L-Galaxies Dust

This repo holds the implementation of the detailed dust model into the public released version of the Henriques2015 L-Galaxies model. This adds a model of dust production from AGB stars, supernovae as well as grain growth in molecular clouds. It also includes a model of dust destruction from supernovae shocks. Documentation of the public version of the code can be found at http://galformod.mpa-garching.mpg.de/public/LGalaxies/

Changes to Henriques 2015 you might expect

• Detailed dust modelling. Each galaxy now outputs the dust mass as a function of constituent elements (i.e. C,O,Si etc.). You should sum these for the total dust mass.
• The dust model is called in main.c (update_dust_mass) immediately AFTER the call to the chemical enrichment model (update_yields_and_return_mass)
• Changes to various models, transfer functions, header files etc. to account for the dust.
• New file: dustyields_read_tables.c - Similar to yields_read_tables but for dust. Run before L-Galaxies. Reads in the mass and metallicity dependent dust yield tables.
• New file: dustyields_integrals.c - Similar to yields_integrals.c but for dust. Run before L-Galaxies. Pre-calculates the normalised ejecta rates at every timestep, assuming 1 Msun populations.
• New file: model_dustyields.c - Holds the new detailed dust modelling code

Changes to Henriques 2015 you might not expect

• If dust is switched on, the call to SN_feedback that was at the end of model_yields is now called AFTER the call to the dust model in main.c.
• My_Makefile_options has a new option, REDUCED_OUTPUT, which is switched on by default. The dust model requires SFHs to be switched on, but the SFH output when combined with the chemical enrichment model and the full dust model is very large. In order to make it easier to play with, as well as to take up less hard drive space, REDUCED_OUTPUT cuts the SFH from the output.

Installing and Running the model

A step by step series of instructions to run this version of L-Galaxies on your system.

• git clone https://github.com/scottclay/Lgalaxies_Dust.git

You will probably need to change the filepaths or obtain treefiles/coolfunctions etc. from elsewhere (check out http://galformod.mpa-garching.mpg.de/public/LGalaxies/running_the_model.php for general instructions on running the Henriques 2015 version of L-Galaxies, as well as where to find the tree files and Stellar Population Synthesis tables):

• Edit the following in the input file:

  • FirstFile
  • LastFile
  • CoolFunctionsDir
  • SpecPhotDir
  • SimulationDir
  • OutputDir

• The following compiler options have been added to My_Makefile_options

  • OPT += -DDETAILED_DUST - Switch on dust model with default options
  • OPT += -DDUST_AGB - Switch on AGB dust production (default on)
  • OPT += -DDUST_SNII - Switch on SNII dust production (default on)
  • OPT += -DDUST_SNIA - Switch on SNIA dust production (default on)
  • OPT += -DDUST_GROWTH - Switch on grain growth dust production (default on)
  • OPT += -DDUST_DESTRUCTION - Switch on dust destruction (default on)
  • OPT += -DFULL_DUST_RATES - Output dust production/destruction rates (default on)
  • OPT += -DREDUCED_OUTPUT - Shorten output (mainly cutting SFH bins)
  • OBJS += ./code/model_dustyields.o
  • OBJS += ./code/dustyields_integrals.o
  • OBJS += ./code/dustyields_read_tables.o

To compile the model make

To run the model locally on box 5 of MR ./L-Galaxies ./input/input_MR_W1_PLANCK_local.par

To run the model locally on box 40 of MRII ./L-Galaxies ./input/input_MRII_W1_PLANCK_local.par

To run the model on apollo for all MR at Sussex qsub batch_MR_apollo.sh

To run the model on apollo for box 1 and 2 of MRII at Sussex qsub -l m_mem_free=220G batch_MRII_apollo1.sh

To run the model on apollo for box 3-512 of MRII at Sussex qsub batch_MRII_apollo2.sh