most of the sky has been analyzed. Analysis of the internal consistency of large numbers of stars in multiply-observed calibration fields indicates that the systemmatic variations of the 2MASS flux calibration frame of reference are of order 1% around the sky.

vi. Tile Overlap Regions

Duplicate Source Rectification

Successive 2MASS Survey Tiles overlap each other by approximately one arcminute in the Right Ascension direction and 8.5´ in Declination. Therefore, approximately 15-20% of sources in the Survey can be detected multiple times.

To avoid artificial sensitivity biases in the Tile overlap regions, the Point and Extended Source Catalogs contain data from only a single measurement for multiply detected sources, and not averaged data. The measurement that is included in the Catalogs is from the apparition that falls farthest from the edge of its respective Survey Tile (or equivalently closest to its Tile center). Very faint sources in Tile overlap regions may not be detected in all scans of those regions. Such a faint detection is selected for inclusion in the Catalog only if it falls farther from the edge of its Tile than its virtual counterparts in overlapping Tiles fall from the edges of their Tiles. Section V.3 provides a more detailed explanation of the duplicate source rectification procedure.

The duplicate source selection procedure operates only on the edges of Tiles that have an overlapping Tile that is also in the Second Incremental Data Release. For Tiles in the Release with "unbounded" edges, all qualifying sources out to within 10" (PSC) or 15" (XSC) of the unbounded edges are included in this Incremental Release. In subsequent Incremental Release Catalogs, these apparitions may be replaced by the appropriate source in an available overlapping Tile, or in the case of very faint sources, the source may not appear at all.

Consequences for Atlas Images

The 2MASS Atlas Images are derived directly from the 2MASS tile observations. The sky region in the westernmost 10% of an Atlas Image may appear on the eastern edge of the adjacent Image. The 2MASS Survey Visualizer is designed to return the Atlas Image for which the input coordinate is closest to the tile center. If the input coordinates for an object of interest are not accurate (as sometimes occurs for name-resolved positions), the target may lie on extreme edge of the Atlas Image. In these cases, it is possible that a query aimed at returning the adjacent Atlas Image may yield a more centered image of the source. During the Incremental release period, however, the adjacent Atlas Images may not be available.

vii. Artifacts

Meteors and Artificial Satellites

Meteors and artificial satellites produce trails that frequently appear on 2MASS raw data frames. Since the camera observes each position on the sky in six consecutive frames, the coaddition of frames mitigates the contamination from the trails. However, meteor and satellite trails persist into the Atlas Images, where they often trigger point and/or extended source detections. These "sources" are usually characterized by high 2 values and low "N/M" statistics in the case of point sources (see IV.4c). Many of these false detections have been removed from the Catalogs, but the algorithm to identify trails is not 100% efficient. Users are encouraged to review the images of interesting sources to rule out the possibility that they may be meteor or satellite trail detections.

Insects, Electronics Glitches, etc.

The rigorous quality assurance procedures applied to all 2MASS data products have largely identified rogue artifacts such as airplanes, insects walking across the camera window, and rare failures of the camera readout electronics, and identified those survey tiles for re-observation -- thus excluding them from the release data. Inspection of a fraction of the Images in the release suggests a few such artifacts may remain. Again, users are encouraged to review the images of sources of interest.

Anomaly Lists

The 2MASS Catalog products are static. No deletions will be made after a particular release, even for sources that are known to be artifacts, until the Catalog is rereleased after reprocessing. Separate anomaly lists will be provided that are updated at regular intervals.

[Last Update: 2000 January 26; R. Cutri, M.Skrutskie, T.Chester]

Previous page. Next page.
Return to Explanatory Supplement TOC Page.