Appendix 1. Introduction to the 2MASS Extended Mission and Ancillary Data Products

4. Cautionary Notes

a. General Cautionary Notes for the 2MASS Extended Mission Ancillary Data Products

  1. Additional Information Available
  2. Using the 2MASS Extended Mission Products
  3. Sky Coverage Non-Uniformity, Variable Depth-of-Coverage and Limitations
  4. Sensitivity Variations
  5. Low Quality Scans
  6. The 6x READ1/READ2-READ1 Sensitivity Gap
  7. Astrometric Biases in the Calibration Scan Data Products

i. Please see General Cautionary Notes for the 2MASS All-Sky Release (I.6.a)

More thorough descriptions of the general cautionary notes for the Survey Reject and 6x and Calibration Working Database Tables and Images are available at:

ii. Using the Extended Mission Databases and Image Atlases

Because all 2MASS data were obtained with well-characterized facilities and reduced with the same processing software that incorporates multiple levels of quality assurance, the overall characteristics of the 2MASS Survey Reject Tables, 6x and Calibration Working Databases and their corresponding Image Atlases are expected to be comparable to the those from the All-Sky Release. However, the Extended Mission Products contain, in part, extracted source data and images that do not meet the strict standards of reliability and uniformity required for the primary 2MASS products. In addition, the Extended Mission Databases and Images have not been as carefully analyzed the All-Sky Release Catalogs prior to their release, so all areas of performance have not been characterized and validated. Users should therefore use these products with care and follow these recommendations:

iii. Sky Coverage Non-Uniformity and Limitations

The effective sky coverage of the 2MASS Extended Mission Image Atlases and Source Databases are non-uniform and complex. Consequently, these products are not suitable for extracting statistically complete samples of objects sources.

Full Survey Atlas and Point and Extended Source Reject Tables

The Full Survey Image Atlas and the Survey Point and Extended Source Reject Tables (PSRT and XSRT) are drawn from all 70,712 6°-long survey mode scans that covered 99.998% of the sky made in photometric conditions. However, the content and depth-of-coverage of the PSRT and XSRT in specific scans depend on whether that scan was used in the 2MASS All-Sky Release, and how many times that position on the sky was observed.

Approximately 30% of the sky was observed more than once during 2MASS survey operations (see A2.1.a.i). Half of that area is contained in the overlap region between survey tiles, and half is contained in the 8,266 tiles that were repeatedly scanned. The Full Survey Image Atlas contains two or more, independent images of all regions that were observed multiple times. There may be up to ~500 images covering locations near the equatorial poles where survey tiles converged. One measurement of sources that were observed and detected multiple times in the survey scans are usually found in the All-Sky Release PSC or XSC. The remaining detections are contained in the PSRT or XSRT. Objects observed and detected three or more times may have multiple, independent measurements in the PSRT or XSRT.

6x Image Atlas and Point and Extended Source Working Databases/Catalogs

2MASS 6x observations covered a much more limited area than the main survey. Approximately 589 deg2 of sky was observed with the longer exposure times, distributed among 30 discrete fields containing targets of astronomical interest (see Figure 1 in A3.2). Most of this area is contained in two large regions covering the Large and Small Magellanic Clouds, in which 383 deg2 and 127 deg2 are surveyed, respectively. The remaining ~80 deg2 is in 28 smaller fields, the largest covering 28.3 deg2 in the Lockman Hole region.

The 6x field observations were conducted using either 6°-long or 1°-long scans, depending on the size of the region to be covered. As illustrated in Figure 1, the coverage within each field was often complex and usually not defined by a simple rectangle on the sky. There are often coverage gaps within the fields that were resulted from tiles being missed due to poor weather and because of telescope commanding errors. The footprints of the scans covering each field are presented in the detailed 6x field description pages in A3.2.b.

Figure 1 - Sky map showing the outlines of 6x scans covering the Pleiades field (blue lines), and the distribution of 6x-PSWDB extractions (red dots). Note the coverage gaps at the declination boundaries between some tiles, and the missing tile coverage near the eastern edge of the field.

Calibration Scan Image Atlas and Point and Extended Source Working Databases

2MASS calibration observations covered the smallest area of all the 2MASS data sets - approximately 6 deg2 of sky distributed among the 40 fields shown in Figure 1 in A4.1. What the calibration observations lack in area, though, they make up for in depth-of-coverage and variety in sampled astrophysical environments.

Each time a calibration field was observed, a 1°-long area was scanned six consecutive times in alternating north-south directions. The 2MASS telescopes were stepped 5" east in RA between each scan within a calibration observation, so the total area covered is usually ~0.15 deg2. Over the course of the 2MASS survey operations, the calibration fields were observed between several hundred and several thousand times.

In most fields, the footprints of the scans were relatively well-registered on the sky (Figure 1). However, a small fraction (<5%) of the scans in a few of the fields were displaced from the nominal position due to occasional telescope pointing variations. This results in irregular coverage patterns such as shown in Figure 2. The depth-of-coverage in the calibration fields ranges from one scan at the extreme edges, up to several hundred or thousand along the central parts of the field.

