andyto1234 / demcmc-fip

demcmc-fip

This repository contains scripts for estimating the first ionisation potential bias (FIP bias) using demcmc to do the differential emission measure (DEM) calculations. The scripts here are designed to be run on Myriad, one of the high performance computing systems at University College London (UCL). It shouldn't be too hard to adapt them to work elsewhere.

Ingredients

To start with you'll need

• the original EIS observation as data/header h5 files. eispac has instructions one how to search for and download these.
• a numpy .npy file with fitted intensities and intensity errors across the whole EIS map.
• a .sav file with pre-computed contribution functions for each of the observed lines.

The lines in each of these files need to be a number of Fe lines for calculating the DEM, and the S and Si lines for estimating the FIP bias. Put these three files in the same folder, and make a note of that folder for later.

Recipe

Calculating DEMs

Make sure your ingredients are prepared as per the above section. Then run_dem.py runs the DEM calculation. Two things need to be configured at the top of this script:

• The number of parallel threads used to process the DEM data.
• The input and output data directories. Running this script will output progress to a log file in the output directory. Processing of the DEM for each pixel will happen independently and in parallel. When a whole row of pixels is processed the DEMs will be output to a netCDF file called dem_x.nc, where x is the number that gives the x-pixel of the row that contains DEMs for all y-pixels. Additional dem_x_lines.nc files will be produced, which give information on what lines were used to calculate the DEM. A '1' value means that line was used to calculate the DEM for that pixel.

Running on Myriad

The submit_demcmc_myriad.sh script provides a job script to submit the MCMC calculations to Myriad.

Calculating FIP biases

run_fip.py runs the FIP esitmation, using the DEMs calculated in the previous step. The only things that need configuring in this file are the various input/output file path locations. When you run this a progress bar should pop up in your terminal that shows you how long is left to process all the pixels. When the script finishes running it will output a file called fip_map.fits in the previously specified output directory.

Plotting the FIP map

plot_fip.py plots the generated FIP map, using some nice deafult colormap settings.