L-Galaxies Dust Analysis

This repo holds the data analysis pipeline used to obtain all the plots in Clay17 from the detailed dust implemenation of L-Galaxies available here: https://github.com/scottclay/Lgalaxies_Dust

To use this pipeline, you should first go to https://github.com/scottclay/Lgalaxies_Dust and follow the instructions for running the model.

Running the pipeline

A step by step series of instructions to run this pipeline is given below.

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

Sorting the output data

If you have run the model on the full millennium box, you will have an output directory with a very large amount of files. The first thing you should do is run the tools in /prepare_output/tools/ to sort these files into their respective snapshot directories.

./rehome.sh snaplist_file model_output_dir where_you_want_the_files

• rehome.sh and the snaplist_file for MR/MRII Planck is available in /prepare_output/tools/
• model_output_dir is the location of the L-Galaxies Dust output files from running the model at https://github.com/scottclay/Lgalaxies_Dust
• where_you_want_the_files is the directory you want to store the files

Pickling the data (Python)

You should now pickle the data.

• cd Lgalaxies_Analysis/prepare_output/
• edit the file paths in script_read_allz_apollo.py to match the where_you_want_the_files from above and for where you want the pickled data to go
• run run_script.sh

run_script.sh sends script_read_allz_apollo.py to apollo and pickles the data for redshifts z=0-15.

NOTES:

• You can change what information is in the pickled output in script_read_allz_apollo.py
• You can see all available information from the L-Galaxies dust output in the snap_template located at /data/snap_template.py
• This pipeline assumes the data is stored in the following format:

Clay2018 (this can be named anything)
└───MR
│   │   Pickled
│   
└───MRII
    │   Pickled

Plotting the graphs

You should first edit the file path in the get_data_dir function in /src/bin_data.py to match that of the top output data directory (in the case above, that would be Clay2018 (don't include MR or MRII or pickled in this directory as the pipeline can be used to call MR or MRII or both itself).

• Each plot has its own script within the /dust_analysis/ subdirectory
• For example - running python SM_DM.py will output the SM_DM.eps to /dust_analysis/figs/ and the binned data to /dust_analysis/binned_data/
• Alternatively, running ./produce_all.sh will create all plots

As many of these plots require reading in data at 9 redshifts, you may prefer to send these plotting scripts to apollo. You can find submission scripts for all the plots in /dust_analysis/apollo_runscripts/, or alternatively to submit all plots to apollo qsub run_all.sh