Figure 2 - 90182 Field CoverageFigure 3 - 92202 Field Coverage
Coverage maps showing footprints of 1703 scans of the 90182 calibration field (left) and 2802 scans of the 92202 field (right). The maps show a 1x2 area in cartesian projection with the RA scale exaggerated to emphasize the scan cross-stepping. The scans of 90182 were all well-registered, while ~1% of the scans of 92202 were off the nominal field position.

iv. Sensitivity Variations

Within each of the 2MASS survey, calibration and 6x data sets, the relative photometric sensitivity achieved varied by up to one magnitude because of variations in atmospheric transparency, seeing and background emission levels at the time of each observation.

Figures 4, 5 and 6 show distributions of SNR=10 photometric sensitivities levels all 2MASS survey, calibration and 6x scans, respectively, estimated using the empirical relationship between observing condition and photometric repeatability described in VI.2. The average point source sensitivity level achieved for survey and calibration observations is J=16.3, H=15.4 and Ks=14.7 mag, and the 6x sensitivities are approximately one magnitude fainter because of the longer exposure times. Within each data set, the achieved sensitivities span a range of up to one magnitude because of the differences in atmospheric observing conditions.

The estimated SNR=10 magnitude level for each 2MASS survey, calibration and 6x scan is given by the [jhk]_msnr10 values in the respective Scan Information Tables.

A number of the survey and calibration scans have estimated sensitivities below the thresholds specified in the 2MASS Level 1 Requirements. Data from survey scans with sensitivities below the required levels were generally not selected for inclusion in the All-Sky Release, but are still contained in the survey Reject Tables and Full Survey Atlas. The sensitivity of survey scans not used in the All-Sky Release is shown in Figure 4 by the dashed lines. Calibration scans were not required to meet the survey's sensitivity thresholds, so there are proportionally more low sensitivity scans in the calibration data set.

Expect to find elevated photometric uncertainties for fainter sources and brighter completeness limits than in the All-Sky or 6x Catalogs in scans of tiles with low sensitivities. Because most scans with poor sensitivity were reobserved under better conditions, it is often possible to find data from alternative scans of the same tile with better sensitivity among the various 2MASS data sets.

Figure 4 - Survey ScansFigure 5 - Calibration ScansFigure 6 - 6x Scans
Distribution of estimated SNR=10 point source sensitivity levels achieved in all 2MASS survey (left), calibration (center) and 6x scans (right), in the absence of confusion. For the survey scans, the dashed lines denote the subset that were not selected for the All-Sky Release Catalogs. Note that some scans in the survey and calibration data sets fall below the sensitivity requirements for the main survey (J=15.8, H=15.1, Ks=14.3 mag)

v. Low Quality Scans

The 2MASS Extended Mission Image Atlases and Source Tables contain data from survey, 6x and calibration scans that may not meet all of the survey's quality requirements, or that exhibit other anomalies.

Survey and 6x Image Atlas and Source Tables

The calibrated images and source lists extracted from each 2MASS survey and 6x scan were subjected to a rigorous and uniform Quality Assurance process during survey data reduction. This process resulted in the assignment of a numerical quality score for each scan indicating the likelihood that the data satisfy the survey's Level 1 Requirements (adjusted by +1 mag for the 6x data products). Scans of tiles that have a 100% chance of meeting these requirements were assigned quality=10, and those with the lowest probability of meeting the requirements but were still taken under photometric conditions were assigned quality="1". Quality="0" was reserved for scans that were known to fail the requirements, usually because of non-photometric weather. Only data from survey scans with quality>0 were loaded into the Survey and 6x Working Databases and Image Atlases. The quality score for the survey and 6x scans is given in the qual column in the Survey and 6x Scan Information Tables.

During the course of the 2MASS survey and 6x observations, tiles with scans having poor quality scores were scheduled for reobservation, and most of these were successfully observed under improved conditions. The scan of each tile with the best relative quality was selected for the All-Sky Release and 6x Catalogs and Image Atlas. However, the Full Survey and 6x Image Atlases and Survey Source Reject Tables and 6x WDBs contain the calibrated images and extracted source data from all q>0 survey and 6x scans, including those with quality that may fall below various thresholds and requirements imposed on the All-Sky Release and 6x Catalogs.

The Survey Scan Information Table contains several flags that identify observations that were found to exhibit anomalies or peculiarities during quality review. These flags and their corresponding anomalies are summarized in Table 1. Scans that are affected by one or more of these anomalies will have a relevant flag value of "1". Scans with flag values of "0" are not affected. Use caution when utilizing source and image data from any scan in which the quality anomaly flags listed in Table 1 have a value of "1". There may be data from alternative scans of the same regions of sky available in the All-Sky Catalogs or Survey Reject Tables. No anomalies are identified among the 6x scans.

