Documentation/UserGuide/Grid Based Output

Grid-binned particle output is essentially the same as that for gas quantities. One example is given as below

<output1>
  out_fmt	= ppm		/* ppm image */
  out		= M1par	 	/* particle x1-momentum */
  id		= M1par
  x2		= 1.0
  dt		= 1.0		/* time increment between outputs */

Here the variable M1par denotes the particle x1-momentum. All built-in quantities for the particle output include dpar, M1par, M2par, M3par, V1par, V2par, V3par, which denote particle density, momentum and velocity respectively. The item pargrid is automatically set to 1 for these built-in output variables. By default, particle property selection function is set to all, but the user is free to define their own property selection functions.

The built-in variables can be retrieved in the grid from pG->Coup[k][j][i], where pG is the pointer to the grid, and Coup is of type GPCouple (see /athena/src/athena.h). It is mainly used for storing the intermediate step gas quantities for the particle integrators, but is also used in storing the grid-binned particle quantities for output purposes. Here its elements include grid_d (particle density), grid_v1, grid_v2, grid_v3 (particle momentum). The user can define their own particle-related output variables based on these grid-binned quantities in the same way as for gas variables (see User-defined Output Variables), but in such cases pargrid in the output block must be set to 1 explicitly.

For binary/vtk/tab data dumps, they contain binned particle variables dpar, M1par, M2par and M3par with property selection function all, which can NOT be changed.