A comparison run with 128x128 gridcells in an inertial frame (Omega=0) with Rmin=0.4, Rmax=2.5 is presented. The radial boundaries (inner and outer) are non-permeable (closed box). Only mass accretion onto the planet is allowed. The initial setup in both codes was idential.
Three dimensional mountain plots are presented for both codes. In (Fig 1) the results of RH2D and in (Fig 2) of Nirvana are given. All features concerning the spiral wave structure in particular within the gap region are nearly identical.
The resulting total Mdot rate (which includes accretion from the inner and outer parts of the disc) is given in (Fig 3). (The reference model refers to the one stated in the paper). The Mdot rates are nearly identical. Thus, the agreement of the two codes for runs in the inertial and rotating frame is quite remarkable. Additionally, two results where the size of the accretion radius in the Roche lobe has been varied are given by the green and yellow curves.
Now, two comparison mountain plots in the corotating frame are shown. In (Fig 4) the results of RH2D and in (Fig 5) of Nirvana are given. Both codes use here an explicit viscosity. Also here the agreement with each other and the inertial frame calculations is very good.
In Fig.8 contour plots of the surface density for the three grids are plotted. The main grid is (Fig 8a), the first subgrid (Fig 8b), and the 2nd subgrid in (Fig 8c). The results are given at 180 orbits. The units are given in grid point number, and contour lines are superimposed. The resolution of the finest grid around the planet is such that there are about 11 angular and 30 radial gridcells within the Roche lobe of the protoplanet.
Results of a single grid calculation with the PPM method by Pawel Artymowicz which shows even more details near the protolanet is given in (Fig 9).
A little Movie shows how the initial phase of the gap opening evolves.
