A capita selecta Masters course.
Chemical Evolution of Galaxies
Galaxies are factories that convert gas into stars, and in the process stars produce all the chemical elements in the Universe, other than those that were already created in the Big Bang (Hydrogen, Helium and a little Lithium). Stars are continuously being produced, starting in the early Universe and continuing until the present day, and in their death they enrich with chemical elements the interstellar gas from which the subsequent generations form. Linking this chemical build up to the formation and evolution of galaxies requires theories of cosmology, galactic dynamics, nuclear physics, stellar evolution and interstellar processes, and above all careful measurements of the abundances of chemical elements in stars. This is also crucial to understand why some stars have terrestrial planets. Stars can be used like cosmic fossils to answer a variety of topical questions in astrophysics, especially relating to the early Universe, and also with ongoing galaxy evolution processes up to the present day.
Stellar Abundance Measurements trace a variety of physical processes, both local and global. For example, the formation of terrestrial planets requires the build up and segregation of heavy elements, such as iron. A galaxy, like our own Milky Way, forms and evolves by continuously converting gas into stars. This process is repeated in all galaxies throughout time, often at very different rates, and it determines our entire perception of the Universe, as starlight is what makes galaxies visible. To determine the properties of dark matter and dark energy we have to assume a relation between starlight and galactic mass, and this is obviously dependent on an accurate understanding of star formation and evolution.
Stars are relatively well understood; they are luminous because they burn hydrogen into helium in their cores, and so produce heavier elements such as carbon, nitrogen, and oxygen. Stellar deaths in supernova explosions release these products and also create heavier and more exotic elements. The critical point is that low-mass stars, like our Sun, have lifetimes comparable to the age of the Universe. This means that many stars that formed in the early Universe will still be around today, and with the same chemical properties as the gas out of which they originally formed. Stars are fossils, time capsules, and their study is a sort of galactic palaeontology, with many applications and broad ranging implications that encompass my research goals.
A link to the early Universe, when the first stars and galaxies were forming, can be made through the small and ancient dwarf galaxies that surround our Milky Way. Dwarf galaxies do not undergo major transformations, except perhaps that they can loose their gas, however this appears to have little effect on their global properties. This means that most dwarf galaxies look now much as they did in the distant past, unlike the Milky Way. Dwarf galaxies are also relatively simple systems, compared to the Milky Way, and they typically form stars slowly and inefficiently, and so it is possible to follow their evolutionary history more easily star by star. In some particularly useful cases dwarf galaxies have only formed stars for a few billion years in the early Universe, and this lack of younger stars lying on top of the older populations affords an unobstructed view back to the earliest times. An important observational challenge is the hunt for the least chemically enriched stars, in our Milky Way and in nearby dwarf galaxies, to detect the starting point of star formation in galaxies.
Merging dwarf galaxies form the Milky Way is one of the fundamental assumptions of the currently favoured theory of galaxy formation and evolution, called hierarchical structure formation, or lambda-CDM. It predicts that smaller structures (dwarf galaxies) interact and merge together over time to form the large galaxies we see today, such as the Milky Way.
The detailed abundance patterns of the individual stars that make up the different components of the Milky Way (halo, thin/thick disk, bulge) all show evidence of coherent large-scale star formation that efficiently, uniformly and often very rapidly enriched them. In contrast, in small dwarf galaxies star formation is never an efficient or continuous process, it typically progresses in fits and starts. This is most likely because in small systems the energetic effects of star formation, and particularly supernovae, will tend to disrupt the gas so severely that only a low star formation rate is ever possible. This irregular and inefficient mode of star formation leaves its mark in how chemical elements build up over time. This can be traced by stellar abundance patterns, which are markedly different for each dwarf galaxy and for each component of the Milky Way. The oldest stars in dwarf galaxies are the least chemically enriched, and they do overlap with the oldest galactic halo stars. This overlap defines the time period during which dwarf galaxies could have contributed significantly to the build up of the Milky Way halo. This is during the first billion years of star formation, at redshifts, z > 5.
Thus at present it seems that large and small stellar systems with (galaxies) and without (globular clusters) dark matter all seem to form stars differently most of the time, resulting in distinct chemical abundance patterns. This suggests that the properties of large and small galaxies are not compatible with one being made out of the other, except at early times. This conclusion is controversial as it is inconsistent with the predictions of cosmological simulations, and so it needs to be extremely carefully examined. At the moment it rests on incomplete data sets, and so it is left open to widely different interpretations.
Making the connection to high-redshift observations is also an important step to fit nearby galaxies into a more complete understanding of the early Universe. Dwarf galaxies are currently the most common and most widely dispersed galaxies in the Universe, and according to cosmological predictions they were even more common in the distant past. They were most likely the primary drivers of reionisation at redshifts, z > 6. Connecting the observations of ancient stars in nearby dwarf galaxies with observations of the early Universe is necessarily indirect, as dwarf galaxies are typically much too faint to be observed directly at high-redshift. However, supernova-driven outflows from dwarf galaxies are likely to be one of the most efficient mechanisms to spread ionising radiation and chemical elements into their surroundings. The dramatic effect that a large population of dwarf galaxies will have should be detectable in the abundance patterns of chemical elements in the intergalactic medium.