Infrared studies of the structure of the galaxy

Abstract

A simple three-component model of the Galaxy is presented. The Galaxy is represented by (1) a disk whose stellar density decreases exponentially with distance from the galactic centre as well as perpendicularly as a function of height above or below the galactic plane; plus a (2) ring with a radius of R₀sin25 kpc considered to be an enhancement of the disk and finally by (3) an oblate de Vaucouleurs spheroid meant to represent the galactic bulge. The absorbing material is also assumed to be distributed as an exponential layer. The model is used to fit the 2.4 μm integrated flux observations of the Galaxy, which it does remarkably well, and to determine some of the model constants such as the radial length scale for the disk and the central densities for the bulge and ring components. Based on this model and the derived values of the constants, a series of stellar count predictions for the Km agnitude (2.2 μm) were made. These predictions were com pared to those made by the Bahcall and Soneira model of the Galaxy. The model predicts an almost complete dominance of the bright K-magnitude (K ≤ 14-15) realm by the disk population. At fainter magnitudes the bulge population dominates. A transition region at which disk and bulge contributions are similar occurs at different K -m agnitudes and depends on galactic longitude (from K ~ +9.5 at l ~ 0 to K ~ +20 at l ~ 180). Stellar counts along the galactic plane in the range 220 ≤ l ≤ 60 have been obtained, and have been fitted to the galactic model described above. The model constants utilised in fitting the stellar counts are those derived from the model fit to the 2.4 μm observed integrated flux. Considering the general nature of the model (same constants for all regions), the agreement between the model predictions and the observations is remarkable. The slight discrepancies which have arisen between theory and observations have been interpreted, in most cases, as inhomogeneities in the absorbing material. In the longitude range 0 ≤ l ≤ 60, there appears to be a slight theoretical excess over the observed counts at bright magnitudes. A carefully chosen alteration of the absolute K-magnitude of all the stars in the Luminosity Function (LF) may be necessary to produce a perfect fit between predicted and observed Cumulative Counts Functions (CCF’s). A low absorption window in the direction l — 0, 6 ~ -4, known as Baade’s Window (BW) has been observed photometrically as well as spectroscopically. A K-scan of this region provided an observational CCF which, with the aid of the model discussed above, was decomposed into possible disk plus bulge contributions. The following results were obtained: • At bright K-magnitudes the slope of the disk CCF is different from that of the bulge CCF. This suggests the presence of two radically different stellar populations within a radius of ± 3 kpc from the galactic centre. • Possible similarities between the real BW bulge population and globular clusters stars is suggested from the similar slope of both stellar populations’ CCF’s. • A significant spread in the (J-H) vs. (H-K), as well as in the CO vs. (J-K) diagram suggests an intrinsic spread in the colour of the sources, which may indicate a possible spread in the metallicity of the objects observed in BW. • What may be identified as the brightest infrared (IR) magnitude locus for stars in BW, is composed of young (2 x 10⁹ years) Asymptotic Giant Branch (AGB) stars with a luminosity of the order ~ 10⁴Lʘ. • In BW ‘true’ bulge members appear to be CO-poor as opposed to ‘true’ disk members which seem to have higher CO-index values. This suggests there is a difference in metallicity of the stars in the disk and the bulge. ‘True’ bulge members appear to be more metal poor than 47-Tuc. There is a hint of kinematical differences between CO-weak and CO-strong stars which is consistent with our picture of ‘true’ bulge and ‘true’ disk populations respectively. Further photometric, spectroscopic and kinematical studies of IR sources in low absorption regions will help to confirm the points put forward in this thesis.