AAS Meeting #194 - Chicago, Illinois, May/June 1999
Session 82. A Stellar and Variable Star Melange
Display, Thursday, June 3, 1999, 9:20am-4:00pm, Southwest Exhibit Hall

## [82.01] Observations of RR Lyrae in M31 using Difference Imaging Photometry

B.E. Sugerman, R.R. Uglesich, A.P.S. Crotts (Columbia University)

Due to distance and crowding, the incidence and population statistics of short-period variables in the inner parts of M31 are extremely poorly-determined. Specifically, RR Lyrae have an R~5 superposed on a surface brightness between 18 and 20 mag arcsec-2 at the inner disk, making them extremely enigmatic to detect. Pritchet & van den Bergh (1987, ApJ, 316, 517) found only 30 RR Lyrae candidates in the halo [40\prime (9 kpc) from the nucleus] using the CFHT 3.6-m.

Using the Difference Image Photometry'' (Tomaney & Crotts, 1996, AJ, 112, 2872) technique to isolate variable sources within unresolved fields, we seek to better determine the population of RR Lyrae in M31. Observations (including the VATT 1.8-m and INT 2.5-m) were made between Oct.\ 15 and Oct.\ 21, 1996 on a 10\prime \times 10\prime far-side field 1.\prime5 (0.3 to 2.5 kpc) from the nucleus in the bulge/inner disk. Our current study is restricted to INT data. Identification of RR Lyrae is nontrivial since they lie deeply embedded within the unresolved bulge/disk. To achieve the required high S/N to identify candidate varirables, we median-combine differenced data by folding frames along a given test period at evenly-spaced phases. Since the constant background has been removed, any high-\sigma residual is a potential variable-source. We compare the resulting light-curves against custom templates for a variety of short-period variables (including RRab, RRc and eclipsing binaries) to identify true variables and ignore sporadic signal.

Adopting the metallicity and specific-incidence of RR Lyrae in M31 from Pritchet & van den Bergh, KPNO 4-m observations of M31 indicate that we expect to find \lesssim 0.9 RR Lyrae per square arcsec. This corresponds to \lesssim 3 \times 105 RR Lyrae in our field. Since lower incidence indicates higher metallicity, this study is also a good diagnostic for metallicity, which is otherwise difficult to determine due to crowding.

Preliminary estimates from findings on a small subsample of the full dataset indicate that there are gtrsim 2 \times 105 variables with periods between 3 and 24 hours. Statistics for the full dataset, as well as completeness estimates will be presented.

If the author provided an email address or URL for general inquiries, it is a s follows: