In projects funded by the NSF and NASA, I am studying how of a number of dwarf galaxies, which are satellites of our own Milky Way, have evolved. My collaborators and I have been using the 4-meter and 1.5-meter telescopes at the Cerro Tololo Inter-American Observatory in Chile to image in the Carina dwarf spheroidal galaxy (dSph), Fornax dSph and the Large Magellanic Cloud. We have additional imaging of fields in the Large Magellanic Cloud scheduled to be made with the Hubble Space Telescope this fall. We are determining the luminosities and colors for more than 30,000 low mass stars in each galaxy. By comparing the observed color-magnitude diagrams to Monte-Carlo simulations of theoretical diagrams, we deduce the star formation histories. However, we also need to know the chemical enrichment history since color-magnitude diagrams are degenerate in age and chemical abundance (as well as interstellar reddening and distance, too). Initially, we have had to use the diagrams themselves to judge what chemical abundances to adopt. But in the fall, we return to Chile to use the 4-meter telescope and its multi-fiber spectrograph to obtain spectra of 125 stars in the Carina dSph. With this spectrograph, each of 24 fiber-optic cables are placed on a different star and 24 spectra are obtained simultaneously. With colleagues in Australia, I will use the Anglo-Australian Telescope's new 300 fiber spectrograph to obtain spectra of hundreds of stars in the Fornax dSph. By measuring the depth of calcium absorption lines in the stellar spectra, we will determine the distribution of chemical abundances in the galaxies directly.
My initial analyses of the color-magnitude diagrams of the Carina and Fornax dSph have produced some exciting results. Previously, a dSph galaxy was thought to have such a low mass (10^7 to 10^9 solar masses including the dark matter halo) that it could only have formed stars in a brief episode early in the history of the Universe before the explosion of hundreds of supernovae swept the gas out of the galaxy. To put things in perspective, supernovae generate shock waves that travel at greater than 1000 km/sec whereas the stellar velocity dispersions of dSphs are only 10 to 20 km/sec. Indeed we find no hint of gas in today's dSphs. However, our new color-magnitude diagrams show that dSphs had surprisingly complex star formation histories. Only in the last few Gyr (giga, or billion years) have they lost enough gas to curtail ongoing star formation. The Carina dSph had distinct episodes of star formation approximately 2 through 7 Gyrs, and 11 through 14 Gyr ago. The Fornax dSph had a more constant star formation rate that decreased to very low levels about 2 Gyr ago, although stars as young as 500 Myrs old are seen in Fornax. Star formation in these dSphs lasted much longer than the theoretical dynamical or cooling timescales of approximately 0.1 Gyr. Therefore, star formation must be self-regulated, even in these low mass galaxies, by a delicate balance between cloud cooling and ongoing star formation, and the ensuing photoionization by massive stars and violent eruptions of supernoave. The small spread in chemical abundances in the Carina dSph proves that chemically-enriched gas was expelled in galactic winds. However, the more massive Fornax dSph shows a wide range of chemical abundances. Clearly these galaxies, which were once thought so simple, are teaching us valuable lessons about star formation on a global scale. Why the Carina dSph remained quiescent for as long as 4 Gyr is a mystery. This winter, I will collaborate with colleagues at UC Santa Cruz to explore photoionization/hydrodynamical models to see if the meta-galactic flux of UV photons could have kept the gas in Carina ionized during this time.
Studying the dSph companions of our Milky Way also gives us valuable constraints on the evolution of our own Galaxy. Astrophysicists continue to debate whether large galaxies form primarily in a monolithic collapse of a single gas cloud, or whether their formation is more hierarchical in nature where star formation first occurs in dwarf-galaxy sized fragments that later merge to form larger galaxies. The latter is suggested by many cosmologies (Cold Dark Matter being one well-known example). The discovery of the Sagittarius (Sgr) dSph, which has gotten too close to the Galaxy and is being ripped apart by gravitational tides, has renewed interest in whether mergers were important in the Galaxy's history. Sgr dSph stars stretch over more than 20 degrees of the sky, and eventually these stars will disperse and be incorporated into the Galaxy. I am exploiting the tremendous light-gathering power of the Keck I Telescope to study Sgr dSph stars. By obtaining high dispersion, high signal-to-noise ratio spectra, my colleagues and I are measuring the abundances of many different chemical elements in Sgr dSph stars. Analysis of our first star shows surprisingly high chemical abundances. The star is a metal-rich as the Sun. After analyzing our data on seven stars taken this summer, we can use the specific patterns of elemental abundances to determine more about the evolution of the Sgr dSph and whether or not the Galactic halo formed from the merger of similar satellite galaxies.