Spitzer Photometry Program

Estimate 5σ flux (uJy) sensitivity for Spitzer/IRAC observations of magnetars. The pipeline places synthetic PRFs, performs circular aperture photometry (IDL), aggregates intermediate measurements into .coldat, and fits SNR–flux to derive the flux at SNR=5 (Python).

Repository structure

• configs/ — example configuration tables (.coldat) for IDL procedures
• mag_info/ — object lists and helper tables
• prf/ — PRF libraries for IRAC channels
• simtar_partial/ — directory layout expected by the pipeline (mosaici1..4/)
• src/ — IDL and Python sources, plus a convenience runner script
• results/ — example outputs and intermediates
• plots/ — example figures produced by the Python fit

Requirements

• IDL (tested with IDL 8+), with supporting routines available at run time:
  • readcoldat, irac_limit, adxy, readfits, get_annulus (by request), circapphot (by request) must be on !PATH
• Python 3.7+
  • Install Python deps via src/requirements.txt (NumPy, SciPy, Pandas, Matplotlib)
• Download Cryogenic IRAC Core PRF images for each channel

Quick start: Python-only (compute SNR=5 from an existing coldat)

If you already have an intermediate photometry table (see format below), run:

cd src
./run_py.sh ../results/intermed/result_1e2259.coldat ../results/result_py.coldat

This will:

• create a virtualenv in src/.env (if missing),
• install requirements, and
• run flux_snr5.py -i <input> -o <output> --plot.

The output is a .coldat table with one row per source containing estimated ch1–ch4 fluxes at SNR=5 (NaN where unavailable). With --plot, diagnostic SNR–flux fits will display.

Full pipeline (IDL → Python)

1. Generate intermediate photometry with IDL

In IDL:

.r src/main.pro
run_irac_limit, 'configs/sample_config.coldat'

This reads the config, iterates over sources and channels, places PRFs, performs circular aperture photometry, and writes an intermediate .coldat specified by the config. The file has one line per placement with columns (see below).

2. Optional: Extract local exposure time (IDL)

In IDL:

.r src/texp.pro
extract_texp, 'configs/texp.coldat'

This scans mosaic FITS files in simtar_partial/<target>/mosaici<ch>/Coadd/ and writes per-channel exposure time around the target to the output coldat listed in the config.

3. Fit SNR–flux to get 5σ flux (Python)

cd src
./run_py.sh ../results/intermed/intermed_result.coldat ../results/result.coldat

Or directly:

cd src
python3 -m venv .env && source .env/bin/activate
pip install -r requirements.txt
python3 flux_snr5.py -i ../results/intermed/intermed_result.coldat -o ../results/result.coldat --plot
deactivate

Example output

File formats

• Intermediate photometry (.coldat) — produced by run_irac_limit and consumed by flux_snr5.py. Space-separated with header like:

  • # name ra dec ch x y factor phot sigma
  • Multiple rows per source and channel across PRF scaling factors.

• SNR=5 result (.coldat) — produced by Python, one row per source:

  • # name ch1_sens5(µJy) ch2_sens5(µJy) ch3_sens5(µJy) ch4_sens5(µJy)

• Exposure time (.coldat) — produced by extract_texp:

  • # name ch1_texp(s) ch2_texp(s) ch3_texp(s) ch4_texp(s)

Configuration (IDL)

configs/darkhan_config.coldat drives run_irac_limit. Fields include (by row index as used in code):

• [0] input_path — magnetar list (.coldat) with folder, RA, Dec
• [1] output_path — intermediate photometry output path
• [2] data_path — base path containing simtar_partial
• [3] factors — array literal like [1.,5.,10., 20., 30.] for total PRF
• [4] rap — aperture radius (pix)
• [5] rbackin, [6] rbackout — background annulus (pix)
• [7] spacing — PRF placement spacing (pix)
• [8] prf_path — PRF library root (e.g., prf/070131_prfs_for_apex_v080827_ch1)
• [9] result_path — final SNR=5 output path (used by spawn of Python step)
• [10] channels — array literal like [1,2,3,4] for IRAC channels
• [11] test_path — extra diagnostics output to store internal test results

configs/texp.coldat drives extract_texp (see src/texp.pro).

Notes on the Python fit (src/flux_snr5.py)

• Reads the intermediate table, groups by source and channel, computes SNR at each PRF factor and performs a linear regression SNR = a·Flux + b to solve for SNR=5.
• Use --plot to display diagnostics and an estimated per-point scatter.
• Defaults: sliding window size is 9; channel constants can be tuned in the script.

Data layout expected

simtar_partial/
  <source>/
    mosaici1/Coadd/*.fits
    mosaici2/Coadd/*.fits
    mosaici3/Coadd/*.fits
    mosaici4/Coadd/*.fits
prf/
  070131_prfs_for_apex_v080827_ch1/...
  070131_prfs_for_apex_v080827_ch2/...
  070131_prfs_for_apex_v080827_ch3/...
  070131_prfs_for_apex_v080827_ch4/...

Troubleshooting

• Ensure IDL helper routines (readcoldat, irac_limit, etc.) are on the IDL path.
• If Python plots do not display in headless environments, remove --plot or use a non-interactive Matplotlib backend.
• Verify paths in configs/*.coldat are absolute or correct relative paths.