Appendix 3. Long Exposure (6X) Scan Databases, Catalogs and Images


2. General Properties of the 6x Observations, Images and Working Source Databases

a. Scope of the 6x Observations

b. Detailed 6x Field Properties

Figure 1 - Interactive sky map showing the locations of the 2MASS 6x fields (in red). The green line denotes the Galactic Plane. Click on a field for a detailed summary of its observations and properties of the extracted sources.
Abell 262 - 6x Abell 347 - 6x Abell 401 - 6x Abell 407 - 6x Abell 426 - 6x Abell 553 - 6x Abell 754 - 6x Abell 3420 - 6x Abell 3558 - 6x Cancer - 6x Chameleon II - 6x Cygnus - 6x Hydra - 6x IC348 - 6x IC1396 - 6x Large Magellanic Cloud - 6x Lockman Hole - 6x Lupus - 6x M31 - 6x Perseus - 6x Perseus - 6x Pisces - 6x Pleiades - 6x rho Ophiuchus - 6x Small Magellanic Clouds - 6x 3C 129 - 6x Test Fields - 6x Test Fields - 6x Test Fields - 6x Test Fields - 6x Test Fields - 6x

Figure 1 is an interactive sky map showing the fields observed using the 2MASS long exposures. Clicking on the red shaded boxes or field labels will take you to a web page that summarizes the observations and detailed properties of the 6x data for each field.

A brief description of the tabular and graphical contents of the field summary pages is given in the list below. A more detailed discussion of the field information and how it demonstrates the performance achieved by the 6x observations is given in the sections that follow. Click on the Table or Figure number in the list to go to the corresponding details in the text.

Detailed 6x Field Summary Page Contents
Table 16x field centers, length and number of tiles, areal coverage and boundaries
Table 26x point and extended source WDB, Catalog and metadata table information.
Figure 1Map of 6x point source WDB distributions and scan outlines.
Figure 2Map of 6x extended source WDB distributions and scan outlines.
Figure 3Raw 6x point and extended source counts from the respective WDBs.
Figure 4Area-normalized 6x point and extended source counts from the 6x-PSC and 6x-XSC. All-Sky PSC and XSC counts in same area shown for comparison.
Figure 56x Point and extended source color-magnitude diagram.
Figure 66x Point and extended source color-color diagram.
Figure 76x-survey point source photometric residuals plotted as function of 6x point source magnitude.
Figure 86x-survey extended source photometric residuals plotted as function of 6x extended source magnitude.

i. 6x Observations and Sky Coverage

Table 1 in each summary page describes the sky coverage and the number of 6x scans made in each field. The tables give the scan length, the number of pre-defined tiles covering each field, and the number of scans made on the field. Because certain tiles in some fields were scanned more than once, the number of scans can exceed the number of tiles.

The first two figures in each summary page show the distributions on the sky of the 6x point and extended source Working Database (WDB) extractions for each field. The 6x WDBs contain all extractions from all 6x scans, so the source distribution sky maps may show apparent overdensities in tiles observed more than once and in tile overlap regions (e.g. Abell 3558). Because the WDBs also also contain flagged but spurious detections of image artifacts, patterns that trace out stellar diffraction spikes, latent images, etc. are seen in the vicinity of bright stars in the point source distribution maps, such as those seen in the map of the Abell 407 6x PSWDB in Figure 2 below. The 6x Catalog Generation process removes duplicate source detections and filters out spurious detections of low SNR noise and image artifacts, so the distribution of 6x-PSC sources is much more uniform. The 6x-PSC source distributions for the Abell 407 field are compared to the 6x-PSWDB extractions in Figure 3 below.

The scan and source distribution sky maps also illustrate coverage gaps at tile declination boundaries in some fields resulting from a telescope commanding software error, and missing tile coverages due to poor observing condition. Both effects can be seen in the 6x PSWDB sky distribution in the Pleiades field shown in Figure 4.