Table 1 - Quality Anomaly Flags in the Survey Scan Information Table
cldClouds may affect data from a small portion of scan
xphNon-cloud related photometric anomaly in scan
anomUnusual problem found in Atlas Images such as residual meteor trails, satellite trails, insect trails, and unusual reflections or glints from bright stars. Examples of these are shown in the gallery of Image Anomalies.
hglElevated "noise-glitch" may affect one or more J or H band images in scan

Calibration Scan Image Atlas and WDBs

Because the principal objective of the 2MASS calibration observations was to derive photometric zero point and atmospheric stability information during each night of survey operations, they were not subject to the same rigorous quality assessment as the main survey scans. In particular, calibration scan data were not required to meet the sensitivity thresholds required for the main survey, or to be free from astrometric errors as long as standard stars could be identified and their photometry extracted. Calibration scans were assigned a binary quality score (10 or 0 = pass or fail) that was based only on their photometric quality. A passing score of 10 was assigned to scans in which standard stars could be unambiguously identified, photometric zero points could be derived, and that were part of a six-scan calibration observation that was made under photometric conditions. Data from scans receiving a passing score were loaded into the calibration scan Atlas Image archive and WDBs. However, these products may contain other anomalies that were not identified during the quality assurance process.

vi. The 6x READ1/READ2-READ1 Sensitivity Gap

Photometry for bright sources that saturate the 6x 7.8 s READ2 exposures (rd_flg="1") should be obtained from the 2MASS All-Sky PSC. In addition, photometry for any point source with high SNR measurements in the All-Sky Catalogs should take precedence over the 6x measurements because they have received a greater level of review and validation

For the 6x observations, the READ2 integration time was increased from 1.3 s to 7.8 s, but the READ1 integration time remained at 51 ms, the same as in the main survey (see A3.4.a). The brightest unsaturated sources in the main survey's 1.3 s READ2-READ1 images were also detectable as the faintest sources in the 51 ms READ1 frames. This photometric overlap was essential for providing contiguous and internally consistent magnitude coverage across the dynamic range of the survey. The 6x observations do not necessarily have overlapping READ1 and READ2-READ1 photometry because of the increased READ2 integration time. Thus, many saturated sources in the 6x READ2-READ1 exposures are too faint to be detected in the 51 ms READ1 frames, or are detected only at very low SNR levels.

The resulting incompleteness and/or faint detections in the 8-11 mag READ1/READ2-READ1 boundary introduces artificial features in source counts and color-magnitude diagrams as illustrated in Figures 7 and 8. Small discontinuities in the differential 6x Lockman Hole field point source counts (Figure 7) near 10.5-11 mag are caused by the incompleteness and systematic flux overestimation of low SNR measurements in the READ1 exposures. The same effects cause the horizontal discontinuity of points near Ks=10.5 mag in the 6x color-magnitude diagram from the same field shown in Figure 8.

The 2MASS 6x data primarily exist to report the faint extension of the original survey results provided by the 6x increase in READ2 exposure time. Photometry of the brighter 51 ms READ1 extractions are included in the 6x Catalogs given their utility for variability and proper motion studies.

Figure 7 - Differential point source counts from the 6x Lockman Hole observations showing the discontinuities at the READ1/READ2-READ1 boundaries. Figure 8 - Color-magnitude diagram for point (red) and extended source (green) extractions 6x WDBs in the Lockman Hole field. The black contours trace the density of point sources. The horizontal discontinuity near Ks=10.5 mag is at the READ1/READ2-READ1 sensitivity boundary.

vii. Astrometric Biases in the Calibration Scan Data Products

2MASS calibration scan images and source positions have systematic biases of up to ~0.6" with respect to the 2MASS All-Sky PSC and XSC.

Position reconstruction for the calibration scan observations (A4.4.b) used the USNO-A2.0 as the primary astrometric reference catalog, rather than the Tycho-2 Catalog that was used as the reference for the main survey. Because of small systematic astrometric differences between USNO-A2.0 and Tycho-2, the reconstructed astrometry of 2MASS Calibration scan data products exhibit residual biases of up to ~0.6" with respect to the 2MASS main survey images and source positions.

These astrometric biases are position-dependent within calibration scans, and they differ in amplitude and structure between calibration fields. However, they are consistent among all scans of a given calibration field. The average offsets with respect to the main survey have been computed for each calibration field using the Merged Calibration Point Source Database and are shown in A4.2. The largest biases are found in the 90004 calibration field, which has an offset relative to the All-Sky PSC of approximately 0.6" in declination as shown in Figures 9 and 10.

Source positions in the All-Sky PSC and XSC and Survey PSRT and XSRT are more accurate relative to the J2000 reference frame than those in the calibration WDBs, and should be be used wherever possible. However, the internal consistency of the position reconstruction within scans of individual calibration fields is excellent. This consistency, combined with the large number of independent scans of the calibration fields, make the Cal-PSWDB a resource for measuring source very small-scale motions.

Figure 9 - Mean RA offset between Calibration Scan and All-Sky PSC positions for stars in the 90004 Calibration Tile, plotted as a function of RA and Declination within the Tile. Figure 10 - Mean Declination offset between Calibration Scan and All-Sky PSC measurements for stars in the 90004 Calibration Tile, plotted as a function of RA and Declination within the Tile.

[Last Updated: 2006 December 20; by R. Cutri]

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