Next Contents Previous

3. WIDE-FIELD MAPPING SURVEYS OF M31

The relative proximity of M31, while advantageous for detailed study, also poses a problem in that even the main body of the galaxy subtends a substantial angle on the sky. In the mid-to-late 90s, wide-format CCD detectors became increasingly available on medium-sized telescopes and this development opened up new possibilities for surveying the faint outlying regions of M31. With a distance modulus of mM = 24.47 ± 0.07 (McConnachie et al. 2005), stars near the tip of M31's RGB (MI ≈ −4) have I ≈ 20.5 and thus could be easily detected in modest exposures with a 2-m class telescope. As a result, it became feasible to contiguously map resolved stars at the bright end of the luminosity function over very large areas.

RGB stars are the evolved counterparts of low to intermediate mass (M ∼ 0.3−8 M) main-sequence stars with ages of at least ≳ 1 Gyr. While asymptotic giant branch stars and high mass main sequence stars can be even more luminous than RGB stars, RGB stars are the most interesting in the context of tidal stream research since they are long-lived and can be used to trace the old stellar components of galaxies, where signatures of hierarchical assembly are expected to be most prevalent. Additionally, for a fixed age, the colour of an RGB star depends almost entirely on its metallicity; thus, for a roughly uniform age for a stellar population, individual stellar metallicities can be derived purely from photometry.

Resolved star surveys map the spatial distribution of individual RGB stars which can in turn be used to infer the surface brightness distribution of the underlying light. This technique allows much lower surface brightness levels to be reached than typically achievable with conventional analyses of diffuse light. As a simplistic illustration of how the method works, consider a population of M31 RGB tip (TRGB) stars with I0 ∼ 20.5 and (VI)0 ∼ 1.5. A surface density of 105 such stars per square degree corresponds to µV ≈ 27 mag arcsec−2 while a surface density of 103 stars per square degree corresponds to µV ≈ 32 mag arcsec−2. This calculation is crude since it neglects the fact that there is a range of RGB luminosities within a population, and also that some sizeable fraction of the total light will come from stars fainter than the magnitude limit, but these are corrections that can be easily calculated for any given survey (e.g. Pritchet & van den Bergh 1994, McConnachie et al. 2010). Nonetheless, it is sufficient to demonstrate that, in the low crowding outer regions of M31 (and external galaxies in general), the resolved RGB star technique is clearly the optimal means to search for and map very faint structures. The main challenges in using this method are to image to sufficient depth to detect a statistically significant population of RGB halo stars, and to disentangle genuine RGB stars from contaminating populations, which typically consist of foreground Milky Way dwarf stars and unresolved background galaxies (see the left panel of Fig. 1).

Figure 1

Figure 1. (Left) A Hess diagram of the point sources in the PAndAS survey at distances beyond 2 of M31 (reproduced from Ibata et al. 2014). A series of fiducial tracks spanning [Fe/H] = −1.91, −1.29, −0.71 and −0.2 are superimposed on the RGB while contaminating Milky Way foreground disk and halo sequences are indicated by dashed boxes. Blueward of (gi)0 ∼ 1 and fainter than i0 ∼ 23, unresolved background galaxies become the primary contaminant. The orange box shows the adopted colour-magnitude selection for M31 RGB stars. (Right) An early map of metal-poor RGB star counts around M31 from the INT/WFC survey (reproduced from Ferguson et al. 2002). The outer ellipse has a flattening of 0.6 and a semimajor axis length of 55 kpc.

The breakthrough in our ability to search for low surface brightness structure in the outskirts of M31 led to the discovery of a plethora of faint streams and substructure in the the inner halo, including the dominant Giant Stellar Stream. The pioneering INT Wide Field Camera survey (e.g. Ibata et al. 2001, Ferguson et al. 2002, Irwin et al. 2005) mapped ≈ 40 square degrees (163 contiguous pointings) around M31 in the V and i passbands, reaching to ∼ 3 magnitudes below the TRGB (see the right panel of Fig. 1). M31 was also targeted by the SDSS (Zucker et al. 2004a), where stars near the TRGB were mapped to large distances along the major axis. These surveys also uncovered several previously-unknown M31 dwarf satellites and globular clusters (e.g. Zucker et al. 2004b, Zucker et al. 2007, Irwin et al. 2008, Huxor et al. 2008).

In parallel to these efforts to survey RGB stars, other groups began to explore the halo and outer disk populations using planetary nebulae (PNe) (e.g. Merrett et al. 2003, Merrett et al. 2006, Morrison et al. 2003, Kniazev et al. 2014). Although PNe offer some advantages over RGB stars as tracers of halo light (for example, they are more luminous, suffer much reduced sample contamination, and can provide simultaneous information on radial velocities), they are far rarer. From a survey of PNe in the M31 bulge, Ciardullo et al. 1989 found that the ratio between the number of PNe in the top 2.5 magnitudes of the luminosity function and the V-band luminosity was α2.5 ∼ 30.8 × 10−9 PNe per L. This leads to the expectation of ≈ 50 PNe per square degree at µV ∼ 24 mag arcsec−2 but only ≈ 1 per square degree at µV∼ 28 mag arcsec−2. This makes surveys for resolved RGB stars far more efficient than those for PNe in the outer regions of M31. However, the situation changes for more distant galaxies, where RGB stars become too faint to resolve while PNe can still be detected (e.g. Coccato et al. 2013, Foster et al. 2014).

Based on the success of these early studies, exploration of the M31 outer halo began in the mid 2000s. Due to the lower stellar density in these parts, deeper photometry was required in order to sufficiently sample the RGB luminosity function and this in turn required larger telescopes. The Pan-Andromeda Archaeological Survey (PAndAS) was conducted using the MegaCam instrument on the 3.8-m CFHT to contiguously map over 380 square degrees around the M31-M33 region and detect stars to ∼ 4 mag below the TRGB (e.g. Ibata et al. 2007, McConnachie et al. 2009, Ibata et al. 2014). Other wide-field ground-based work concentrated on deep pencil beam studies of the outer halo (e.g. Ostheimer 2003, Tanaka et al. 2010).

Next Contents Previous