Figure 2 - Sky map showing outlines of Abell 407 6x scans in blue and distribution of 6x-PSWDB extractions (red dots). Note the spurious detections of diffraction spike around several bright stars. Figure 3 - Click this figure to view a blink comparison of the 6x-PSC and 6x-PSWDB source distributions in the Abell 407 6x field. 6x Catalog selection removes spurious detections of low SNR noise and artifacts detections around bright stars and redundant detections of sources in scan overlaps. Figure 4 - Sky map of Pleiades 6x scans in blue and 6x point source WDB extractions illustrating gaps in coverage due to a missing tile and scan declination declination offsets that affected some 1°-long 6x scans.

ii. Achieved Sensitivity and Source Counts

Figure 5 contains histograms of the SNR=10 brightness level for all 2MASS 6x scans, estimated statistically from the measured atmospheric transparency, seeing and background conditions using the method described in VI.2. Compare these with the distributions from the main survey shown in Figure 18 of VI.2 to see that the 6x measurements have predicted sensitivities approximately one magnitude fainter than the survey as expected from the increased exposure time.

Figure 5 - Distribution of the predicted SNR=10 magnitudes for all photometric 6x scans, based on atmospheric transparency, seeing and background. These estimates do not take into account the effects of confusion.

The achieved sensitivity of the 6x measurements is illustrated by the third and fourth figures in the field summary pages that show the raw and area-normalized 6x differential point and extended source counts as a function of magnitude, respectively. These figures were constructed using the default magnitudes for point sources and the Ks=20 mag arcsec-2 elliptical isophotal magnitudes for extended sources. Source densities from the same areas of the All-Sky PSC and XSC are shown in the area-normalized count plots for each field to illustrate the sensitivity improvement.

Source confusion moderates the sensitivity gain relative to the main survey in 6x fields that cover regions of high source density. This is illustrated in Figures 6 and 7 below which show the area-normalized source counts for the Lockman Hole and Cygnus fields, respectively. 6x source counts in the high galactic latitude Lockman Hole field extend approximately one magnitude fainter than those in the main survey Catalogs. Counts from the 6x observations of the Cygnus field, which has an order of magnitude larger source surface density than the Lockman Hole region, extend less than one magnitude fainter then the main survey.

Figures 6 and 7 also demonstrate that the surface density of extended sources (primarily galaxies) is systematically lower in high source density fields because of incompleteness due to confusion with foreground stars and because of obscuration by foreground dust in the Milky Way. Because a number of the small 6x fields target low redshift galaxy clusters that contain a relatively high concentration of bright galaxies, their extended source count curves can be flatter than seen in the field (e.g. compare the Abell 426 counts with those of the Lockman Hole).

The 6x point source counts show a pronounced discontinuity near J~11, H~10.5 and Ks~10.5 mag that is caused by the reduced overlap in READ1/READ2-READ1 sensitivity in the 6x observations. This discontinuity, illustrated in the raw point source counts from the Lockman Hole in Figure 8, is produced from the "pile-up" of detections at the faint (low SNR) end of the READ1 regime due to statistical flux overestimation, and incomplete detection of bright sources because of saturation near the high SNR end of the READ2-READ1 regime.

Figure 6 - Area-normalized 6x-PSC and 6x-XSC differential source counts in the Lockman Hole field, compared with those in the same area in the All-Sky PSC and XSC. 6x-PSC and 6x-XSC counts are shown by the red and blue lines, respectively. All-Sky PSC and XSC counts are shown in the grey and cyan shaded regions. Figure 7 - Area-normalized 6x-PSC and 6x-XSC differential source counts in the Cygnus field, compared with those in the same area in the All-Sky PSC and XSC. 6x-PSC and 6x-XSC counts are shown by the red and blue lines, respectively. All-Sky PSC and XSC counts are shown in the grey and cyan shaded regions. Figure 8 - Differential point source WDB counts from the 6x Lockman Hole observations showing the discontinuities at the READ1/READ2-READ1 boundaries.

