STIX–MWA Data Analysis

Part of the solar-radio-imaging-spectroscopy sabbatical (FHNW / i4DS). MWA imaging itself lives in solar-burst-multiview.

This repository provides tools for analyzing and comparing data from STIX (Spectrometer/Telescope for Imaging X-rays aboard Solar Orbiter) and the MWA (Murchison Widefield Array).

---

Table of Contents

• Overview
• Environment Setup
• Repository Layout
• Data
• Paper
• Known Issues / TODO

---

Overview

This project integrates STIX and MWA observations to:

• identify overlapping solar flare events,
• download and preprocess MWA visibilities,
• generate dynamic spectra and light curves,
• compare flare timing and locations across instruments.

It also includes optional support for e-CALLISTO data and imaging routines using WSClean/CASA.

---

Environment Setup

This codebase supports Python 3.8–3.10 to ensure compatibility with recent versions of CASA.

Environment Setup

1. Conda Environment

Ensure you have Conda or a compatible tool installed. Then:

conda env create -f environment.yml -n stixmwa
conda activate stixmwa

2. Additional Dependencies

Some required libraries must be installed manually.

Manta-ray

git clone https://github.com/ICRAR/manta-ray-client.git
cd manta-ray-client
pip install -e .

Then replace the api.py in manta-ray-client/mantaray/api with api.py from src/helper_functions.

casacore

git clone https://github.com/aaijmers/wsclean.git
cd wsclean
mkdir build && cd build
cmake .. \
  -DCMAKE_INSTALL_PREFIX=$HOME/casacore-install \
  -DCMAKE_BUILD_TYPE=Release
make -j$(nproc)
make install

WSClean

# Navigate to the WSClean source directory (adjust path as needed)
cd ~/STIX-MWA/src/wsclean

# Patch CMakeLists: change C++20 to C++17
sed -i 's/set(CMAKE_CXX_STANDARD *20)/set(CMAKE_CXX_STANDARD 17)/' CMakeLists.txt
sed -i 's/set(CMAKE_CXX_STANDARD *20)/set(CMAKE_CXX_STANDARD 17)/' external/aocommon/CMakeLists.txt

# Clean and configure build
rm -rf build
mkdir build && cd build
cmake .. \
  -DWSCLEAN_WITH_CHGCENTRE=OFF \
  -DWSCLEAN_WITH_PYTHON=OFF \
  -DWSCLEAN_WITH_RADLER=OFF \
  -DBUILD_TESTING=OFF \
  -DCMAKE_CXX_FLAGS="-march=x86-64 -mtune=generic"
cmake --build . -j$(nproc)

After sucessful casacore instalation, install casatools and casatasks with pip:

pip install --index-url https://casa-pip.nrao.edu/repository/pypi-casa-release/simple casatools
pip install --index-url https://casa-pip.nrao.edu/repository/pypi-casa-release/simple casatasks

Then export the following paths (update to your install locations):

export PATH="$HOME/STIX-MWA/wsclean/build:$PATH"
export LD_LIBRARY_PATH="$HOME/casacore-install/lib:$LD_LIBRARY_PATH"
export PKG_CONFIG_PATH="$HOME/casacore-install/lib/pkgconfig:$PKG_CONFIG_PATH"

3. Environment Variables

A sample .env.example file is provided. Copy it to .env and fill in required values (e.g., API keys, data paths).

---

Repository Layout

• src/find_flares_in_mwa.py – queries STIX flare list and checks overlap with MWA obs times.
• src/mwa_stix_overlap.py – 2026 rework of the flare/MWA match: Sun-pointed obs from any project (not only G0002), STIX science flare list to 2026-02-28, Earth light-travel-time correction. Two steps: mwa (TAP query) and match.
• src/get_mwa_data.py – downloads raw data from MWA ASVO.
• src/compare_mwa_stix_locations.py – cross-comparison of event positions.
• src/plot.py – generates light curves and spectrogram plots.
• src/run_wsclean.py – runs WSClean imaging pipeline.

helper functions (src/helper_functions/)

• stix.py – STIX flare parsing and lightcurve handling.
• mwa_asvo.py – querying MWA ASVO API.
• spectrogram.py – building spectrograms from visibilities.
• ecallisto.py – optional integration of e-CALLISTO data.
• plot_flare.py – visualization helpers.
• utils.py – shared helpers.
• calibration.py, selfcal.py, mwa_imaging.py – calibration and imaging routines.

---

Data

All observations are already downloaded and are located on the /mnt/nas05/data02/predrag/data/mwa_data

---

Paper

Start of the paper containing introduction, detailed related-work and methodology is located in the _results folder.

---

Known Issues / TODO

• CASA compatibility can be version-sensitive.
• Spectrogram generation is slow for full observations → parallelization would help.
• Some helper modules (calibration.py, selfcal.py) are experimental.