学术文献库

The standard method to generate dynamical models with a finite extent is to apply a truncation in binding energy to the distribution function. This approach has the disadvantages that one cannot choose the density to start with, that the important dynamical quantities cannot be calculated analytically, and that a fraction of the possible bound orbits are excluded a priori. We explore another route and start from a truncation in radius rather than a truncation in binding energy. We focus on the simplest truncated density profile, the uniform density sphere. We explore the most common inversion techniques to generate distribution functions for the uniform density sphere, corresponding to a large range of possible anisotropy profiles. We find that the uniform density sphere cannot be supported by the standard isotropic, constant anisotropy or Osipkov-Merritt models, as all these models are characterised by negative distribution functions. We generalise the Cuddeford inversion method to models with a tangential anisotropy and present a one-parameter family of dynamical models for the uniform density sphere. Each member of this family is characterised by an anisotropy profile that smoothly decreases from an arbitrary value $β_0\leqslant0$ at the centre to completely tangential at the outer radius. All models have a positive distribution function over the entire phase space, and a nonzero occupancy of all possible bound orbits. This shows that one can generate nontrivial self-consistent dynamical models based on preset density profile with a finite extent.