RAMSES AMR file formats

Default

The default settings for the AMR data file formats is as follow:

from pymses.sources.ramses.output import *
RamsesOutput.amr_field_descrs_by_file = \
{   "2D": {"hydro" : [ Scalar("rho", 0), Vector("vel", [1, 2]), Scalar("P", 3) ],
            "grav"  : [ Vector("g", [0, 1]) ]
        },
    "3D": {"hydro" : [ Scalar("rho", 0), Vector("vel", [1, 2, 3]), Scalar("P", 4) ],
         "grav"  : [ Vector("g", [0, 1, 2]) ]
        }
}

which means that in the hydro_*.out* files :

• the first read variable corresponds to the scalar gas density field
• the next 3 read variables corresponds to the gas 3D velocity field
• the fifth read variable corresponds to the scalar gas pressure field

and in the grav_*.out* files :

• the 3 read variables corresponds to the 3D gravitational acceleration field

User-defined

If you use a nD (n not equal to 3) or a non-standard version of RAMSES, you might want to redefine this AMR file format to :

• make additional tracers available to your reader
• read nD (n not equal to 3) data

from pymses.sources.ramses.output import *
# 2D data format
RamsesOutput.amr_field_descrs_by_file = \
{   "2D": {"hydro" : [ Scalar("rho", 0), Vector("vel", [1, 2]), Scalar("P", 3) ],
            "grav"  : [ Vector("g", [0, 1]) ]
        }
}

# Read additional tracers : metallicity, HCO abundancy
RamsesOutput.amr_field_descrs_by_file = \
{   "3D": {"hydro" : [ Scalar("rho", 0), Vector("vel", [1, 2, 3]), Scalar("P", 4), Scalar("Z", 5), Scalar("HCO", 6) ],\
         "grav"  : [ Vector("g", [0, 1, 2]) ]
        }
}

To take into effect these settings, make sure you define them before any call to amr_source():

from pymses.sources.ramses.output import *
RamsesOutput.amr_field_descrs_by_file = \
{   "3D": {"hydro" : [ Scalar("rho", 0), Vector("vel", [1, 2, 3]), Scalar("P", 4), Scalar("Z", 5)],\
         "grav"  : [ Vector("g", [0, 1, 2]) ]
        }
}
ro = RamsesOutput("/data/metal_simu/run001", 20)
amr = ro.amr_source(["rho", "Z"])

User-defined output file

In order to have user defined variables updated automatically by PyMSES when an output is opened, you can create a “pymses_field_descrs.py” in your current directory with this structure :

from pymses.sources.ramses import output
self.amr_field_descrs_by_file = \
    {"2D": {"hydro" : [ output.Scalar("rho", 0), output.Vector("vel", [1, 2, 3]),
            output.Vector("Bl", [4,5,6]), output.Vector("Br", [7,8,9]),
            output.Scalar("P", 10),output.Scalar("Z", 11)],
                    "grav"  : [ output.Vector("g", [0, 1, 2]) ]
                    },
    "3D": {"hydro" : [ output.Scalar("rho", 0), output.Vector("vel", [1, 2, 3]),
            output.Vector("Bl", [4,5,6]), output.Vector("Br", [7,8,9]),
            output.Scalar("P", 10),output.Scalar("Z", 11)],
                    "grav"  : [ output.Vector("g", [0, 1, 2]) ]
                    }
    }
print "amr_field_descrs_by_file updated for MHD !"