The stellar contents of the Milky Way and nearby galaxies make up the fossil records that were produced during the galaxy formation and evolution. The spatial, temporal and abundance distributions of these stars provide important clues to the underlying astrophysical processes involved and also complement the study of galaxy formation and evolution at high redshift. We discuss a number of outstanding questions, answers to which are expected from TMT observations.
A large fraction of stars in the Milky-Way are part of binary or higher order systems, with separations < 0.04 parsecs (Larson 1995). Binaries can be identified by direct imaging or spectroscopically. The spectroscopic detection of binaries can be a laborious process that typically requires observations that span a long time baseline. Two possible imaging programs are discussed here that exploit the unprecedented angular resolution of the TMT to probe the binary frequency in very different environments:
Stars, and in particular the most massive stars, rarely form in isolation. In fact, it is now well established that the vast majority of active star formation occurs in clusters of some sort. Despite significant recent progress, the evolutionary connection between the recently formed young massive clusters (YMCs) in starbursts and old globular clusters in the nearby Universe is still contentious. The evolution and survivability of young clusters depend crucially on their stellar initial mass function (IMF): if the clusters are significantly depleted in low-mass stars compared to, for example, the Solar neighborhood, they will likely disperse within about a Gyr of their formation. As a simple first approach, one could construct diagnostic diagrams for individual YMCs, of mass-to-light ratio (M/L; derived via dynamical mass estimates using high-resolution spectroscopy and the virial approximation) versus age (derived from spectral features), and compare the YMC locations in this diagram with models of “simple stellar populations” (SSPs) governed by a variety of IMF descriptions. However, such an approach has serious shortcomings and suffers from a number of fundamental problems.
The essential conditions to make a major leap forward are to obtain high-resolution spectroscopy and imaging of a significantly larger cluster sample than available at present (to distinguish between trends and coincidences), covering a much more extended age range. These observations will need to be backed up by detailed N-body simulations of clusters containing both a realistic number of test particles (upwards of several x 105) and all relevant physical processes occurring over the clusters’ lifetimes. Using TMT-sized apertures will allow us to probe both the dynamics and the luminosity function of young and intermediate-age star clusters (and their host systems) out to cosmologically interesting distances, where we can obtain statistically significant samples of galaxy types spanning the entire Hubble sequence, and of their YMC systems. Using this approach, the initial conditions for cluster formation (and evolution) can be modeled fully self-consistently. This will, for the first time, provide us with the firmest handle yet on potential IMF variations in external galaxies. For the expected velocity dispersions of ~7−15 km s−1 (for masses of ~ 105 − 106 M蕵), a spectral resolution of R > 40, 000 is essential to efficiently sample the prevailing IMF conditions in a statistically significant number of YMCs. As long as the masses of the clusters are not too small (Mcl > 105 M蕵, depending on the cluster’s size), one can extract their velocity dispersions using suitably chosen cool giant and supergiant template stars (Ho & Filippenko 1996). Using R = 40, 000, one can resolve velocity dispersions down to σv ~ 6.5 km s−1 at a wavelength of 0.85 μm, i.e. to Mcl ~2 × 105 M蕵 for YMCs of globular cluster size. The spectral range around the CaII triplet (λcentral = 860 nm) with a large number of metal lines in this spectral range can be used, and cross-correlated with a number of properly selected (super-)giant velocity template stars over the entire observed wavelength range, to obtain our velocity dispersion measurements.
Complementary imaging observations covering – ideally – at least 4 pass bands spanning a minimum of the entire optical wavelength range will allow to independently and robustly determine the cluster properties (age, mass, metallicity) using the sophisticated multi-parameter AnalySED algorithm (de Grijs et al. 2003, Anders et al. 2004), which has been shown to produce robust results.
The first, so-called Population III stars are believed to have been formed from gas unpolluted by heavy elements, after which all subsequent generations of stars contained increasing fractions of metals. This process has continued during the entire lifetime of the Universe. Hence the most metal-poor stars in the Milky Way and other galaxies are the oldest stars. In their atmospheres these old objects preserve details of the chemical composition of their birth gas cloud. These very old stars are hidden amongst a vast number of stars formed later in time. Once found, understanding the chemical composition of the oldest stars provides a direct probe of the initial conditions of star formation as well as the details of chemical evolution and nuclear astrophysics in the early universe.
Spectral comparison of stars in the main-sequence turn-off region with different matellicities. Several atomic absorption lines are marked. The variations in line strength reflect the different matellicities. From top to bottom: Sun with [Fe/H] = 0.0, G66-30 with [Fe/H] = -1.6 (Norris et al. 1997), G64-12 with [Fe/H] = -3.2 (Aoki et al. 2006), and HE 1327-2326 with [Fe/H] = -5.4 (Source: Frebel, 2010)
In the spectra of the metal-poor stars, since the relevant absorption lines appear quite weak, the necessary chemical composition studies require high spectral resolution (R = λ/δλ~ 50 000 or greater), high SNR (~ 100 per spectral resolution element) observations. These requirements lead to extremely long integration times on the largest telescopes currently available. For example, an 18-hour integration with UVES on the VLT was obtained in an attempt to determine the oxygen abundance in HE1327-2326, the most extreme metal-poor turn-off star known ([Fe/H] ~ -5.5 dex, U 13.8 V 13.5), but even with such a long exposure, the UV-OH band was not detected; its spectrum showed only very weak metal absorption lines with weak CH and NH bands detected was required to determine abundances for HE1327-2326, the most extreme metal-poor turn-off star ([Fe/H] ~ 5.5 dex); its spectrum shows only weak metal absorption lines (Frebel et al. 2006).
The number of stars accessible to TMT will be much larger; for the seeing-limited high-resolution spectrograph HROS, the total improvement relative to VLT/UVES or Keck/HIRES or Subaru/HDS is expected to be 15 – 20. For example, a 4-hour integration with HROS will enable R = λ/δλ ~ 40 000 spectroscopy with SNR = 100 per spectral resolution element for stars as faint as V ~ 21 in the visible.
Positions of planetary nebulae (PNe; the magenta dots) overlying on the ten mosaic fields of M31 that were targeted in the Local Group Survey by Massey et al. (2006). The mosaic images are in V band (in the logarithm scale, inverted color) and each field is roughly 36’ × 36’ in size. The PNe sample (in total ~3000) includes those observed by Merrett et al. (2006) and LAMOST (Yuan et al. 2010; Yuan et al. 2015). Horizontal divisions of the dashed grid are 0.75°, and vertical divisions are 1°. (Source: Yuan et al. 2010).
Recent estimates suggest that close to 70% of the stars in the local Galactic field population are M dwarfs, and about 6% are white dwarfs. The local population of old stars from the Galactic halo is expected to hold a larger fraction of M dwarfs and white dwarfs. These stars however have low luminosities, and are not amenable to detailed spectroscopic analysis with current telescopes unless they lie within about 100-200 parsecs of the Sun.
Low-mass stars and white dwarfs can be highly useful to map out the halo population for three reasons: 1) they are by far the dominant stellar population of the halo; they are the old population, holding the majority of the baryonic mass, 2) their 3D velocity components can be determined to much higher accuracy, because of their large proper motions, 3) physical properties of low-mass stars, and in particular their metallicity, can be constrained using low to medium resolution spectroscopy, because their spectra are dominated by broad molecular bands, and with suitable spectrum synthesis techniques, can now be modeled in detail.
Existing and upcoming deep imaging surveys such as Pan-STARRS and Rubin Observatory are expected to identify low-mass stars in huge numbers (millions) through proper motion analyses. These surveys will effectively provide a statistically complete census of the halo population out to several kiloparsecs from the Sun. Spectroscopic data is required to measure and constrain the temperature and metallicity of M dwarfs, and also to determine the mass and ages of white dwarfs. As the photometric distances of M dwarfs and white dwarfs are generally unreliable, spectroscopic data are needed to constrain their luminosities. Thus a better understanding of the spatial distribution and kinematics of these objects depends on spectroscopic follow-up observations.
The sensitivity of the TMT along with the multi-object capabilities provided by WFOS will make possible a large-scale spectroscopic follow-up of low-mass stars and white dwarfs of the Galactic halo. These observations will enable:
Along with proper motion and photometric data from large imaging surveys, such as GAIA, the spectroscopic data from TMT will produce a detailed census of the nearby halo population, drawing their detailed distribution in velocity space. The combination of metallicity information from the M dwarfs and of age information from the white dwarfs will identify possible substructure in the velocity space distribution. This will provide additional constraints on the shape of the halo, and also on the possible existence of accretion events, that would show up as streams in the local volume.
