Mapping stationary axisymmetric phase-space distribution functions by orbit libraries
Mapping stationary axisymmetric phase-space distribution functions by orbit libraries
复制标题
DOI：
10.1111/j.1365-2966.2004.08072.x
复制
发表时间：
2004-06
影响因子：
4.8
通讯作者：
J. Thomas;R. Saglia;R. Bender;D. Thomas;K. Gebhardt;J. Magorrian;D. Muenchen;Mpe Garching;U. T. A. Austin;U. Oxford;U. Michigan
中科院分区：
文献类型：
--
作者：
J. Thomas;R. Saglia;R. Bender;D. Thomas;K. Gebhardt;J. Magorrian;D. Muenchen;Mpe Garching;U. T. A. Austin;U. Oxford;U. Michigan
这是一系列论文中的第一篇，致力于揭示后发星团中扁平的早期星系样本的质量组成和动力结构。我们描述了我们对Richstone等人的Schwarzschild码的修改。在截面表面应用Voronoi镶嵌，我们能够为单个轨道分配精确的相空间体积，并重建任何轴对称轨道库的完整三积分相空间分布函数（DF）。已经进行了两种类型的测试，以检查df可以用适当的轨道库表示的准确性。首先，通过将球形γ-模型和扁平Plummer模型的DF映射到库中，我们发现库的视距速度分布和内部速度矩与直接从DF导出的速度分布和内部速度矩相匹配，其精度优于当前的观测误差。其次，通过将库拟合到相同DF的投影运动学，我们表明，从拟合库重建的DF与输入DF在覆盖库质量90%的相空间区域内匹配到约15%的均方根。所达到的精度使我们能够实现有效的基于熵的正则化来拟合真实的、有噪声的和空间不完整的数据。
This is the first of a series of papers dedicated to unveiling the mass composition and dynamical structure of a sample of flattened early-type galaxies in the Coma cluster. We describe our modifications to the Schwarzschild code of Richstone et al. Applying a Voronoi tessellation in the surface of section, we are able to assign accurate phase-space volumes to individual orbits and to reconstruct the full three-integral phase-space distribution function (DF) of any axisymmetric orbit library. Two types of tests have been performed to check the accuracy with which DFs can be represented by appropriate orbit libraries. First, by mapping DFs of spherical γ-models and flattened Plummer models onto the library, we show that the resulting line-of-sight velocity distributions and internal velocity moments of the library match those derived directly from the DF to a precision better than that of present-day observational errors. Secondly, by fitting libraries to the projected kinematics of the same DFs, we show that the DF reconstructed from the fitted library matches the input DF to a rms of about 15 per cent over a region in phase space covering 90 per cent of the mass of the library. The accuracy achieved allows us to implement effective entropy-based regularization to fit real, noisy and spatially incomplete data.