=== ABSTRACT ===

The ATLAS All-Sky Stellar Reference Catalog

J. L. Tonry, L. Denneau, H. Flewelling, A. N. Heinze, C. A. Onken, S.J. Smartt, B. Stalder, H. J. Weiland, and C. Wolf

The Asteroid Terrestrial-impact Last Alert System (ATLAS) observes most of the sky every night in search of dangerous asteroids. Its data are also used to search for photometric variability, where sensitivity to variability is limited by photometric accuracy. Since each exposure spans 7.6◦ corner to corner, variations in atmospheric transparency in excess of 0.01 mag are common, and 0.01 mag photometry cannot be achieved by using a constant flat field calibration image. We therefore have assembled an all-sky reference catalog of approximately one billion stars to m ∼ 19 from a variety of sources to calibrate each exposure’s astrometry and photometry. Gaia DR2 is the source of astrometry for this ATLAS Refcat2. The sources of g, r, i, z photometry include Pan-STARRS DR1, the ATLAS Pathfinder photometry project, ATLAS re-flattened APASS data, SkyMapper DR1, APASS DR9, the Tycho-2 catalog, and the Yale Bright Star Catalog. We have attempted to make this catalog at least 99% complete to m < 19, including the brightest stars in the sky. We believe that the systematic errors are no larger than 5 millimag RMS, although errors are as large as 20 millimag in small patches near the galactic plane.

=== REFCAT2 ===

Refcat2 is normally provided as a bzip2 compressed tar archive of 64800 files, one for each (coordinate) square degree in the sky. The file names reflect the coordinate location, rrr+dd.rc2. For example 270-20.rc2 is the square degree with 270 <= RA < 271 and -20 <= Dec < -19. The stars in each square degree file are sorted by increasing RA. The data are given in comma separated variable format (CSV), using scaled integers for all real numbers as described in Table 4. Abbreviations include "10ndeg" for 1e-8 degree, "10uas" for 1e-5 arcsecond, "mas" for 1e-3 arcsecond, and "mmag" for 1e-3 magnitude.

                 ===  Refcat2 table columns ===

Field Name Entry Units Meaning Description

1 RA 28000001672 [10ndeg] 280.00001672 deg RA from Gaia DR2, J2000, epoch 2015.5 2 Dec -1967818581 [10ndeg] -19.67818581 deg Dec from Gaia DR2, J2000, epoch 2015.5 3 plx 98 [10uas] 0.98 mas Parallax from Gaia DR2 4 dplx 10 [10uas] 0.10 mas Parallax uncertainty 5 pmra 114 [10uas/yr] 1.14 mas/yr Proper motion in RA from Gaia DR2 6 dpmra 16 [10uas/yr] 0.16 mas/yr Proper motion uncertainty in RA 7 pmdec -1460 [10uas/yr] -14.60 mas/yr Proper motion in Dec from Gaia DR2 8 dpmdec 15 [10uas/yr] 0.15 mas/yr Proper motion uncertainty in Dec 9 Gaia 15884 [mmag] 15.884 Gaia DR2 G magnitude 10 dGaia 1 [mmag] 0.001 Gaia DR2 G magnitude uncertainty 11 BP 16472 [mmag] 16.472 Gaia GBP magnitude 12 dBP 10 [mmag] 0.010 Gaia GBP magnitude uncertainty 13 RP 15137 [mmag] 15.137 Gaia GRP magnitude 14 dRP 1 [mmag] 0.001 Gaia GRP magnitude uncertainty 15 Teff 4729 [K] 4729K Gaia stellar effective temperature 16 AGaia 895 [mmag] 0.895 Gaia estimate of G-band extinction for this star 17 dupvar 2 2 Gaia flags coded as CONSTANT (0), VARIABLE (1), or NOT_AVAILABLE (2) + 4*DUPLICATE 18 Ag 1234 [mmag] 1.234 SFD estimate of total column g-band extinction 19 rp1 50 [0.1"] 5.0" Radius where cumulative G flux exceeds 0.1x this star 20 r1 50 [0.1"] 5.0" Radius where cumulative G flux exceeds 1x this star 21 r10 155 [0.1"] 15.5" Radius where cumulative G flux exceeds 10x this star 22 g 16657 [mmag] 16.657 Pan-STARRS g_P1 magnitude 23 dg 10 [mmag] 0.010 Pan-STARRS g_P1 magnitude uncertainty 24 gchi 23 [0.01] 0.23 chi^2/DOF for contributors to g 25 gcontrib 1f [\%02x] 00011111 Bitmap of contributing catalogs to g 26 r 15915 [mmag] 15.915 Pan-STARRS r_P1 magnitude 27 dr 12 [mmag] 0.012 Pan-STARRS r_P1 magnitude uncertainty 28 rchi 41 [0.01] 0.41 chi^2/DOF for contributors to r 29 rcontrib 3f [\%02x] 00111111 Bitmap of contributing catalogs to r 30 i 15578 [mmag] 15.578 Pan-STARRS i_P1 magnitude 31 di 10 [mmag] 0.010 Pan-STARRS i_P1 magnitude uncertainty 32 ichi 49 [0.01] 0.49 chi^2/DOF for contributors to i 33 icontrib 0f [\%02x] 00001111 Bitmap of contributing catalogs to i 34 z 15346 [mmag] 15.346 Pan-STARRS z_P1 magnitude 35 dz 12 [mmag] 0.012 Pan-STARRS z_P1 magnitude uncertainty 36 zchi 0 [0.01] 0.00 chi^2/DOF for contributors to z 37 zcontrib 06 [\%02x] 00000110 Bitmap of contributing catalogs to z 38 nstat 0 0 Count of griz deweighted outliers 39 J 14105 [mmag] 14.105 2MASS J magnitude 40 dJ 36 [mmag] 0.036 2MASS J magnitude uncertainty 41 H 14105 [mmag] 14.105 2MASS H magnitude 42 dH 53 [mmag] 0.053 2MASS H magnitude uncertainty 43 K 13667 [mmag] 13.667 2MASS K magnitude

