Star formation
Inroduction
Star formation is one of the hottest areas in astrophysics. The big questions that we are trying to answer are:
Why stars always seem to form as binary or triple systems?
Why stars have the distribution of masses they do (the initial mass function)?
Why stars seem to form in clusters of hundreds or thousands of stars?
What are the initial conditions for planet formation?
What is the origin of brown dwarfs and free-floating planets?
How do star clusters evolve, and how does this effect the stars in the cluster?
How does feedback from the most massive stars effect the Galaxy as a whole?
Simulating star formation
Stars form in dense cores within giant molecular clouds. An increasing body of evidence is showing that star formation, rather than being a quasi-static process, is actually a highly dynamic process driven and strongly influenced by the highly turbulent dynamics of molecular clouds. The complex interactions and dynamics of star formation and star clusters mean that computer simulations are an essential tool.
I have written DRAGON - one of the most advanced star formation N-body smoothed particle hydrodynamics (SPH) codes. During star formation the hydrodynamics are of prime importance and SPH is very well suited to simulating the evolution of complex, non-symmetric, self-gravitating gas.
DRAGON has been used to produce the first statistically significant number of simulations of star formation within dense molecular cores. The chaotic nature of fragmentation in a turbulent environment requires a significant number of realisations before statistically meaningful conclusions can be drawn. Using observational data I have constructed cores with realistic masses, sizes, shapes and turbulence. I have found that moderate scale turbulent random velocity fields within cores very often form binary and multiple systems with a variety of mass ratios and orbital parameters very similar to observed properties. These cores also produce and eject significant numbers of brown dwarfs.
But the models don't match reality: current simulations tend to
produce too many stars in a core, and fail to predict the correct
binary properties. One of my main projects is examining the effects
of thermal physics and radiation transport on the collapse and
fragmentation of cores.
A column density plot of the central 1000-by-1000 au of a collapsing 5 solar
mass core which has fragmented into 5 protostars.
Simulating star cluster evolution
Stars do not form in isolation - they form in clusters of hundreds to millions of stars tightly packed together.
Once star clusters have finished forming stars the evolution becomes
collisionless and can be modeled using N-body simulations. the
evolution depends critically on the initial conditions - and I have
recently shown that substructure in young clusters can significantly
effect the evolution of young clusters. I am extending this work to
investigate the effect of initially fractal structure on the dynamics,
binary population and observations of young clusters. In addition I
am working on simulations of the extremely massive young cluster Wd1.
The Origin of the IMF
The origin of the stellar initial mass function (IMF) is one of the great unsolved problems in astrophysics. Observations of the mass functions of dense cores suggest that they have a similar form to the IMF. But, as most stars form in multiples, the mapping of cores-to-stars cannot be trivial. I am working on models of how core mass functions convolve to give stellar IMFs. These models seem to be sucessful in reproducing observed IMFs and also getting the correct binary fractions.
Here is a poster on the origin of the IMF from the core mass function from the IMF@50 meeting in Italy.
The Faulkes Cluster Survey
The Faulkes telescopes are two two-meter robotic telescopes (one in Hawaii, one in Australia) designated for combined educational and research projects. I am PI with Paul Roach (Cardiff) and Simon Clark (UCL) on the Faulkes Cluster Survey. This survey will provide at least 3-colour photometry of hundreds of previously unobserved open clusters over the whole sky. From the photometry one immediately obtains distances, ages, reddenings and HR diagrams for these clusters as well as the ability to identify massive young stars.
This database will provide a new and unique insight into the cluster luminosity/mass function and age distribution as well as a new probe of Galactic structure. (The only comparable survey uses only two colours and has a limited sky coverage.)
An short I-band image of the open cluster Dutra-Bica 58 taken with the
Faulkes Telescope. Not the most exciting image, but better images
are on the way...