iii. Photometric Performance

Figures 9, 10 and 11 below show the point source J, H and Ks default photometric measurement uncertainties ([jhk]_msig) from the 6x Lockman Hole observations. Compare these with the distributions from the All-Sky PSC shown in Figures 2-7 in VI.3.a.iii. Uncertainties rise at faint magnitudes due to the dominance of background noise. The achieved SNR=10 and 20 levels are shifted approximately one magnitude fainter than in the main survey as expected because SNR scales with the square root of the exposure time for background-limited measurements. For brighter rd_flg='2' sources, the uncertainties are consistently in the range 0.02-0.03 mag, as in the main survey. This is the limiting profile-fit uncertainty attributable to undersampling, due to the coarse pixel size of 2MASS (2.0´´) relative to the seeing-limited image size (typically 1.0´´). Because the saturation level scales linearly with exposure time, the 7.8s 6x READ2-READ1 exposures saturate two magnitudes fainter than in the main survey. Therefore, the default 6x photometry for stars in the 8-11 mag range is supplied by relatively low precision aperture aperture photometry on the READ1 images (rd_flg='1') instead of high precision READ2-READ1 profile fitting as in the main survey. Better photometric accuracy for stars in this brightness range can usually be obtained by using All-Sky PSC measurements. The brightness at which the 51 ms READ1 exposures saturate is the same in the 6x measurements as in the main survey (i.e. the boundary between rd_flg='1' and rd_flg='3'), so the photometry of fully-saturated (rd_flg='3') stars is equally uncertain in both.

Figure 9 - J-band Figure 10 - H band Figure 11 - Ks-band
Point source default photometry uncertainties from the 2MASS 6x Lockman Hole measurements. Grey dots denote individual unsaturated (rd_flg='2') sources, light blue dots are bright (rd_flg='1') stars that saturate the 7.8s READ2-READ1 exposures but not the 51ms READ1 exposures. Extremely bright stars (rd_flg='3') that saturate even the 51 ms READ1 exposures are shown as dark blue points. The filled blue circles with error bars show the trimmed average and RMS photometric uncertainties for all sources in 0.2 mag bins.

The relative, band-to-band photometric performance of the 6x measurements is illustrated by the color-magnitude and color-color diagrams for the point and extended sources shown in the fifth and sixth figures in each 6x summary page. Figures 12-14 below show sample color-magnitude diagrams for the Lockman Hole, 3C129 and Ophiuchus 6x fields, and serve to illustrate the variety of source populations and environments sampled by the 6x observations. The Lockman Hole field lies at high galactic latitude where the detected point sources are predominantly late-type dwarf stars (J-Ks~0.9, with a relatively small fraction of bluer, earlier dwarfs. There is also a prominent population of faint, red point sources extending below the distribution of resolved galaxies. These are unresolved galaxies, and they become an increasingly larger fraction of the near-infrared sources detected at fainter magnitudes probed by the 6x measurements. In 3C129, a field in the Galactic Plane but relatively free from heavy foreground extinction, many more earlier-type dwarfs are detected than in the Lockman Hole, as well as more distant giants. Note that there are relatively few galaxies detected both because of the small area of the field and because of incompleteness due to foreground stellar contamination. The Ophiuchus field samples a heavily extinguished line of sight through a molecular cloud complex. The distribution of stars in its color-magnitude diagram are shifted and stretched to redder colors due to extinction in the cloud, and there are few if any detections of galaxies. The reddening in the Ophiuchus field is also evidenced by its color-color diagram which has points stretched out along the direction of the reddening vector.

The effect of the READ1/READ2-READ1 sensitivity gap can be seen graphically in the color-magnitude diagrams of some of the 6x fields, such as the Lockman Hole shown in Figure 12. The horizontal discontinuity in the point source distribution is produced by an incompleteness due to saturation at the bright end of the READ2-READ1 regime, and a systematic overestimation of the brightness of low SNR READ1-detected sources (rd_flg='1').

