Аннотация
Context. Ultraviolet radiation plays a crucial role in molecular clouds.
Radiation and matter are tightly coupled and their interplay influences the
physical and chemical properties of gas. In particular, modeling the radiation
propagation requires calculating column densities, which can be numerically
expensive in high-resolution multidimensional simulations. Aims. Developing
fast methods for estimating column densities is mandatory if we are interested
in the dynamical influence of the radiative transfer. In particular, we focus
on the effect of the UV screening on the dynamics and on the statistical
properties of molecular clouds. Methods. We have developed a tree-based method
for a fast estimate of column densities, implemented in the adaptive mesh
refinement code RAMSES. We performed numerical simulations using this method in
order to analyze the influence of the screening on the clump formation.
Results. We find that the accuracy for the extinction of the tree-based method
is better than 10%, while the relative error for the column density can be much
more. We describe the implementation of a method based on precalculating the
geometrical terms that noticeably reduces the calculation time. To study the
influence of the screening on the statistical properties of molecular clouds we
present the probability distribution function (PDF) of gas and the associated
temperature per density bin and the mass spectra for different density
thresholds. Conclusions. The tree-based method is fast and accurate enough to
be used during numerical simulations since no communication is needed between
CPUs when using a fully threaded tree. It is then suitable to parallel
computing. We show that the screening for far UV radiation mainly affects the
dense gas, thereby favoring low temperatures and affecting the fragmentation.
Пользователи данного ресурса
Пожалуйста,
войдите в систему, чтобы принять участие в дискуссии (добавить собственные рецензию, или комментарий)