This project is dedicated to perform analysis with ctools and gammalib to compute upper limits of annihilation cross-section and decay lifetime of dark matter particles using CTA observations.
The code presented here is functional at least for versions of ctools and gammalib greater than 1.6.3.
To install ctools and gammalib you can use conda. Also, you can build from source. Please check gammalib and ctools pages.
The core of the analysis tool is csdmatter. This class is based in csscripts installed with ctools. One of the key features is the management of dark-matter spectrum via the gammalib GModelSpectralTable class. This has the advantage to only use one fits file to describe the spectrum for candidates with masses in a determined range, dedicated annihilation channels (following the convention from the PPPC4DMID project) and particular energy ranges. You can check the jupyter notebook tablemodel under the notebooks/ folder to check how to create a spectrum fits file. The tablemodel notebook uses dmtable class to interpolate the spectrum for different masses, energy and channels using dmspectrum class. The spectrum is save into a fits table that is esaily ingested by ctools and gammalib.
The fits table has three parameters to describe the spectrum:
Aditionally, you can select whether or not to include electroweak (EW) corrections in the spectrum. By default, this parameter is set to True.
If you found any issue during test or use of this project, please open an Issue.
The contents of the project are:
You can set the GAMMALIB and CTOOLS environment variables as usual. See the documentation about gammalib and ctools.
ctadmtool is a python package to compute exclusion limits for model-independent dark matter searches with CTA. The package is an effort to have a common set of tools and use as basic example for analysis. There are two subpackages:
The relevan classes (about physics) are:
GModelSpectralTable models for annihilation or decay of dark matter particles.There are also some files in data and pfiles folders:
Finally, ctadmtool contains also the csdmatter app, based on cscripts.
The csdmatter app is based on how the cscripts are implemented within ctools. The csdmatter computes the upper-limits (at this moment, just) for annihilation cross-section for a famlily of mass points of dark matter particles. You can refer to pfiles/csdmatter.par and pfiles/csdmatter.txt to check the full list of input parameters, and the help of the app.
To have ctadmtool package availabe in your system you must to be sure that ctools and gammalib are loaded. Then to install ctadmtool you have two options:
Cloning:
$ git clone git@github.com:sergiohcdna/ctadmtool.git$ cd ctadmtool$ python -m pip install .Using pip directly:
$ python -m pip install git@github.com:sergiohcdna/ctadmtool.gitPlease note, that, if you want to contribute to the development of csdmatter and related classes, you must use the first option. Additionally, you can create a branch.
Note: There is Deprecation message when usin python -m pip install . referring that the local packages will be building in-place. You can take a look at the discussion. To avoid the deprecation message and test this new feature you must use:
$ python -m pip install . --use-feature=in-tree-buildNote: If you are updating a previous installation of ctadmtool, please be sure that you don't have any old csdmatter.par files in other locations. This will create a problem if new parameters were added to the version you are trying to install. A manual solution is to erase the csdmatter.par files from the $CTOOLS/syspfiles and your $HOME:
$ rm $CTOOLS/syspfiles/csdmatter.par
$ rm $HOME/pfiles/csdmatter.parAfter this, you can install ctadmtool.
Because the script is not part of the default cscripts, you must execute it using inside a python script.
To compute the ULs, csdmatter compute the ratio between the expected integrated flux and the upper-limit integrated flux obtained during the fit. Both fluxes are computed between minimum energy emin and 95% of the available energy in the process. The ratio is used to compute the exclusion limit using the reference annihilation cross-section or the decay lifetime used to compute the normalization of the dark matter spectrum. The ULs are computed for masses points (mnumpoints parameter) in the range of masses indicated by the user (via mmin and mmax parameters)
As already it was mentioned, the csdmatter app computes the upper-limits for a family of mass points (you can specify as many points as you need using the input parameter mnumpoints). These masses corresponds to differentes dark matter particles. They are separated logarithmically. For every mass value, the csdmatter app generate the corresponding GModel, and compute the upper-limit. Then, for every mass point, the following results are saved:
At the end, the results for all mass points are saved into a fits file.
If you use the run method, the results table can be accessed via the dmatter_fits method. For example:
from ctadmtool.csdmatter import csdmatter
thistool = csdmatter()
# After all the parameter initialization
thistools.run()
results = thistool.dmatter_fits()
print(results)You can take a look at the jupyter notebooks get_decayllimits and get_dmulimits to learn how to run the csdmatter tool to get exclusion limits for a Toy dark-matter halo.
There are several jupyter notebooks to show you how to use the package. The notebooks are:
dmspectrum ➜ Shows how to use thedmspectrum class to get dark matter spectrum for annihilation or decaytablemodel ➜ Shows how to use dmtable class to create GModelSpectralTable models for dark matterplottingdmSpectra➜ Shows how to plot spectra from table modelsdmflux ➜ Shows how to compute gamma-ray fluxes using dmspectrum class. comparison with Clumpy gamma-ray flux for decay.dmsimulation ➜ Simulate the a dark-matter signal using GModelSpectralTable modelsget_dmulimits ➜ Shows how to use csdmatter tool to compute exclusion limits for annihilation of dark matter particlesgetdecayllimits ➜ Shows how to use csdmatter tool to compute exclusion limits for decay of dark matter particlesPlease feel free to check the code and test. If you find any issue, bug, problem or you have a suggestion, open an issue.