-
Notifications
You must be signed in to change notification settings - Fork 0
Manual
The MuSCAT2 transit analysis pipeline consists of a set of Python scripts and classes that aim to make the analysis of MuSCAT2 photometry easy and painless. The pipeline covers the reduction of generic (non-transit) photometry, transit analysis, and more specific TESS follow up analysis. The pipeline is mainly aimed to be used from inside a Jupyter notebook, but it can also be used from inside a Python script.
- Visualisation of the raw data.
- Selection of the comparison stars and best target and comparison star apertures.
The pipeline can either aim to identify the best apertures and reference stars automatically, or it can be given a set of reference stars and apertures to use by the user. The reference star and aperture selection is done during a global optimisation step where the reference star apertures are free parameters in the light curve model.
The pipeline has one main high-level component, the TransitAnalysis classes. The class contains the methods for the comparison star selection, transit model fitting, MCMC sampling, etc.
The pipeline requires a set of Python packages that can be easily installed either using pip or
conda:
numpy, scipy, astropy, tqdm, traitlets, pandas, xarray, photutils, matplotlib, astroquery, corner,
seaborn, numba, uncertainties
The pipeline needs the latest version of emcee that is not available through conda or pip at the time of writing. It
is best installed directly from GitHub.
git clone https://github.com/hpparvi/PyTransit.git
cd PyTransit
python setup.py install
git clone https://github.com/hpparvi/ldtk.git
cd ldtk
python setup.py install
git clone https://github.com/hpparvi/MuSCAT2_transit_pipeline.git
cd MuSCAT2_transit_pipeline
python setup.py install
python setup.py develop
These steps are common to all analyses (TFOP, transit modelling, reduction of transitless light curves, etc.)
- Execute
m2init <target_name>to create an analysis directory<target_name>with the default directory structure. - Copy the photometry from each night to
<target_name>/photometry/<yyyymmdd>subdirectories. - Move into the analysis directory and execute
m2nbtemplate <target_name> <night>to create a template notebook. - Open the template notebook in Jupyter.
The template notebook begins with a cell initialising the main TransitAnalysis class
ta = TransitAnalysis(target, night, tid=TID, cids=CIDS)
where target is the target name, night is the observing night, tid is the target ID, and cids is a list of IDs
of reference stars to be included into the reference star optimisation (marked in the photometry reference frames).
The TransitAnalysis class has options to tailor the analysis, but this information should be enough for the basic use
cases.
- After the generic pre-analysis steps, copy one .fits file with its corresponding .wcs file from the MuSCAT2 NAS directory photometry_org to the photometry directory (this step will be removed in the future)
- Follow the template notebook
First, if we're not expecting to see a transit in the light curve, it is useful to let TransitAnalysis know that. This
can be done by setting the with_transit argument to False.
ta = TransitAnalysis(target, night, tid=TID, cids=CIDS, with_transit=False)
The pipeline can use PyTransit's OpenCL transit model for transit modelling, which can significantly accelerate the
analysis. This can be done by initialising TransitAnalysis with the with_opencl argument set to True.
ta = TransitAnalysis(target, night, tid=TID, cids=CIDS, with_opencl=True)
The light curves can be trimmed from the beginning and end by setting the mjd_start and mjd_end in the TransitAnalysis
initialisation. This may be necessary if a part of the light curve is corrupted, or has strong systematics due to large
airmass or similar.
The apreture ranges used in the optimisation can be constrained by the aperture_lims argument in TransitAnalysis
initialisation.