Astronomy and Astrophysics · 2010 · 234 citations · 18 references
Natural SciencesChemistryPhysicsComputational ChemistryMolecular PhysicsRadiation TransportComputational AstrophysicsMolecular CloudsAtomic Excitation SpectraDelaunay Triangulation AlgorithmsExcitationPhysical ModelsRadiation Field ModelingRadiative Transfer ApproximationRadiative Transfer In Molecular CloudsRadiative Transfer Modeling
We present a new code that solves the molecular and atomic excitation and radiation transfer problem in molecular gas to predict emergent spectra. The code operates in arbitrary 3‑D geometry using unstructured Delaunay lattices, samples input models weighted by density to create an opacity‑scaled grid, iteratively converges level populations and radiation fields, then ray‑traces the grid to produce images, and can handle overlapping lines of multiple species. The code achieves efficient photon transport, making slow convergence in opaque models traceable, and its high dynamic range makes it particularly suitable for modeling ALMA data.
We present a new code for solving the molecular and atomic excitation and radiation transfer problem in a molecular gas and predicting emergent spectra. This code works in arbitrary three dimensional geometry using unstructured Delaunay latices for the transport of photons. Various physical models can be used as input, ranging from analytical descriptions over tabulated models to SPH simulations. To generate the Delaunay grid we sample the input model randomly, but weigh the sample probability with the molecular density and other parameters, and thereby we obtain an average grid point separation that scales with the local opacity. Our code does photon very efficiently so that the slow convergence of opaque models becomes traceable. When convergence between the level populations, the radiation field, and the point separation has been obtained, the grid is ray-traced to produced images that can readily be compared to observations. Because of the high dynamic range in scales that can be resolved using this type of grid, our code is particularly well suited for modeling of ALMA data. Our code can furthermore deal with overlapping lines of multiple molecular and atomic species.
18
Least squares quantization in PCM
Sheelagh Lloyd · IEEE Transactions on Information Theory · 1982 · 15.6K citations · Full text
Self-similar collapse of isothermal spheres and star formation
Frank H. Shu · The Astrophysical Journal · 1977 · 1.7K citations
Natural Sciences, Astronomy, Physics +16