This is a set of programs for calculating the autocorrelation of Lyman-α forests from SDSS BOSS data.
Note that deployment is not very user friendly.
- Some tinkering with configuration files and scripts may be required.
- 500GB of storage space
- 32GB of RAM (preferrably more)
- 100GB of fast local storage (RAM or SSD)
- 16 CPU cores or more (recommended)
- Python 2, or Python 3
- Cython
- matplotlib
- lmfit
- h5py
- healpy
- astropy
- scipy
- numpy
- mpi4py
To build and install the cython module run:
cd cython_helper
python setup.py install --user
- SDSS BOSS DR12 spPlate fits files, with the original directory structure
- QSO table from a CasJobs query, saved as a FITS file, for example:
SELECT S.specObjID,S.mjd,S.plate,S.fiberID,S.ra,S.dec, S.zOffset,S.z,S.zErr,S.zWarning,S.class,S.subClass,P.extinction_g INTO mydb.QSOs FROM SpecObj AS S JOIN SpecPhotoAll AS P ON S.specObjID=P.specObjID WHERE S.instrument='BOSS' AND (S.zwarning | 0x10) = 0x10 AND S.class='QSO' AND (S.boss_target1 & 0x1FF) = 0 AND (S.z>2.1) AND (S.z<3.5) AND (S.plate>3523)
- PCA continuum fit tables from Suzuki et al. 2005 and Paris et al. 2011.
- Broard Absorption Line catalog
DR12Q_BAL.fits
- DLA catalog from Garnett et al. 2016
There is an annotated configuration file at example_config/.
The default configuration file is lyacorr.rc
at the current directory.
The configuration file path can be overridden using the LYACORR_CONF_FILE
environment variable, e.g.:
export LYACORR_CONF_FILE=/path/to/config_file.rc
There is an annotated script file at example_config/.
The output files are in native numpy format (.npy
,.npz
).
They can be viewed using matplotlib in a Jupyter notebook.