This work is about models of Milky Way formation and evolution models in two ways. It continues and expands a program to map the Galactic halo structure using photometric data for 4,000 RR Lyrae stars from the Lincoln Near-Earth Asteroid Research (LINEAR) survey. This extends the distance limit of current maps and can place better constraints on halo structure. In addition, numerical simulations of galaxy formation are compared to the observational results and to further data sets such as the Sloan Digital Sky Survey (SDSS). This is a quantitative test whether cosmologically motivated state-of-the-art N-body models reproduce the observed spatial, kinematic, and metallicity distributions of the Milky Way stars.
The LINEAR photometric database will be made publicly available and will be a good resource for other researchers in the field. Some part of this project will be done by a doctoral student. A large number of undergraduate students will be involved in follow-up observations of LINEAR-detected RR Lyrae stars and in publication of the results; which gives them hands-on experience doing research.
This project consisted of two main logical parts: a comparison of modern computer models for galaxy formation and evolution with data from the contemporary digital imaging sky surveys,and a statistical analysis of variable stars discovered using the LINEAR sky survey. We describeeach part in more detail below. We tested and influenced theories of galaxy formation and evolution by comparing recent observational results on spatial, kinematic and metallicity distributions of the Milky Way stars with predictions by state-of-the-art galaxy models. The strength of these tests derives from recent significant observational advances based on modern large-area sky surveys (such as the Sloan Digital Sky Survey, SDSS) that characterized these distributions on scales from very near (a few hundred light years) to very far (a few hundred thousand light years, all the way to the presumed edge of the halo of our galaxy, the Milky Way) using unprecedentedly large samples of up to 100 million stars. Another enabling development for our program were rapid improvementsin hydrodynamic simulations of galaxy formation and evolution which are approaching sufficient fidelity to aid interpretation of these observations. Our main accomplishment is that we affirmatively answered a simple but far-reaching question: "Do state-of-the-art galaxy formation and evolution models produce spatial, kinematic and metallicity distributions in agreement with recent SDSS and other datasets?" These results are relevant not only to studies of the Milky Way, but may have transformative impact on studies of galaxy formation and evolution in general. For example, statistical comparison of large observational and model-based catalogs fostered novel applications of modern statistical methods to astronomical data analysis, such as automated search for overdensities and hidden correlations in high-dimensional spaces. We also extended observational constraints on the Milky Way outer halo structure by utilizing the largest sample of halo RR Lyrae ever assembled: 4,000 RR Lyrae stars discovered using LINEAR survey. RR Lyrae stars represent an excellent tool for studying halo structure because they are very luminousand thus can be seen to large distances, and their distances are easy to estimate. The LINEAR database, with several hundred observations per source, for about 20 million sources, is the best such resource currently available. The analysis of LINEAR RR Lyrae sample provided constraints on the outer halo structure that currently cannot be accomplished with any other dataset (for illustration see the figure). We performed follow up time-resolved photometric observations of a subset of these stars with a dual purpose: to provide more accurate characterization of selection efficiency and metallicity (a measureof chemical composition, related to star's age) accuracy, and to engage a large number of undergraduate students in publication-quality hands-on research. About a dozen undergraduate students were involved in all phases of this research, including the publication process. The science-ready LINEAR database, which we made publicly available, offers many ways of producing smaller and well-defined projects that are ideal for engaging undergraduate and graduate students in cutting-edge research, including both data mining and observational follow up (indeed, we expect that there are many variable stars still unrecognized in the LINEAR database). These stand-alone but discovery-based projects are excellent for introducing the scientific process to students without prior research experience.
