Data Reduction Overview
The CAMPFIRE pipeline wraps the standard STScI JWST calibration pipeline with custom processing steps to better handle backgrounds, detector noise, and spectral extraction. The pipeline is fully automated, but all products undergo visual inspection — automated fitting provides a starting point, and human vetting of individual exposures and redshift solutions is essential for building a reliable catalog.
Pipeline Architecture#
The pipeline is organized into sequential stages for each instrument:
| Stage | NIRSpec | NIRCam |
|---|---|---|
| Stage 1 | Detector processing + background subtraction | Detector processing + artifact removal |
| Stage 2 | WCS assignment (2a), stuck shutter masking, nodded background subtraction (2b) | Flat fielding + sky subtraction |
| Stage 3 | Spectral extraction + 1D combination | Alignment + mosaicking |
| Redshift Fitting | Template-based chi-squared fitting | — |
Heritage#
The CAMPFIRE pipeline builds on the work of several teams:
- NIRSpec — Largely based on the CAPERS pipeline developed by A. Taylor and P. Arrabal-Haro
- NIRCam — Largely based on the CEERS and COSMOS-Web pipelines developed by M. Bagley and M. Franco, respectively
Instrument Pipelines#
- NIRSpec Pipeline — Multi-object spectroscopy: detector processing through 1D extraction and redshift fitting
- NIRCam Pipeline — Imaging: detector processing through mosaic assembly (documentation coming soon)