next up previous
Next: Characteristics of warping Up: No Title Previous: Introduction

Disc model and the perturbation equation

We take a non-self-gravitating, axisymmetric disc as an unperturbed equilibrium disc. It rotates around a black hole of mass , of which the Schwarzschild radius . To simulate the weak general relativistic effect, we use the pseudo-Newtonian potential (Paczynski & Wiita 1980). We assume that the geometrically-thin maser disc is in hydrostatic equilibrium in the vertical direction, and has an inner radius, , and an outer radius, . We neglect radial advective motions and viscous effects. We use cylindrical coordinates ; the z-axis is normal to the disc midplane and is not the vertical axis of the galaxy.

We assume that the gas obeys a polytropic relation with index . The unperturbed density distribution is then given by

 

where is the unperturbed density in the equatorial plane, and H the half-thickness of the disc given by . In our model, we are free to specify a functional form of . For simplicity, we adopt a toy model for ,

 

where is the radius at which the density reaches its maximum [], and C a constant defined by . This density distribution was used by Miyama et al. (1994) to study the stability of circumstellar discs.

Since global features of the eigenmodes are not sensitive to the values of , C, and , we adopt a particular density distribution such that , , and . Here, and are, respectively, the inner and outer radii of the region from which the maser emission arises. For this particular choice of parameters, , , and .

We normalize H such that the maximum half-thickness of the region in which becomes (hereafter, we refer to this region as the maser emission region). Figure 1 shows the density distribution in the unperturbed disc by contours. In Fig. 1, the dashed line denotes the maser emission region.

  
Figure 1: Density distribution in the unperturbed disc. The dashed line denotes the maser emission region, in which . The contours are separated by 0.1

We consider a linear, adiabatic m=1 perturbation superposed on the unperturbed disc.

The perturbation varies as , where is the frequency of the perturbation. After some manipulations, we finally have the following equation describing the linear, adiabatic m=1 perturbation:

 

where is the enthalpy perturbation defined by , the Eulerian perturbation of the pressure, the sound speed in the unperturbed disc, and and the angular frequency of disc rotation and the epicyclic frequency, respectively.

As the boundary conditions for Eq. (3), we require that the Lagrangian perturbation to the pressure vanishes on the disc surface and in the equatorial plane. The latter condition is for warping modes. We solve Eq. (3) in the -plane with 101 grid points in the radial direction and 51 grid points in the vertical direction. To avoid singularities, we pose the inner and outer boundaries of the computational domain at and with , respectively. We have verified that the value of does not affect the global features of the eigenmodes as long as .



next up previous
Next: Characteristics of warping Up: No Title Previous: Introduction



Atsuo Okazaki
平成8年12月20日 (金), 午後 3時14分19秒