Authors: Jens Kammerer, Thomas Vandal, Katherine Thibault, Frantz Martinache

Supported instruments: NIRCam, NIRISS, MIRI

This package provides a pipeline to extract kernel phases from JWST data. The code aims to replicate the interface of the official JWST data reduction pipeline to provide a custom stage 3 kernel phase pipeline that outputs KPFITS files. The pipeline is based on the XARA¹ package and uses stage 2 calibrated ("cal" or "calints") products from the official JWST data reduction pipeline.

It is recommended to install the pipeline in a new virtual environment either with conda:

conda create -n <env-name> python
conda activate <env-name>

or venv:

python -m venv <env-name>
source <env-name>/bin/activate

To install the Kpi3Pipeline in the new virtual environment, you can either clone and install locally:

git clone https://github.com/kammerje/jwst-kpi
cd jwst-kpi/
python -m pip install .

or directly with pip from the GitHub repository:

python -m pip install git+https://github.com/kammerje/jwst-kpi

Examples of how to use the pipeline are available. There are examples for processing NIRCam, NIRISS, and MIRI cal and calints data.

A discrete representation of the pupil is required for the kernel phase extraction. Default pupil models for all supported instruments are provided with the package, but users can generate their own pupil models with XARA or XAOSIM and use them in the pipeline. We provide a script showing how the default pupil models were generated. It can be used and adapted to generate custom pupil models. If you find a pupil model that performs better than the default one, feel free to open an Issue or a PR!

The most recent major update brings new functionality and simplified usage:

• Both 2D cal and 3D calints data can now be processed.
• There is a new trim frames step. It is possible to specify the center and the size of the trimmed frames. With this new step, the pipeline is now running the steps in the following order: trim frames, fix bad pixels, recenter frames, window frames, extract kerphase, empirical uncertainties.
• It is now possible to provide a list of good frames and extract the kernel phase only from those.
• Improved pupil models and new pupil models for the NIRCam coronagraphy Lyot stops.
• Improved diagnostic plots.
• Improved file outputs.
• Simplification of the code by always transforming input data to 3D calints data.