legend-bkg-scripts

Python scripts for looking at the LEGEND background data. Authors: Elisebetta Bosio, Sofia Calgaro, Toby Dixon

Currently there is:

1. build_data.py: Read the event tier files into histograms.
2. time-analysis.py: Look at (normalised) counting rates per run
3. plot-spectra.py : Plot the overlayed spectra
4. remove-OB.py : Script used to estimate the sensitivity to removing the OB

The build_data.py script can be run in LNGS in the current production enviroment as loaded with:

source "/data2/public/prodenv/setup.sh"

As:

dataprod-load-sw /data2/public/prodenv/prod-blind/ref-v1.0.0/config.json -- python build_data.py --output <OUTNAME> --p <LIST OF PERIODS> --proc <PROC EVENT> --recompute <RECOMPUTE> --target <TARGET KEY>

Where:

• <OUTNAME>: is the output file name (eg l200a-p10-r000-dataset-tmp-auto)
• <LIST OF PERIODS>: Is the list of periods to proces eg ['p10']
• <PROC EVENT>: is a boolean to get the data from evt files (True) or from parquet (if it exists - False)
• <RECOMPUTE> : is a boolean of whether to recompute QC flags based on a list of bad detectors
• <TARGET KEY> : is a string of type YMDTDHMZ, eg '20240317T141137Z', up to which data are included; later cycles are not loaded

The rest of the scripts do not depend on pygama and can be run in the LEGEND container on LNGS. Both have argsparsers and can be run with python <SCRIPT> -h to see the options: More details:

plot-spectra.py plots two spectra normalising the first one by exposure: It can be run fom inside the LEGEND container at LNGS (can also be run locally with the correct python packages) run:

python plot-spectra.py -o <OUTPUT PDF> -i <INPUT> -I <INPUT_P10> -e <ENERGY> -b <BINNING> -s <SPECTRUM>

Where:

• <OUTPUT PDF> is the name of the pdf file to save the plot (in the plots directory)
• <INPUT>: is the root file for the vancouver dataset for comparison (default /data1/users/tdixon/build_pdf/outputs/l200a-p34678-dataset-v1.0.root)
• <INPUT P10>: Is the root file for period 10
• <ENERGY> : is the energy range to plot (comma seperated eg 0,4000)
• <BINNING> is the binning
• <SPECTRUM> is the spectrum to plot, default mul_surv could also be mul_lar_surv etc (look in the root files)

time-analysis.py plots the rate of events in a certain energy window, it can also be run in the LNGS container:

python time-analysis.py -o <OUTPUT> -i <INPUT> -I <INPUT_P10> -e <ENERGY> -p <PLOT_HIST> -s <SPECTRUM> -B <BAT_OVERLAY>

Most of the options are the same as plot-spectra.py, the difference is:

• <OUTPUT> is the path to save both a ROOT file with the time-histogram and a PDF, it will be appended with the enegry range.
• <PLOT_HIST>: is a boolean to plot the data as a histogram not a graph
• <BAT_OVERLAY> is a flag to overlay a BAT fit to the data, the argument is the path to the directory contaning the BAT output.