The most basic definition of a galactic bulge is that of an over-density that swells up from the plane of the disc. The idea of this natural conception originated from the observations of other disc galaxies, in particular the bulges of edge-on spirals, which allow us to compare them more easily with our own in a purely morphological way.
Within spiral galaxies, we could in principle distinguish between classical bulges and those bulges which are formed via secular evolution, based almost solely on their morphological signatures. Clearly, a proper characterisation of the structural properties of the Milky Way bulge would provide us with a valuable set of constraints needed to find its place in the more general scheme of external bulges and, as extensively shown by galaxy formation models, to connect these constraints to the different mechanisms of origin.
After it was first postulated by de Vaucouleurs (1964), followed by Sinha (1979) and Liszt and Burton (1980) among others, Blitz and Spergel (1991) predicted the presence of the bar-like structure for the inner regions of the Galaxy based on infrared observations. Later on, continuing with the exploitation of infrared imaging in order to overcome the strong dust obscuration towards the inner galaxy, the COBE/Diffuse Infrared Background Experiment (Smith et al 2004) data was used by Weiland et al (1994) to unambiguously establish the presence of the bar. Soon after, the COBE data further revealed the global B/P morphology of the Milky Way bulge (Dwek et al 1995).
An important number of Bulge structural studies have been based on the stellar counts of red-clump stars, which are the metal-rich counterpart of the well known globular cluster horizontal-branch stars. The absolute magnitudes of red-clump stars, are found to have little dependence on age and metallicity, making them one of the most powerful tools for deriving distances towards the bulge and therefore tracing its global morphology. This method is based on the construction of the luminosity function of the Bulge towards a given line of sight where the red-clump feature can be easily identified and fitted with a Gaussian distribution to obtain the mean red-clump magnitude (Stanek et al 1994). Zoccali (2010) and McWilliam et al (2010) presented the discovery of a split within the red-clump when investigating the luminosity function of the Bulge at latitudes |b| > 5, along the minor axis. Soon enough, McWilliam and Zoccali (2010) and Nataf et al (2010) provided a wider mapping of this split red-clump in the color magnitude diagram, providing substantial evidence for the bright and faint red-clumps to be the consequence of having two over-densities of stars located at different distances, namely the two southern arms of an X-shaped structure both crossing the lines of sight. Detailed three-dimensional maps were later constructed by Saito et al (2011), based on the use of red-clump stars observed in the near-IR survey 2MASS, which confirmed the suggestion of McWilliam and Zoccali (2010) that the Bulge is in fact X-shaped due to the prominent vertices of the B/P. Wegg and Gerhard (2013) modelled the distribution of red-clump stars, observed in Vista Variables in the Via Lactea (VVV) ESO public survey, providing the first complete mapping of the X-shaped Bulge. These X-shaped bulges are commonly observed in external edge-on galaxies and belong to the case of a pronounced B/P structure, that is simply the inner regions of the bar that grow out of the plane of the disc.
Currently, the axial ratios of the bar are constrained to be about 1:0.4:0.3 with a bar size of about 3.1−3.5 kpc diameter and the near end of the bar pointing towards positive Galactic longitudes. Until recently, the position angle of the bar was constrained to a relatively large range of values between ∼ 20−40 degrees with respect to the Suncentre line of sight. The uncertainty in the bar position angle is likely to be a consequence of measurements done across different latitudes in each study, thus finding a different position angle when looking at different distances from the Galactic plane. As a matter of fact, specific evidence for a longer flatter component of the bar, referred to as the Galactic long bar, has been presented in the literature based on near-IR star counts near the Galactic plane (Blitz and Spergel 1991, Stanek et al 1994, Dwek et al 1995, Binney et al 1997, Bissantz and Gerhard 2002, Benjamin et al 2005, Babusiaux and Gilmore 2005, Rattenbury et al 2007, Cao et al 2013). This long bar is found to have an axis length of 4−4.5 kpc and ratios of 1:0.15:0.03. The position angle of this longer component has been constrained to ∼ 45 degrees in such studies (e.g. López-Corredoira et al 2007, Cabrera-Lavers et al 2007, Hammersley et al 2000, Churchwell et al 2009, Amôres et al 2013). On the other hand, the recent model of the global distribution of red-clump stars from Wegg and Gerhard (2013) provided a precise measurement for the B/P Bulge position angle of 27± 2 degrees, in agreement with the studies done at larger distances from the Galactic plane. The nature of the long bar has been debated extensively in the literature. Recently, Garzón and López-Corredoira (2014) provided a theory where two co-existing bars, the long bar restricted to the plane latitudes and the B/P thick bar, could be present in the inner Galaxy. However, model observations of barred galaxies led Martinez-Valpuesta and Gerhard (2011), Romero-Gómez et al (2011), and Athanassoula (2012) to strongly argue that the apparent long bar is an artefact associated with leading spiral features at the end of the shorter primary bar (the B/P Bulge). Furthermore, the co-existence of such independently large scale structures has not been seen in external galaxies. For this reason, the observed properties attributed to different bars in the Galaxy are more likely corresponding to a unique B/P bulge and bar structure formed by the buckling instability process. The long bar would then be explained by the interaction of the outer bar, with the adjacent spiral arm near the plane which produced leading ends that ultimately results in the measurement of a larger position angle (Martinez-Valpuesta and Gerhard 2011).
