III. 2MASS Details


3. Data Processing

d. Extended Source Identification and Photometry

Overview: The extended source processing in 2MAPPS (GALWORKS) identifies sources that are resolved relative to the PSF and uses various apertures to measure the flux of resolved sources. Extended source processing operates independently on each of the individual Atlas Images in a scan, and does not have available the individual frame measurements for each source. Due to the survey strategy, extended source identification is complete only for galaxies smaller than the scan overlap size of 50", since some larger galaxies will not be contained on a single 2MASS scan or Atlas Image within a scan. GALWORKS operates on larger galaxies, flags galaxies that run into a scan boundary, and saves image "postage stamps" of all galaxies, including large ones (these postage stamps are not available for the 2MASS Sampler Release, but will be for the large Spring 1999 data release).

Two very important steps must occur for proper discrimination between point and extended sources. First, the seeing must be characterized throughout the scan to accurately determine the PSF used to measure whether a source is resolved. Second, the structure of the background across each Atlas Image must be fit and subtracted from each Image.

Extended source processing occurs after all detected sources have been characterized by the point source processor through artifact identification and band-merging, and hence the point source measurements for each source are available as a "seed list" for GALWORKS. This seed list contains nearly every extended source because of the robustness of the detection step (cf III.3.c). Sources that pass an initial screening are intensively examined for extent and their fluxes are measured. Sources passing further thresholds for extent are placed in the Extended Source Database. Catalog sources are eventually selected from that Database.

A publication is in preparation to accompany the Spring 1999 release of 2MASS galaxy data. The draft version is written for the user of the Extended Source Catalog, and already contains significant detail on the algorithms used for the extended source processor. It may be consulted with the caveat that this is a work in progress. Also, internal 2MASS working documents are referenced below to give further information on various topics, but note that these documents were written for the 2MASS team, and may refer in some parts of them to problems that were later fixed. The reader must be more alert in reading those documents than in reading this mini-Explanatory Supplement.

Seeing Characterization

An accurate characterization of the PSF is essential in reliably determining whether a source is resolved. Detecting galaxies in a ground-based mission like the 2MASS is thus exquisitely sensitive to atmospheric seeing and variations in telescope focus.

Most of the time, the seeing and focus vary slowly enough so that even in low source density areas the stars detected by 2MASS can be used to determine the PSF as a function of time. Such an estimate of the seeing is done previously in the pipeline to support profile-fitting photometry for point sources, but that estimate does not have the entire scan available at once. Thus the first step in GALWORKS is to more accurately determine the variation of seeing with time in each scan.

The seeing is measured by determining a size for each source above some magnitude thresholds, and then using a robust estimator to determine the mean size as a function of time, rejecting extended sources and single pixel events from being used by the estimator.

Infrequently, especially when the seeing FWHM is large, the variation in seeing occurs too rapidly to be tracked by the number of sources available. If the seeing FWHM is underestimated, true unresolved sources will be falsely identified as extended. A diagnostic has been developed that attempts to measure when this occurs, which seems to work well most of the time. This diagnostic shows that the Sampler night appears not to be troubled by untracked seeing.

Background Removal

Background removal is crucial to determining whether there is extended flux surrounding a source. If the background-removed image contains residual background near a source, the source will incorrectly appear to be extended, degrading the reliability of the catalog. If the image has removed too much background, flux around truly extended sources may disappear, causing incompleteness in the catalog.

Most of the time, the background variation in the Atlas Images is smooth enough to be fit with a cubic polynomial, after care is taken to mask out regions affected by sources. The cubic polynomial removes most structure at scales larger than 4-5'. Thus sources comparable to this size or larger will have compromised photometry.

The background-removal algorithm appears to work quite well most of the time. However, two sources of higher-frequency noise exist that are not removed by the current algorithm: electronic noise pickup in all three bands and rapid airglow variations at H.

Low-level electronic noise pickup can occasionally survive to the Atlas Images. Normally, the electronic noise pickup in the Images is negligible. However, the phase of the noise pickup can sometimes match the frame frequency and is large enough to cause photometric problems for extended sources. In the northern 2MASS camera, one side of the array exhibits variation with maximum amplitudes of ~0.20 DN and periods of 50-75", that can cause extended source flux errors of ~15%.

The background-removal algorithm is normally extremely successful in removing airglow variation. However, infrequently the airglow varies too rapidly and a small portion of airglow emission remains in the images. This appears to be a problem only at H band because this is the band in which the OH emission is strongest.

For more information, consult the working document Data Artifacts (see caveat below).

Identification

About ~1600 galaxies brighter than K~13.5 are part of the 2MASS Sampler data set. A description of the expected completeness and reliability for the extended source catalog can be found here.

