My research interests are particularly in atomic processes that take place in the atmospheres (the outer, directly visible parts) of cool stars, and in applying this knowledge to interpretation of stellar spectra to gain better understanding of stars, stellar evolution, the history of star formation and the origin and evolution of the elements. Current instruments allow us to determine much from stellar spectra through interpretation of line strengths, shapes and asymmetries, provided we understand the underlying physics. Recently I also began working on atomic physics problems related to ionospheric physics.
Below I give an outline of the main areas of my research. If you are interested a more complete or detailed description then please see my publications.
In photospheres of cool stars such as the sun, the gas consists mostly of neutral hydrogen atoms. In fact they outnumber electrons by a factor of 104, even more in metal-poor stars (around 106). Understanding the collisional broadening of lines by H collisions is very important for interpreting stellar spectra, particularly strong lines. My research includes work on fundamental broadening theory, interatomic potentials, and calculation of line broadening data for astrophysical applications. For spectral lines of metals, H atoms are usually the dominant perturber in collisional broadening of spectral lines, since Stark broadening by electrons is not so strong. I have made available code for interpolating the data computed by our group using the so-called Anstee, Barklem & O'Mara (ABO) theory. I have also written a description on how to employ these data in calculations. This work has a natural connection to work with the Vienna Atomic Line Database, particularly updating the neutral hydrogen broadening data for astrophysically important lines. Data for nearly 50000 lines have been added to VALD.
In the case of spectral lines of H, due to the accidental degeneracy of the levels, Stark broadening by electrons and ions is usually the dominant collisional broadening mechanism. However, collisions with other neutral hydrogen atoms, a process called self broadening, cannot be ignored, and becomes quite important in metal-poor stars where there are fewer electrons and ions. I also do research on this process for Balmer lines, and codes for calculation of H lines, including both my work and the work of others, have been made available. It should be noted that theoretical understanding of strong lines is useless if we cannot reliably observe these lines. Determination of the continuum placement in the regions of strong lines, particularly Balmer lines, contains many difficulties. Together with Eric Stempels, I have developed techniques for reliable continuum rectification of broad lines from echelle spectra. These techniques and the broadening theory have been applied to a wide range of observations of A, F, G and K dwarfs and giants, from metal-rich to metal-poor objects.
Determinations of stellar abundances, particularly in old halo dwarf stars, are hampered by our lack of understanding of inelastic collision processes, data for which are needed in non-equilibrium (non-LTE) calculations. Once again the dominant perturbers are hydrogen atoms and electrons. Hydrogen atoms may be important because of their number. Electrons are important due to high collision frequency due to their speed and the fact that such collisions are expected to generally be efficient (non-adiabatic) at the speeds of interest. Andrey Belyaev (St. Petersburg) and I have performed calculations for Li+H collisions. We are currently performing calculations for Na+H. Recently I have begun working on a project to compute inelastic collision processes for electron collisions with Li and O. Calculations are being performed using the R-matrix approach.
The interpretation of the linearly polarised solar spectrum, observed at the limb, is a rapidly growing field of research in solar physics. One may for example measure weak magnetic fields in the solar atmosphere. In order to interpret the polarisation one must decouple collisional depolarisation mechanisms. This project, in collaboration with Moncef Derouich and Sylvie Sahal-Brechot in Paris, employs the ABO methods for computing interatomic potentials to compute data for depolarisation of metal lines by H collisions.
The nearest stars are of great interest. They tell us both about our place in the Galaxy, and also provide test beds for studies of stellar physics. I have been a part of the Spectroscopic Survey of Stars in the Solar Neighbourhood (S4N ) project led by Carlos Allende Prieto (Texas). We have produced spectral atlases for 118 nearby stars, and derived stellar parameters and chemical compositions of these stars.
The stars of the galactic halo are characterised by the fact that they are old and have relatively little amounts of heavy elements (metals). These stars are very interesting for a number of reasons. For example, they allow us to infer a lot about the origin and evolution of the elements, star formation and mixing in the early galaxy, and the physics of supernovae. I have worked a considerable amount on analysing the spectra of metal-poor stars, including applying my Balmer line theory and techniques to deriving stellar temperatures and luminosities. These methods have been applied to the most metal-poor stars known. I have also developed automated spectrum analysis software, based on the SME package by Jeff Valenti and Nik Piskunov, which we have applied to derive chemical compositions (abundances for 22 elements) in 253 metal-poor stars. I have also been involved in a study of chemical diffusion in globular cluster stars.