How to run ProDiMo¶

TODO: Integrate that in other pages (e.g. Beginners)

TODO For the installation one should just run any fast model (make a dedicated one). Just to see if it runs, and also try some plotting script. All the other stuff here is also described somewhere else. Also the description of the outputfiles is likely outdated and also unnecessary. Maybe put some description for that into the Beginners section (e.g. looking at the log file(s))

1. Create a working (model) directory somewhere, say TWHya, and link the executable, if it is not in your PATH anyway (see OS environment).

2. Copy all *.in files from another folder of the ProDiMo source distribution (see examples). These files should at least include (more input files would be required for advanced models or you restart from a previous run but do not want to overwrite the finished run).

   Parameter.in      : parameter of the model
   Elements.in       : selection of elements
   Species.in        : selection of chemical species and initial abundances

3. Modify the input files according to your needs. Start simple by using a small grid resolution, a small disk mass, and by putting most of the switches to .false. in Parameter.in. In particular, use restart=.false., because a restart is not yet possible (see more in How_to_set_parameters).

4. run the executable in your working (model) directory. Before you do this please take a look at the OS Environment Variables. See the section on how to setup the parameters to accelerate_ProDiMo for running it in parallel.

   cd /path/to/models/TWHya
   prodimo > prodimo.log 2>&1

   Please always pipe the output of ProDiMo into a log file (i.e. prodimo.log). This is especially important if any problems occur and the developers need to take a lock. The 2>&1 at the end makes sure that both stdout and stderr are written in the same log file (is not required).

   To run in the background (i.e. you can exit the terminal and ProDiMo will run in the background)

   nohup prodimo > prodimo.log 2>&1 &

5. After completion, you should find at least the following output files in your working (model) directory (N.B. if older files with the same name are present, they are overwritten). The *.log and *.out files are human-readable. However the format is complicated and the user does not need to look at them.

   prodimo.log    : the logfile (if piped correctly see above). Take a look !
   ChemInit.log   : remarks about the chemical rate coefficients during initialisation
   Elements.out   : list of elements and abundances
   Reactions.out  : complete list of reactions merged from different sources
   Species.out    : list of species and masses taken into account
   dust_opac.out  : output of monochromatic dust opacity meant for the Ratran interface
   LineList.out   : output of cooling/heating lines considered
   ProDiMo.out    : human-readable output file containing all physical and chemical quantities
   pop_*_*.out    : output files containing calculated population numbers for line flux predictions   (if parameter write_pop=.true.)
   restart.*      : checkpoint file. You can use '''restart=.true.''' next time, which is faster
   Parameter.out : human-readable output of the parameters as used by the code

   Other output files (*.out) are listed below

   area.out
   BgSpectrum.out
   dust_sigmaa.out
   FlineEstimates.out
   FlineList.out
   initial_abund.out
   RTconv.out
   RTdata.out
   RTinterpolation.out
   SED.out
   SEDana.out
   StarSpectrum.out
   tau_vertical.out
   time_records.out

   Most of these files, except for ProDiMo.out, restart.fits.gz (restart.dat) and pop__.out, are available right after the start. If you are running models that iteratively solve the hydrostatic disk structure, which requires global iterations, these out files are renewed after every global iteration step. So you don't have to wait for visualisation until the global iteration has converged.

   There is no need to understand the content of each of the output files. A suite of python/IDL routines are provided to gather the output and make plots.

Now you can look at the results.