Figure 12 - Color-magnitude diagram for point and extended source extractions 6x WDBs in the high galactic latitude Lockman Hole field. Point sources are shown as small red dots and extended sources as larger green dots. The black contours trace the density of point sources. Figure 13 - Color-magnitude diagram for point and extended source extractions 6x WDBs in the low galactic latitude 3C129 field. Point sources are shown as small red dots and extended sources as larger green dots. The black contours trace the density of point sources. Figure 14 - Color-magnitude diagram for point and extended source extractions 6x WDBs in the heavily extinguished Ophiuchus field. Point sources are shown as small red dots and extended sources as larger green dots. The black contours trace the density of point sources.

iv. Photometric Comparison to All-Sky PSC and XSC

The relative photometric calibrations of 6x and survey point and extended source photometry are compared in Figures 7 and 8 in each of the 6x field summary pages. These figures show the distribution of 6x-versus-survey photometric residuals as a function of 6x magnitude for point and extended sources, respectively. To construct these diagrams, the closest All-Sky PSC/XSC sources within 2" (point sources) or 5" (extended sources) corresponding to each 6x-PSC and 6x-XSC entry were identified. Simple differences between the point source default magnitudes ([jhk]_m) and extended source 7" circular aperture magnitudes ([jhk]_m_7) were formed band-by-band if there were detections in both the 6x and survey source entries. The grey dots in the figures indicate individual sources and the contours trace the density of the individual source residuals. The filled blue circles with error bars indicate the mean and RMS residuals for all sources in 0.25 mag bins.

Representative examples of the point and extended source photometric residuals are shown in Figures 15 and 16 below. These figures, drawn from the Pisces 6x field data, demonstrate that there is <1-2% bias between the 6x and survey photometry in the brightness intervals corresponding to high SNR in the 51 ms READ1 and 7.8 s READ2-READ1 exposures. These diagrams also illustrate systematic biases between the 6x and survey photometry in certain brightness regimes. These biases are described in A3.1.d.

The point source residuals exhibit a downward trend in the 10-11 mag range, where the default 6x photometry comes from low SNR measurements on the 51 ms READ1 exposures (rd_flg="1"). These low SNR measurements systematically overestimate source fluxes relative to the All-Sky PSC photometry which comes from high SNR measurements on the READ2 exposures (rd_flg="2"). The mean 6x/survey photometric residuals become increasingly positive towards the faint end of the point and extended source distributions as a result of statistical flux overestimation of the low SNR sources measurements from the survey photometry.

In the southern 2MASS 6x fields, 6x point source photometry appears systematically fainter than photometry from the All-Sky PSC near ~11 mag. This bias is probably due to non-linearity in the 2MASS detectors because the default 6x measurements come from bright sources that are near the saturation limit of the 7.8 s 6x READ2 exposures (rd_flg="2"). Sources at this brightness are well removed from the saturation limit of the shorter 1.3 s survey READ2 exposures. As can be seen in the photometric comparison plot for the Abell 3558 6x field, the bias is most prominent in the southern J-band data, present but less prominent in the southern H-band data, and negligible in the southern Ks-band data. This bias is not seen in the northern 6x point source photometry.

In certain northern and southern high source density 6x fields, such as Cygnus, the 6x point source photometry for sources in the range 12-15 mag can be a few percent brighter on average than that from the All-Sky PSC. The origin of this offset is not understood and it is not seen in all high density fields, nor in all bands.

Because of potential biases between 6x and survey photometry in the transition region between 51 ms READ1 and 7.8s READ2 measurements from the 6x exposures, high SNR measurements from the 2MASS All-Sky Catalogs should always take precedence over 6x measurements of the same objects. The 2MASS All-Sky Catalogs have also benefited from greater validation and scrutiny.

Figure 15 - Differences between 6x and main survey default point source photometry as a function of 6x source default magnitude in the Pisces 6x field, observed with the northern 2MASS facility. Black contours trace the density of individual sources that are shown as light grey points. The large blue points and error bars show the trimmed average and RMS of the 6x-survey magnitude differences for all sources in 0.25 magnitude wide bins. Figure 16 - Differences between 6x and main survey extended source photometry in 7" circular apertures ([jhk]_m_7) as a function of 6x extended source magnitude in the Pisces 6x field. Black contours trace the density of individual sources that are shown as light grey points. The large blue points and error bars show the trimmed average and RMS of the 6x-survey magnitude differences for all sources in 0.25 magnitude wide bins.

c. Analysis of 6x Position Reconstruction

[Last Updated: 2008 February 14; by R. Cutri]


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