ART - Advanced Radiative Transfer code

LTE RT code for solar ALMA applications (Formerly known as RTma.)

Installation

To run & compile, you will need C++ & Fortran compiler with MPI. Additional that you will need a parallel version of the HDF5 library. On MacOS (M1, or Intel) the easiest is to install all requirements via Home Brew. You will need the following packages:

shell brew install gcc openmpi hdf5-mpi

On Linux machine @ITA you most likely have all necessary compilers and libraries installed, and you just need to load the right modules (see compilation section).

Compilation

To compile, enter the src subfolder and type make MACHINE=xxx, where xxx is the file's name from the machine subfolder. These files are for machine-specific setup (compilers, flags etc.). You might want to create a new one for your own machine (host) or use an existing one. For example, if you have installed gcc & openmpi on Mac Os through Home Brew you can simply use:

shell make MACHINE=mac_gnu

Successful installation should generate RTma.x executable:

shell ... mpicxx -O2 -mtune=native -std=c++11 -I/opt/homebrew/include/ -g -c main.cc -o main.o mpicxx -o RTma.x cop.o eoswrap.o witt.o model.o io.o clte.o main.o -lstdc++ -lc++ -L/opt/homebrew/lib -lhdf5 -lhdf5_hl

Input files

To run ART you need an input atmosphere in *.h5 (HDF5) format. In the example folder you can find a small model with following structure:

```shell

h5ls model.h5

dens Dataset {1, 25, 15, 269} dx Dataset {15} dy Dataset {25} temperature Dataset {1, 25, 15, 269} xne Dataset {1, 25, 15, 269} z Dataset {1, 25, 15, 269} ```

Where:

* `dens` - is density in CGS units
* `dx`,`dy` - mesh spacing in the X-Y direction
* temperature - ..well obvious but make sure it is in [K]
* xne - electron density in CGS [ 1/cm^3 ]
* z - z-axis spacing in [cm]

There is also an example of a python script for generating input models from Bifrost snapshots in raw format (or official FITS) in utils.

Input file with frequencies

Additional to the input model atmosphere, you will also need an input file to define a list of frequencies for which you wish to calculate intensities. Example of such file (you can use it for template) can be also found in example folder.

```

well obvious ..input and out files

inputmodel = model.h5 outputprofiles = synthetic.h5

do not touch

linesfile = lines.vald.input rtsolver = 0 verbose = 1 mu = 1.0

EOS can be witt (default) or pisk (slower)

eos = witt

Cut the model from above until T = temperature_cut

temperature_cut = 50000.0

Compute contribution function (default = 0)

getcontributionfuntion = 0

Options (stderr, stdout, master, file, null)

logfile = master

Finally, a list of frequencies.. wavelengths in Angstrom units // first is ~1 mm, and second ~3.1 mm

region = 10000000., 1.0, 1, 1.0, none, none region = 31000000., 1.0, 1, 1.0, none, none region = 6300.000, 0.01, 16, 1.0, none, none region = 5000.000, 1.0, 1, 1.0, none, none ```

Running ART

Ok, we have the model and input file.. to run use mpirun

shell cd example mpirun -np 4 ./RTma.x -i input.cfg

expected output:

```shell

➜ mpirun -np 4 ./RTma.x -i input.cfg

Using input file: input.cfg info: readInput: inputmodel -> model.h5 info: readInput: outputprofiles -> synthetic.h5 info: readInput: linesfile -> lines.vald.input info: readInput: rtsolver -> 0 info: readInput: verbose -> 1 info: readInput: mu -> 1.0 info: readInput: eos -> witt info: readInput: temperaturecut -> 50000.0 info: readInput: getcontributionfuntion -> 0 log target 0x20952b978 info: readInput: region -> 10000000.,1.0,1,1.0,none,none info: readInput: region -> 31000000.,1.0,1,1.0,none,none info: readInput: region -> 6300.000,0.01,16,1.0,none,none info: readInput: region -> 5000.000,1.0,1,1.0,none,none log target 0x20952cda8 log target 0x20952cda8 log target 0x20952cda8 test log 0/4 info: io, readlines: read [3] lines info: initReadIO: Found dims [ nt=1, ny=25, nx=15, ndep=269 ] info: initReadIO: Found vars [ z ][ temperature ][ dens ][ xne ] info: processing 100% ```

Output file

After a successful run you should get an HDF5 under the name you set in the input file ( synthetic.h5 in the example ) with the following structure:

```

➜ h5ls synthetic.h5

Stokes_I Dataset {1, 25, 15, 19} Tau1 Dataset {1, 25, 15, 19} Wavelength Dataset {19} ```

Where:

*  `Stokes_I` - Calculated intensities, the first dimension is time (not used), then `x`,`y` and `wavelength`.
*  `Wavelength` - is the axis for 4th dimension (list of wavelengths in Angstroms) 
*  `Tau1` - is formation hight in `cm` i.e., value on `z-axis`  where `Tau_{wavelenght} ~= 1.0`.

These should be easy to plot via Python // see example folder again.