parallax-jwst 1.5.0

A Python library for exploring publicly available JWST data from the MAST archive.

Parallax

Parallax is a Python library for exploring publicly available JWST data from the MAST archive. It finds astronomical sources that existing catalogs have not yet characterized, cross-references them against SIMBAD, NED, and Gaia, and presents the results in terms an educated non-specialist can act on.

Installation

Requires Python 3.12 or later.

git clone https://github.com/teaguejg/parallax-jwst
cd parallax-jwst
pip install -e .

This installs all dependencies. PyQt6 is included for the desktop GUI.

Quick start

import parallax as par

# run the full pipeline on M92
report = par.survey.reduce("M92", instrument="NIRCAM")

print(f"Detected: {report.n_sources_detected}")
print(f"Unverified: {report.n_unverified}")

# launch the GUI
from parallax.gui import launch
launch()

reduce() downloads JWST images from MAST, runs source detection, queries a configurable set of catalogs, and writes JSON and markdown reports to data/reports/.

Validated targets

These targets have been tested end-to-end:

• M92 (NGC 6341) - globular cluster. ~23,800 sources detected, ~18,700 unverified candidates. Runs in under two minutes.
• NGC 3132 - Southern Ring Nebula. Use "NGC 3132" as the target string; "Southern Ring Nebula" does not resolve on MAST.
• Orion Bar - star-forming region. MAST has no Level 3 mosaic products for this target, only detector-level i2d files.

Configuration

All settings live in config.yaml in the project root. Key knobs:

• detection.snr_threshold (default 3.0) - minimum signal-to-noise for a detection. Lower finds fainter sources but more noise.
• detection.min_pixels (default 5) - minimum connected pixel area for a source to be reported. Raise to suppress noise detections.
• detection.background_box_size (default 50) - background estimation tile size. Smaller values (20-25) work better near bright extended emission.
• resolver.search_radius_arcsec (default 2.0) - catalog cross-match radius. Increase if WCS alignment is imprecise.

Environment variables override config values using the prefix PARALLAX_ with underscores between segments (e.g. PARALLAX_DETECTION_SNR_THRESHOLD=5.0).

Requirements

• astropy >= 5.3
• astroquery >= 0.4.8
• photutils >= 1.9
• matplotlib >= 3.7
• numpy >= 1.24
• PyYAML >= 6.0
• scipy >= 1.11
• PyQt6 >= 6.4