HLRB Project h1341
The self-consistent formation and evolution of giant molecular clouds
ITA Heidelberg
Proposing Institution
ITA Heidelberg
Project Manager
Dr. Paul Clark
Albert-Ueberle-Str, 2
69120 Heidelberg
Abstract
We intend to perform the first numerical calculations to follow the formation and internal evolution of a giant molecular cloud (GMC) from the scale of the spiral shock out of which it forms down to the formation of the star-forming cores within it. This study will make use of our modified version of the smoothed particle hydrodynamics (SPH) code GADGET-2, which includes a `sink' particle implementation as well as a sophisticated chemical network designed to follow the chemistry of the major coolants in GMC gas, as well as importantobservational tracers such as HCO+. The purpose of this current application is to benchmark our modified version of GADGET-2 on the new Altix 4700. Once the scaling has been established, we intend to make a full proposal for CPU time at the LRZ.Our study combines the dynamic model of Bonnell et al. (2006), which models the formation of clouds in spiral shocks, with a greatly expanded version of the non-equilibrium chemical model of Glover & Mac Low (2007). This constitues a great improvement over current work in the field, which typically does not include both chemical evolution and high-resolution 3D hydrodynamics. The combination of these techniques allows us to study the formation of GMCs from warm atomic gas (at the 100 pc scale), along with the subsequent chemical and dynamical evolution of the cloud as it forms star-forming cores (with typical sizes of 0.1 to 0.01 pc). This project will address a number of poorly understood issues that are central to our understanding of both GMC and star formation. These include:*) What is the chemical structure of a GMC? How does the molecular fraction depend on the density of the gas?*) To what extent is CO an accurate tracer of molecular hydrogen? Does a simple conversion factor, such as the ubiquitously used `X'-factor, actually exist?*) What is the time-scale for the onset of star formation? How quickly are stellar clusters and associations formed?*) How does the turbulent driving in the global cloud formation picture differ from that typically used to study turbulently-supported star-forming clouds?The use of SPH is crucial here, as the Lagrangian nature of the scheme not only is a natural fit to the high dynamical range of the problem, but also allows one to track the histories of the individual fluid elements, and thus to determine how the star-forming regions are constructed. Additionally, SPH does not suffer from the problems associated with the self-consistent advection of multiple chemical species highlighted by Plewa & Mueller (1999) that plague most Eulerian grid-based schemes.
