nbody6tt

Description of the code

nbody6tt is based on nbody6 by Sverre Aarseth and includes the treatment of complex galactic tides in a direct N-body simulation of a star cluster, through the use of tidal tensors (tt). nbody6tt actually gathers two complementary methods:

Mode A: the tidal information is extracted from a galaxy or cosmology simulation, in the form of tidal tensors, along one orbit. This information is stored in a file. In a second step, nbody6tt runs a simulation of the cluster, reads the file, and computes the tidal acceleration. This method is described in Renaud, Gieles & Boily (2011).

Mode B: the user defines a numerical function which takes position and time as arguments, and returns the galactic potential. The space and time derivatives of the potential are used to (i) integrate the motion of the cluster on its orbit in the galaxy (starting from user-defined initial position and velocity vector), and (ii) compute the tidal acceleration on the stars. This method is described in Renaud & Gieles (2015).

The advantage of Mode A is the versatility: any kind of galaxy and orbit can be considered, even complex ones. For example, we used it to model star clusters in galaxy mergers (Renaud & Gieles 2013). However, Mode A relies on the tidal approximation (linearisation of the tidal force), and thus cannot be used to study tidal debris. Mode B is not limited by this, and does not require a galaxy simulation. (We used Mode B in Renaud & Gieles 2015)

Which mode should you use?

Do you plan to study tidal debris (tails)?
If yes, use mode B! If no, consider the next question.

Can you describe the galactic potential with a fortran function?
If yes, use mode B! If no, you should run a galaxy simulation, extract the tidal tensors and use Mode A.

What to do if you want to study tidal tails but cannot describe the potential with a fortran function?
Keep dreaming. I will wake you up in a few months when this will become possible! (work in progress)

Download

Download the sources from GitHub.

Reference should be made to Renaud, Gieles & Boily (2011) and/or Renaud & Gieles (2015), and a nice email is always appreciated!

Help, feedback

Questions, bug reports, comments and suggestions are most welcome: florent.renaud<at>gmail.com

Examples of tidal tensors

You want to try nbody6tt but you don't have a galaxy simulation? First, consider using it with MODE B (since v4.0, see above). Or you can use one of the tidal tensor files provided below. (I used them for the nbody6tt simulations presented in Renaud, Gieles & Boily 2011.) Download one of these files and rename it "tt.dat". In the input parameters, set KZ(14) to 9, and run nbody6tt!

tt.dat_pt_circ_r1000	Circular orbit at 1 kpc from a point-mass galaxy of 10¹⁰M_⊙ (named "Orbit A" in Renaud et al. 2011, Fig.5)
tt.dat_pt_circ_r3000	Circular orbit at 3 kpc from a point-mass galaxy of 10¹⁰M_⊙ (named "Orbit B" in Renaud et al. 2011, Fig.5)
tt.dat_pt_ellip_e0.5	Elliptical orbit of pericenter 1 kpc and apocenter 3 kpc (eccentricity = 0.5) around a point-mass galaxy of 10¹⁰M_⊙ (named "Orbit C" in Renaud et al. 2011, Fig.5)
tt.dat_plummer_extensive	Circular orbit at the caracteristic radius of Plummer potential (named "xi = 1" in Renaud et al. 2011, Fig.7)
tt.dat_plummer_compressive	Circular orbit at 0.66 caracteristic radius of Plummer potential (named "xi = 0.66" in Renaud et al. 2011, Fig.7)
tt.dat_antennae_orbitB	Orbit in the Antennae galaxies (see Fig.8 of Renaud et al. 2011, and orbit B in Fig.8 of Renaud et al. 2009)
tt.dat_ngc4038_orbitB	Same as before, but without the second galaxy colliding (see Fig.8 of Renaud et al. 2011)

Example of results

The role of galaxy mergers in the evolution of star clusters (Renaud & Gieles 2013). The orbits are those of Fig. 1.

Star cluster evolving in a time-dependent potential (white), vs. a static one (blue). Details in Renaud & Gieles (2015).