Skip to content

Simulating high-resolution cross-correlation spectroscopy of exoplanet atmospheres.

License

Notifications You must be signed in to change notification settings

arjunsavel/scope

Repository files navigation

scope

Documentation Status License: MIT Code style: black pre-commit CodeQL Tests codecov Maintainability PyPI - Version

Simulating high-resolution cross-correlation spectroscopy for exoplanet atmospheres. Currently supports simulating IGRINS (on GEMINI-S) observations of transmission and emission spectra.

installation

You can install scope using pip:

pip install scope-astr

To install from source, run

python3 -m pip install -U pip
python3 -m pip install -U setuptools setuptools_scm pep517
git clone https://github.com/arjunsavel/scope
cd scope
python3 -m pip install -e .

You'll also need to download some data files. Currently, these data files are about 141 MB large. You can download them (to the correct directory, even!) with the following:

cd src/scope
chmod +x download_data.bash
./download_data.bash

This will create a data directory and plop the relevant files into it. You're also welcome to run the tests to make sure everything's been installed correctly:

pytest .

workflow

For more details, see the documentation.

Ideally, most user interaction with scope will simply occur through the input file (scope/input.txt). Any data field in a row marked with the [DB] flag can be pulled from a local database by inputting DATABASE. In our case, a database simply refers to a CSV containing contents from the Exoplanet Archive, with planet parameters resolved with the planet name.

Simply edit the input file to the desired parameters, then run:

python run_simulation.py

Running the script requires exoplanet, stellar, and telluric spectra. Default spectra and parameters currently correspond to the exoplanet WASP-77Ab.

Once completed, the code will create a directory for the data in output/ (based on the input modelname) with the following types of files:

  • simdata_: the simulated flux cube with PCA performed. That is, the principal components with the largest variance have been removed.
  • nopca_simdata_: the simulated flux cube, including all spectral components (exoplanet, star, blaze function, tellurics).
  • A_noplanet_: the simulated flux cube with the lowest-variance principal component(s) removed.
  • lls_: the log-likelihood surface for the simulated flux cube, as a Kp--Vsys map.
  • ccfs_: the cross-correlation function for the simulated flux cube, as a Kp--Vsys map.

attribution

If you use scope in your work, please cite our paper: https://arxiv.org/abs/2411.07303

About

Simulating high-resolution cross-correlation spectroscopy of exoplanet atmospheres.

Topics

Resources

License

Code of conduct

Stars

Watchers

Forks