44 dK 44 [mmag] 0.044 2MASS K magnitude uncertainty

When there is no matching Gaia star, dGaia = 0. When a magnitude is not available (for example 2MASS at the faint end) the magnitude and its uncertainty are set to 0, otherwise the magnitude uncertainty is given as at least 1~millimag.

The g band total column extinction Ag is computed from the E(B{-}V) values of SFD, multiplied by 0.88 as recommended by Schlafly11, and also by Ag/E(B-V) = 3.613 - 0.0972 (g-i) + 0.0100 (g-i)^2 (Tonry11).

The proximity statistics rp1, r1, and r10 are derived by adding up the cumulative G band flux of all Gaia stars as a function of distance from each star, and reporting the radius where this flux first exceeds 0.1 (rp1), 1 (r1), and 10 (r10) times the flux of the star. These are given the value 999 (99.9") when a star is so isolated that the cumulative flux never reaches the threshold within the 36" search radius.

The griz-contrib entries identify contributors to the griz magnitudes. Bits 0--7 are set when a catalog contributes to the statistical average with magnitude uncertainty less than 0.2: Gaia DR2 (bit 0), GMP (bit 1), Pan-STARRS (bit 2), SkyMapper (bit 3), Pathfinder (bit 4), APASS (bit 5), APASS DR9 (bit 6), and Tycho-2/BSC (bit 7). For example the code rcontrib=06 implies that r contributions with uncertainty less than 0.2 mag came from GMP and Pan-STARRS.

Gaia DR2 does not include all bright stars; Polaris is missing, for example, as well as a m~12 star at RA 93.7759, Dec +15.0451. Stars found in the contributing catalogs that are no closer than 3.6" (PanSTARRS, SkyMapper, and Tycho/BSC) or 10.8" (Pathfinder and APASS) to the nearest Gaia star are added to Refcat2 with zero values for all Gaia-specific quantities. A real star can therefore be inhibited from inclusion because Gaia DR2 lists a faint star nearby, except for Tycho/BSC for which the Gaia match must be within 2 magnitudes of the non-Gaia candidate. A non-Gaia star may be identified in Refcat2 because it will always have dGaia = 0.

Examining a statistical sample of these non-Gaia objects, we see a change in behavior around g~12, where the Tycho-2 catalog ends. About 78% of the non-Gaia inclusions in Refcat2 brighter than g~12 really are stars, about 18% are double star blends mostly from APASS DR9, and the remainder are bright galaxies, completely false triggers (often in the outskirts of a bright star), or coordinates that did not link with the Gaia stars (possibly a transient). The double star blends almost always have the individual stars also present in Refcat2 as Gaia-matching entries. About 0.25% of the stars with m<10 in Refcat2 do not appear in Gaia DR2; 8946 stars in Refcat2 brighter than m<12.5 are missing from Gaia DR2.