A subset of previously cataloged galaxies are automatically measured and extracted into the 2MASS database. This set of objects is selected based on the optical diameter, in this case, galaxies with a diameter greater than 1', as listed in the NASA Extragalactic Database (NED). For the larger Messier objects (and some NGC objects, for example), >5', they are typically too large to process with the 2MASS imaging data, and so are not processed or extracted into the 2MASS database. That leaves the remaining (>99%) of the sky for 2MASS to find and characterize galaxies. Extended sources are identified from point source detections. That is to say, we characterize each point source and decide if it is extended with respect to the point spread function (PSF). This is accomplished using a battery of star-galaxy discrimination parameters, including intensity-weighted moments, radial profile extent measures, asymmetry metrics and mean surface brightness flux measures. This set of operations is designed to eliminate point-like objects (re: stars) and minimize contamination from double stars (the primary reliability obstacle) and other false galaxies (e.g., artifacts from bright stars). An important step that precedes star-galaxy separation is careful removal of the image background, particularly at H-band which is severely affected by atmospheric "airglow" emission. Once a source has been deemed "extended" or a candidate thereof, its flux is measured using a disparate set of apertures, ranging from fixed circular to adaptive elliptical/circular apertures. The extended source information is extracted to a table and a small "postage-stamp" image (typically 30"×30" in size) is cut out from the J,H and Ks Atlas Images. Additional star-galaxy separation is performed as a post-processing step to further refine the reliability and aid in generation of the extended source catalog. The final catalog is expected to meet or exceed the Level-1 Specifications, that include >90% completeness and 99% reliability for most of the sky (free of stellar confusion). The point source sensitivity limits (10-sigma) are 15.8 (0.8 mJy), 15.1 (1.0 mJy), and 14.3 (1.3 mJy) mag at J,H, Ks, respectively. The extended source sensitivity limits (10-sigma) are ~1 mag fainter than the point source limits, or 14.7 (2.1 mJy), 13.9 (3.0 mJy), and 13.1 (4.0 mJy) mag at J, H, and Ks, respectively.

The extended source catalog contains over 340 fields of information per source, most of which are related to photometry. Below we describe the different measures of galaxy brightness, followed by a brief description of each parameter in the extended source catalog.

Photometry

Given the diverse shape, size and surface brightness that galaxies exhibit in the near-infrared, a corresponding diverse array of apertures are used to compute the integrated fluxes. The simplest, and therefore most robust, measures come from fixed circular apertures. A set of fixed circular aperture include the following radii: 5, 7, 10, 15, 20, 25, 30, 40, 50, 60, and 70". We report both the integrated flux within the aperture (with fractional pixel boundaries) and the estimated uncertainty in the integrated flux. The magnitude uncertainty is primarily based up the measured noise in the Atlas image, which includes both the read-noise component and background Poisson component, as well as the confusion noise component (only relevant when the source density is high). The detailed formula is given here. Further information with regard to photometry and expected measurement uncertainty are given below (see URL links below). A contamination or confusion flag is also attached to each flux measurement with the following code:

For most galaxies in the 2MASS catalog, small fixed circular apertures give adequate 'total' flux measurements. In particular, we recommend use of the R=7" aperture for galaxies fainter than Ks ~ 13 mag (see 2MASS Galaxy Catalog: First Results), corresponding to field names:
j_m_7J 7" radius circular aperture magnitude
h_m_7H 7" radius circular aperture magnitude
k_m_7Ks 7" radius circular aperture magnitude
j_msig_7J 1-sigma uncertainty in 7" circular ap. mag
h_msig_7H 1-sigma uncertainty in 7" circular ap. mag
k_msig_7Ks 1-sigma uncertainty in 7" circular ap. mag
j_flg_7J confusion flag for 7" circular ap. mag
h_flg_7H confusion flag for 7" circular ap. mag
k_flg_7Ks confusion flag for 7" circular ap. mag
Adaptive aperture photometry includes isophotal and Kron metrics. The isophotal measurements are set at the 20 mag per arcsec2 isophote at Ks and the 21 mag per arcsec2 at J, using both circular and elliptically shape-fit apertures. For purposes of computing colors, two classes of photometry are carried out: individual and fiducial. The latter refers to adaptive apertures per band, while the former refers to using fixed apertures (both size and elliptical shape) based on either the J or Ks isophotes, respectively referred to as the "J fiducial" and "Ks fiducial" photometry. For the brighter galaxies in the catalog, Ks > 13 mag, we recommend the use of the "Ks" fiducial isophotal elliptical aperture photometry:
r_k20fe20 mag/sq." isophotal K fiducial elliptical aperture semi-major axis (arcsec)
j_m_k20feJ 20 mag/sq." isophotal fiducial ell. ap. magnitude
h_m_k20feH 20 mag/sq." isophotal fiducial ell. ap. magnitude
k_m_k20feKs 20 mag/sq." isophotal fiducial ell. ap. magnitude
j_msig_k20feJ 1-sigma uncertainty in 20 mag/sq." iso.fid.ell.mag
h_msig_k20feH 1-sigma uncertainty in 20 mag/sq." iso.fid.ell.mag
k_msig_k20feKs 1-sigma uncertainty in 20 mag/sq." iso.fid.ell.mag
j_flg_k20feJ confusion flag for 20 mag/sq." iso. fid. ell. mag
h_flg_k20feH confusion flag for 20 mag/sq." iso. fid. ell. mag
k_flg_k20feKs confusion flag for 20 mag/sq." iso. fid. ell. mag
Kron aperture photometry employs a method in which the aperture is controlled/adapted to the first image moment radius. The Kron radius turns out to roughly correspond to the isophotal radii (see above) under typical observing conditions. More information on the Kron aperture can be found here.

The central surface brightness (mag per arcsec2) is computed for the peak pixel and for the central R <= 5" region:
j_peakJ peak pixel brightness
h_peakH peak pixel brightness
k_peakKs peak pixel brightness
j_5surfJ central surface brightness (r<=5)
h_5surfH central surface brightness (r<=5)
k_5surfKs central surface brightness (r<=5)

Additional information with regard to 2MASS galaxy photometry can be found here:

and more specific studies here:

Extended Source Catalog Field Parameters

The user has the option to download pre-selected fields (mini-set, short-set, or standard-set). For user and database convenience, we have defined a set of "default" magnitudes, corresponding to the Ks fiducial isophotal circular metric (see above). The default mag field names are:
j_mJ selected "default" magnitude
h_mH selected "default" magnitude
k_mKs selected "default" magnitude
Note, however, we recommend using the fixed R=7" circular aperture photometry and the "Ks" fiducial isophotal elliptical aperture photometry (see above).


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