Molecular Line Transfer Calculations in Star Forming Regions
Date: 2 July 2010
Thesis or dissertation
University of Exeter
PhD in Physics
This thesis describes the development, benchmarking and application of a non-LTE, co-moving frame Monte Carlo molecular line radiative transfer module for TORUS. Careful attention has been paid to the convergence, acceleration and optimisation of the code. I present the results of the application of the code to various benchmarking ...
This thesis describes the development, benchmarking and application of a non-LTE, co-moving frame Monte Carlo molecular line radiative transfer module for TORUS. Careful attention has been paid to the convergence, acceleration and optimisation of the code. I present the results of the application of the code to various benchmarking scenarios, including a collapsing cloud, a circumstellar disc and a very optically thick cloud of interstellar water. Benchmarking is an essential step in verifying the accuracy and efficiency of the code which is vital if it is to be used to analyse real data. In all cases, the code was able to accurately reproduce either the expected analytical solution or (in the absence of such a solution) was able to produce results commensurate with the results of other codes. In order to facilitate the motivating radiative transfer calculations of a star-forming cluster simulated using smoothed particle hydrodynamics (SPH) performed in this thesis, it was first necessary to devise and test an algorithm that efficiently maps an irregular distribution of smoothed particle hydrodynamics (SPH) particles onto a regular adaptive mesh. Whilst the algorithm was designed with this in mind it has also been used to study the effects of radiative feedback in circumstellar discs as well create a synthetic survey of a simulated galaxy. Bate et al.'s particle representation was resampled onto an adaptive mesh to enable me to use TORUS to obtain non-LTE level populations of multiple molecular species throughout the cluster and create velocity-resolved datacubes by calculating the emergent intensity using raytracing. I compared line profiles of cores traced by N2H+ (1-0) to probes of low density gas (13CO and C18O (1-0)) surrounding the cores along the line-of-sight. The relative differences of the line-centre velocities were found to be small compared to the velocity dispersion, matching recent observations. The conclusion is that one cannot reject competitive accretion as a viable theory of star formation based on observed velocity profiles.
Item views 0
Full item downloads 0