In the innermost regions (l < 4, b < 2) the bar has been found to change its apparent inclination with respect to the line of sight which has been interpreted as evidence for a possible distinct smaller bar, referred to as a nuclear bar. However, models of a single bar (meaning those that do not include a distinct nuclear bar) have also shown such a change in orientation in the inner regions, most likely due to the presence of a more axisymmetric concentration of stars in the central regions (Gerhard and Martinez-Valpuesta 2012).
Red-clump stars, although an excellent distance indicator for the bulge mean population, still suffer from the usual complications when looking towards the inner Galaxy such as disc contamination, extinction, and, to a minor extent, the effects of stellar populations. Furthermore, red-clump stars will map the distribution of the variety of stellar ages of the Bulge population, which is not defined beforehand. This uncertainties can be statistically handled when constructing and analysing the Bulge luminosity function, however their impact on the results will depend on the level of knowledge of the properties of each analysed field. For this reason, Variable stars, specifically RR-Lyrae, have recently provided a new perspective for the Bulge structural properties. Their well defined period-luminosity relation in the near-IR helps to overcome the effects of dust extinction in the inner Galaxy and they are well spatially distributed across the entire bulge. The period-luminosity relation makes RR-Lyrae an exquisitely accurate distance indicator that unequivocally traces the oldest Galactic population.
Surprisingly, given the vast amount of different tracers that have confirmed a dominant barred structure, RR Lyrae have shown a remarkably different spatial distribution compared to, for example, red-clump stars. While red-clump stars trace the position angle of the bar at all latitudes, a direct comparison with the RR Lyrae distance distribution provided by Dékány et al (2013) strongly suggests a different morphology for the oldest population in the inner Galaxy. Unlike the red-clump stars, the RR Lyrae stars show a more spheroidal, centrally concentrated distribution. This structural component, populated by stars with ages larger than 10 Gyr, seems to be overlying with the B/P bulge. Figure 1 shows a comparison of the projected mean distances obtained from RR Lyrae and those from the mean magnitude distribution of red-clump stars. The figure illustrates the structural difference between the components traced by both distance tracers, with only red-clumps stars following the position angle of the bar. This result presented in Dékány et al (2013) is perhaps the first purely morphological evidence suggesting a composite Bulge nature, with two different stellar populations overlapping in the inner Galaxy.
Figure 1. Upper panel: Spatial distribution in Galactic coordinates for the 7663 OGLE-III RRab stars in the bulge area of the VVV Survey, from Dékány et al (2013). The grey scale background shows the interstellar extinction map of Schlegel et al (1998). The black rectangle denotes the region at b = −4 for which the mean distance of RR Lyrae and RC stars are compared in the lower panel. The lower panel shows the projected mean distances of RR Lyrae in black filled circles and of the red-clump stars as red open circles. Isodensity contours for the projected distance distribution of the RR Lyrae sample in the analysed latitude range are also shown. Mean distances of the red clump stars are from the mean magnitudes obtained in Gonzalez et al (2012) and calculated adopting an absolute magnitude of MKs,RC = 1.71 mag. Distances for the sample of RR Lyrae have been presented in Dékány et al (2013) and were used here to derive the projected mean distance and 1σ width (black solid lines) to each line of sight. [Upper panel adapted from Dékány et al. (2013)]
The presence of more than one age/metallicity distribution within a B/P bulge has already been seen in dissipative collapse models (e.g. Samland and Gerhard 2003) and also in bulges from cosmological galaxy formation simulations (e.g. Obreja et al 2013). However, one must be very careful when further linking the different spatial distributions seen in the Galactic bulge with a distinct origin process, namely having a classical bulge and a secularly evolved B/P bulge that originated from the disc. Recently, Ness et al (2014) gives caution to the fact that different spatial distributions and mean stellar ages can be found in pure B/P bulges without the need of a merger-origin structure to be present, as seen in an N-body + smoothed particle hydrodynamics simulation of a disc galaxy. Certainly, the fine details of the shape traced by RR-Lyrae will be achieved when the complete sample of RR-Lyrae from the VVV survey is available. The mapping of a wider area of the Bulge and the larger sample of sources available in each line of sight will allow for the calculation of precise mean distances with high spatial resolution.