The Local Group has been the benchmark for testing and calibrating many aspects of cosmology and galaxy formation, hence a clear understanding of the internal kinematics of the Local Group as well as the nearby groups is necessary. The formation and evolution of the local group of galaxies in the Local Volume is not clearly understood, as we do not have reliable estimates of the space velocities of the galaxies. The major hindrance in the estimation of space velocity is the difficulty in measuring the very small proper motion of the distant galaxies.
Recent estimations found that the Magellanic Clouds, which were once thought to orbit around the Milky Way (MW) are now found to be on the first passage. Sohn et al. (2013) found that for most plausible MW masses, the observed velocity implies that Leo I is bound to the MW, which is based on the proper motion estimates from the HST. Sohn et al. (2012) obtained the first ever proper motion measurement for M31. The proper motion is consistent with a head-on collision orbit for M31 toward the MW (van der Marel & Guhathakurta 2008). The required accuracy to estimate the proper motion for the nearby galaxies is about 50 microarcsec/yr.
However, estimation of the proper motion of the galaxies near M31 is not possible with the existing facilities; these estimates are necessary to reveal the orbits of individual galaxies within the Local Group and its eventual destiny. Sohn et al. (2013) studied the mean orbital history of Leo I. Similar orbital history for all the galaxies in the Local Group is proposed. TMT with at least one order better resolution in astrometry (50 micro arc seconds accuracy) combined with deeper photometry can be used to estimate the proper motion of all the galaxies in the Local Volume. We expect to achieve orbital calculations of all members, which then will let us understand the kinematical history of interactions in this volume and predict the destiny of this group. These in combination with the proposed studies of chemical composition and star formation history will be able to decipher the formation, evolution and future of the Local Group.
Examples of PTF RR Lyr light curves. Top: A RR Lyr at a distance of 61 kpc whose field has 340 PTF images, with 312 detections (upper limits are indicated by small open circles). Bottom: A RR Lyr at a distance of 102 kpc with 149 PTF epochs, 61 detections and 88 upper limits (Source: Judy Cohen).
Inspite of several efforts we still do not know the Milky Way total mass to within a factor of two due to the lack of understanding of the dark matter dominated halo, the issue of the isotropy of the orbits and the shape of the potential (spherical or not) in the outer halo. The total mass of the Galaxy is usually determined through the motion of tracers (stars, dwarf galaxies, etc.) in the outer part of the galaxy beyond 50 kpc. These are combined with assumptions regarding the isotropy of orbits and the radial density distribution of the tracers to determine the total mass of the Milky Way.
In an effort to determine the total mass of the Milky Way, Deason et al. (2014) considered Blue Horizontal Branch (BHB) stars beyond 80 kpc as tracers of the outer halos. However this sample which was selected from the SDSS database (stripe 82 region) was found to contain mostly Blue Stragglers and a few QSOs, with distance uncertainties exceeding 30% in many cases. The sample contained only a few outer halo Galactic satellite dSph galaxies with reliable distances. This study reported that the number density of their outer halo stars falls very rapidly (ρ ~ r-6 at r > 50 kpc) and that the radial velocity dispersion of their sample is quite cold (σvr 50 – 60 km/s ) in the outer halo for (100 < r < 150 kpc); the estimated total mass for the Milky Way is rather small ~ 5 x 1011 M蕵.
Coehn and Sesar (priv. comm.) explored the outer halo of the Milky Way considering RR Lyrae variables found by the Palomar Transient Factory (PTF). They have selected 1257 RR Lyrae variables at distances beyond 50 kpc in the Milky Way halo, specifically to probe the density distribution as well as the total mass of our Galaxy. The distance of this sample could be determined to an accuracy of 5%, and the contamination by QSOs is expected to be minimal. From a detailed analysis they found a density law, ρ ~ r -3.4, between 50 and 85 kpc, which is close to the results found by Watkins et al. (2009) and Sesar et al. (2010) for the inner Milky Way halo.
For stars to be used as tracers of the outer halo, one needs to know their radial velocities and the proper motions. With TMT it will be possible not only to measure the radial velocities of PTF RR Lyr stars out to 100 kpc but also the radial velocities of the RR Lyr stars to be found with the Rubin Observatory out to a distance of 200 kpc and beyond leading to an improved understanding of the total mass of the Milky Way.
