NStars1A: Photometric searches - 2MASS and optical colours

The 1051 sources with K > 9.5 have been cross-referenced against the GSC 2.2 catalogue by Rick White at STScI. That catalogue includes about 435,000,000 sources from the ST scans of POSS (both I and II) plates in the northern hemisphere and from the ESO and UK Schmidt plates in the south. The catalogue lists F, B_J and V magnitudes, and includes some sources as faint as 19th magnitude in R, although it is not complete at those magnitudes. There are a number of 2MASS sources which have (faint) counterparts on POSS II F (R) or N (I), but don't have an entry in the GSC 2.2. However, brighter sources should be reasonably complete, so cross-referencing the ultracool list againt the GSC2.2 will identify most of the bluer stars scattered into the sample due to either photomteric uncertainties in the 2MASS data or intrinsic scatter near the (J-K)=1 cutoff.

It is possible that some images might have no GSC counterpart due to proper motion taking the source outside the 6 arcsecond search radius between the palte-taking and the 2MASS observation. In the north, the mean epoch of GCS2.2 is around 1991, so that would imply a significant annual motion, so the source is probably worth checking anyway. In the south, the ESO R plates were taken in the late 70s and early 80s, and the second epoch UKST F-survey isn't complete - but even so, the annual motion would have to be around 0.3 arcsec yr^-1. Bottom line is that we're interested in sources with no GSC counterpart or (F-J) (effectively (R-J)) colours which match the late-M/L sequence in the RJK plane.

Figure a1.1: The (F-J)/(J-K) two-colour diagram for faint sources with GSC2.2 counterparts. The magenta points mark data for nearby M dwarfs; the yellow points plot known L dwarfs; and the cyan points plot data for the faint sample. The green line is the selection boundary.

80 of the 1051 sources gave GCS2.2 images have GSC counterparts. Figure a1.1 plots the 2-colour RJK diagram for those sources. Data for nearby M and L dwarfs are superimposed for reference. Clearly there are a lot of sources with implausible colours for ultracool dwarfs. There are also a large number of sources right where the M dwarf sequence meets the (J-K)=1 boundary, and a few sources along the L dwarf sequence. We have applied the folowing cutoff to remove the obvious interlopers

Figure a1.2: The (RA, Dec) and JHK distributions for the 809 targets which survive the R-magnitude cut.

Figure a1.2 plots the (RA, Dec) and near-infrared colour-colour and colour-magnitude diatributions for the reduced sample.There are 326 sources at dec > 0^o and 482 south of the equator. The larger number of southern sources probably reflects both the brighter limiting magnitude of the southern ESO R plates and crowding in the Galactic Bulge.

Figure a1.3: The (RA, Dec) and JHK distributions for the 584 viable candidates amongst the faint sample: 307 in the North, 277 in the South.

As a further check, we have gone through the sample and checked each of the surviving objects. A number of targets are obviously in star forming regions, with counterparts not in the GSC2.2; others are associated with the Fornax dwarf, LMC or SMC (probably AGB/carbon stars). Cutting those stars reduces the sample to a total of 584 targets.

Figure a1.4: The 239 targets still requiring observations in the faint sample.

Cross referencing against the literature (via SIMBAD) plus our own observations gives classifications for over 340 of the surviving sample. That sample includes over 45 L dwarfs, 30 of which are new identifications. Figure a1.4 plots the targets still requiring observation. Relatively few remain with (J-K) colours ufficiently red to be L dwarf.

Meeting the neighbours: NStars and 2MASS

The Ultracool sample: I. The faint sources