Astron. Astrophys. 351, L5-L9 (1999)

2. The EROS 2 proper motion survey

2.1. Instrument and data characteristics

The EROS 2 two colour CCD wide-field imager (Bauer et al. 1997) is mounted at the Cassegrain focus of the 1-m Marly telescope at La Silla (Chile). The pixel size is 0".6 and the field of view is . It contains two CCD mosaics, illuminated through a dichroic beam splitter which defines the bandpasses. The visible and red bands are respectively centered close to the Johnson V and Cousins I filters, but considerably broader. Calibration is based on V-I=[0-1] mag stars and magnitudes of redder stars are indicative only.

Proper motion observations are performed one to two hours per dark night, within 90 minutes of the meridian to minimize atmospheric refraction. For the present analysis we used observed close to the South Galactic Pole, and in the Northern Hemisphere, which had been been observed twice or more, down to and . The experiment is expected to last until 2002, and 4 or 5 epochs will eventually be available.

2.2. Proper motion determination

The reduction software for source detection, classification and catalogue matching was written in the framework of the EROS PEIDA++ package (Ansari 1996), and processes data in (11 arcmin)² chunks. As photon noise dominates the astrometric errors for most of the available volume, we use a simple two-dimensional gaussian PSF fitting to determine stellar positions. A rough star/galaxy classification is performed to limit galaxy contamination, with such cuts that few stars are misclassified. The catalogues for the two epochs are geometrically aligned using a linear transformation adjusted to the 40 brightest stars, and matched within a search radius of 9 arcsec. The average distance between matched stars (Fig. 1) provides an upper limit to the total astrometric error, which for bright objects is 25 mas (). For a 25 km/s disk star and a 1-year baseline this corresponds to a 5 detection at 40 pc.

Fig. 1. Matching distance versus EROS flux in the visible band. For clarity only 5% of the sample is displayed. The dotted vertical line shows the minimum accepted flux, and the solid curve the minimum proper motion cut. The larger dots show the stars selected in the visible band only; only LHS 102B is selected in both bands.

2.3. Candidate selection

Since our main goal is to identify dark matter contributors, we select all objects satisfying a magnitude-dependent proper motion cut, set to retain stars faster than V_t=25 km s^-1 and fainter than and , down to our detection limit (Fig. 1). This selection in a proper motion-magnitude diagram is mostly sensitive to two object classes: halo white dwarfs and disk very low mass stars and brown dwarfs. As the current analysis is based on two epochs, we presently have to require a selection in both photometric channels to avoid excessive contamination by spurious faint candidates. Even though this procedure reduces the searched volume very significantly for L dwarfs, the one candidate it does select, LHS 102B, has a very red EROS colour, confirmed by DENIS photometry. Following this early success, we looked for all red objects with EROS colour similar to LHS 102B's, regardless of their proper motion, obtaining 25 additional candidates with .

© European Southern Observatory (ESO) 1999

Online publication: November 2, 1999
helpdesk.link@springer.de