Fainter than g~12 the fractions of stars, galaxies, and false triggers near bright stars change discontinuously. About 31% appear to be real stars, 27% appear to be diffraction artifacts from bright stars, 23% are galaxies, and 19% are probably transients, eruptive stars, or errors. About 0.1% of Refcat2 is not found in Gaia DR2 for these fainter sources.

A Refcat2 user should therefore use these non-Gaia objects according to application. If a relatively complete sample of m>12 stars is needed for an astrometric or photometric solution there is no reason to include any non-Gaia stars and selecting on dGaia > 0 may be advisable. If the user wants to know whether there is a star with m<3 within a degree, then using the non-Gaia entries is mandatory. If a detection appears in a difference image it is a good idea to check the non-Gaia entries as well for the identity of the object, but there is a chance (~0.1%) that a non-Gaia entry may not really be a bona fide star at the specified location.

The catalog is distributed in five magnitude chunks without overlap, * gri<16 (00_m_16.tbz, 105M stars, 5.9GB), * 16 <= gri<17 (16_m_17.tbz, 107M stars, 5.6GB), * 17 <= gri<18 (17_m_18.tbz,204M stars, 9.8GB), * 18 <= gri<19 (18_m_19.tbz,369M stars, 17GB), and * 19 <= gri (19_m_20.tbz,206M stars, 8.7GB). Any star that has (16 <= m<17) where m is the brightest of g, r, and i lies in the second chunk, etc. Red stars with g>16 will appear in the first chunk, but a user who wants a sample with g<17, for example, needs only examine the first two chunks. Stars appear in the fifth chunk with gri>19 because our Pan-STARRS selection also includes red stars with z<19, but that chunk is very incomplete.

The entire compressed catalog amounts to about 50 bytes per star (about 1 byte per field), depending on how many fields are populated. The expectation is that users will select a subset of stars and fields chosen according to their application, which can greatly reduce the size and increase the access speed of a database table. The data are partitioned into square degree files so that a survey such as ATLAS, whose field of view will typically touch ~40 such files, can dispense with a database and simply use the file system to rapidly return all the stars in a given exposure.


ATLAS observations and this work were supported by NASA grant NN12AR55G. The AAVSO Photometric All-Sky Survey (APASS) was funded by the Robert Martin Ayers Sciences Fund. SJS acknowledges funding from STFC Grants ST/P000312/1 and ST/N002520/1.

This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa. int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/ web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

Data products from the Two Micron All Sky Survey were used, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center/California Institute of Technology, funded by the National Aeronautics and Space Administration and the National Science Foundation.

The Pan-STARRS1 Surveys (PS1) and the PS1 public science archive have been made possible through contributions by the Institute for Astronomy, the University of Hawaii, the Pan-STARRS Project Office, the Max-Planck Society and its participating institutes, the Max Planck Institute for Astronomy, Heidelberg and the Max Planck Institute for Extraterrestrial Physics, Garching, The Johns Hopkins University, Durham University, the University of Edinburgh, the Queen’s University Belfast, the Harvard-Smithsonian Center for Astrophysics, the Las Cumbres Observatory Global Telescope Network Incorporated, the National Central University of Taiwan, the Space Telescope Science Institute, the National Aeronautics and Space Administration under Grant No. NNX08AR22G issued through the Planetary Science Division of the NASA Science Mission Directorate, the National Science Foundation Grant No. AST-1238877, the University of Maryland, Eotvos Lorand University (ELTE), the Los Alamos National Laboratory, and the Gordon and Betty Moore Foundation.

The national facility capability for SkyMapper has been funded through ARC LIEF grant LE130100104 from the Australian Research Council, awarded to the University of Sydney, the Australian National University, Swinburne University of Technology, the University of Queensland, the University of Western Australia, the University of Melbourne, Curtin University of Technology, Monash University and the Australian Astronomical Observatory. SkyMapper is owned and operated by The Australian National University’s Research School of Astronomy and Astrophysics. The survey data were processed and provided by the SkyMapper Team at ANU. The SkyMapper node of the All-Sky Virtual Observatory (ASVO) is hosted at the National Computational Infrastructure (NCI). Development and support the SkyMapper node of the ASVO has been funded in part by Astronomy Australia Limited (AAL) and the Australian Government through the Commonwealth’s Education Investment Fund (EIF) and National Collaborative Research Infrastructure Strategy (NCRIS), particularly the National eResearch Collaboration Tools and Resources (NeCTAR) and the Australian National Data Service Projects (ANDS).