1. Introduction
The Two Micron All Sky Survey (2MASS) completed observations in early 2001, with the final source catalogs and image Atlas to be released to the public in the fall of 2002 (Cutri et al 2000). The extended source catalog (XSC; Jarrett et al 2000) contains over 1.65 million galaxies and Galactic extended sources whose integrated fluxes are brighter than Ks < 14.0 (~1.68 mJy). The XSC is complete for all galaxies larger than ~10-15" in diameter, including the largest galaxies in the sky. However, due to their proximity to a survey "scan" edge, galaxies larger than ~1 or 2 arcmin will have photometry that is systematically incomplete. A typical 2MASS survey scan is 8.5 arcmin * 6° with 10% equatorial overlap between scans. Galaxies that are smaller than the overlap, ~50 arcsec, are guaranteed to be fully sampled in at least one survey scan. Larger galaxies may be truncated based on their proximity to a scan edge. Therefore it was necessary to construct an atlas of large objects made from "pieces" of adjoining scans. The net outcome is that we will fully recover galaxies that are currently "lost" or misrepresented in the 2MASS Extended Source Catalog. Image reconstruction of these large galaxies is important because they are among the nearest objects in the local Universe and so their structure can be studied in greater detail than is possible for most objects in the XSC.
The Atlas will consist of galaxies ranging in size from 2 arcmin to 2° (e.g., M31), with a typical spatial resolution of ~2 to 3 arcsec (with 1 arcsec pixels) in the near-infrared (NIR) bands, J (1.2 µm), H (1.6 µm), and Ks (2.2 µm). The completed atlas provides the aggregate flux for each galaxy and a detailed view of the infrared morphology. The Atlas will be part of the XSC and can be accessed through the NASA Extragalactic Database (NED) and Infrared Science Archive (IRSA).
The 2MASS Large Galaxy Atlas will represent a valuable asset to the professional and amateur astronomical communities. The near-infrared forms a natural, complementary, bridge between the well-studied optical wavelengths (e.g., POSS and SDSS) and the longer wavelengths (e.g., IRAS, ISO, NVSS, FIRST). The near-infrared is much less affected by the extinction by interstellar grains (of course, not all galaxies are fully dust-penetrated at 2 µm, as dark dust lanes are seen in some gas-rich spirals and flocculent disks, such as the edge-on galaxy NGC 891 and the more face-on NGC 2841. The latter galaxy's thick dust lane has been studied in detail by Block et al. 1995 and 1999) and more sensitive to the older stellar populations -- the dominant mass component ("backbone") in most galaxies (Aaronson 1978, 1986; Frogel et al 1990, 1996; Rix & Rieke 1993).
The near-infrared provides a distinctived, penetrating view of spiral galaxies. Disk galaxies have two distinct kinematic and evolutionary classes: the older, Population II stars, representing the underlying large-scale "backbone", and the younger, kinematically colder (spiral arm-confined) Population I stars, associated with massive gas-rich star formation regions. Spirals are sub-divided into morphological types ("Hubble Types") based on their optical properties (e.g., surface brightness), and therefore heavily influenced by the visually luminous Population I class -- it is the shape and gas content that separates the spiral types. But it is the underlying Population II "backbone", for the most part decoupled from the Population I stars, that is fundamental to the evolution of the galaxy (see Block et al. 1994, Block & Puerari 1999, Frogel et al. 1996, and references therein.) The 2MASS mosaics convey a clear contrast between the underlying mass component (Population II) and, for example, star formation and GMC regions (Population I elements) as seen in the optical or mid-infrared bands. For this reason, the correlation between the Hubble Classification and the near-infrared morphology for a given galaxy is typically very weak, or even drastically different (cf. Jarrett 2000; and as we show later in this paper). A classification scheme that properly tracks the galaxy evolutionary sequence will probably arise from a fusion of optical and near-infrared measurements. The near-infrared is also sensitive to nuclear rings and large-scale bars, which might be a critical component to fueling active nuclei (Kormendy 1982; Pfenniger & Norman 1990; Eskridge et al 2000). Buta & Block (2001) have proposed to use gravitational potential "bar strengths" as a quantitative metric toward galaxy classification, with NIR imaging central to the technique.
The Large Galaxy Atlas is an ideal data set to combine with imaging data at other wavelengths, the potential for new discoveries by cross-matching is limitless. The Atlas will serve as a mass benchmark for mid/far-infrared imaging from SIRTF, and in particular, the SIRTF-SINGS Legacy survey of nearby galaxies (Kennicutt et al). In addition, the Atlas will function as an astrometric "finding" chart/map for both professionals and amateurs searching for (among other things) supernovae in galaxies. The Atlas may also serve as an astrometric/photometric calibration data set for ongoing and future surveys, such as the OSU multi-wavelength survey of bright spiral galaxies (Frogel et al 1996; Eskridge et al 2000), and the Grauer & Rieke (1998) NIR survey of UGC galaxies.
In this paper we detail the 100 largest galaxies of the Atlas. The initial sample was drawn from the RC3 (de Vaucouleurs et al 1991) using the optical diameter, D25. This B-band diameter is roughly correlated with the NIR diameter, typically a factor of ~1.2 times the NIR equivalent (see Jarrett 2000), and hence is a satisfactory metric for initially selecting the largest angular-sized galaxies in the sky (a notable exception being extreme late-types, whose NIR diameter is <50% compared to the optical diameter, and galaxies obscured by the Milky Way). The sample is then sorted according to the 2.2 µm isophotal diameter, D20. We do not include the Magellanic clouds (or many other Local Group dwarfs) since individual stars in those galaxies are resolved by 2MASS, and they are more appropriately studied with the 2MASS point source catalog. We also employ the RC3 eccentricity and position angle parameters in the background removal process (Section 3). The NIR data is discussed in Section 2, the technique for constructing the mosaics in Section 3, the basic measurements in Section 4. After measuring the NIR sizes of the galaxies from this larger draw, we isolate a sample comprised of the 100 largest galaxies as given by the 2.2 µm size, including NGC 253, presented in Section 5. Finally in section Section 6, we consider the larger sample, ~500 galaxies, compare with the RC3 and discuss the fundamental galaxy measures, including a look at the internal structure of the M51 double galaxy.
