Skip to content

Pipeline overview

Jump to bottom
Tammo Jan Dijkema edited this page Jan 16, 2019 · 14 revisions

Introduction

This brief overview of the AperCal pipeline is based on the AperCal specification document (APERCal.pdf) in combination with the pipeline itself. The goal is to give a generic (but somewhat specific) insight into the various modules within AperCal and what they are doing. More specific/detailed documentation to come elsewhere.

1. Prepare (prepare.py)

Goal: Download data from ALTA and sort into folders
Input: Requested target + calibrator observation task IDs
Output: *.MS data products, in directory structure

  1. Check if observation directories exist; if not, create them.
  2. Check if observation data products exist; if not, download them.

2. Preflag (preflag.py)

Goal: Apply known flags and conduct automatic flagging
Input: *.MS data products, in directory structure
Output: *.MS data products (changed in place), *.flagversions separately (?)

  1. Flag data where antennas are shadowed (preflag.shadow)
  2. Flag subband edges, specifically channels 0-1 and 63-64 (preflag.edges)
  3. Flag ghost channels, specifically channels 16 and 48 (preflag.ghosts)
  4. Apply manual flags (preflag.manual), which can be any of:
  • Autocorrelations (preflag.manualflag_auto)
  • Named antennas, e.g. 'RT2,RTD'? (preflag.manualflag_antenna)
  • Correlations, e.g. 'XY,YX,YY' (preflag.manualflag_corr)
  • Baselines, e.g. 'ant1&ant2'? (preflag.manualflag_baseline)
  • Channels, e.g. '0~5;120~128' (preflag.manualflag_channel)
  • Time, e.g. '09:14:0~09:54:0' (preflag.manualflag_time)
  1. Apply AOflagger strategy to data-set (preflag.aoflagger):
  • Create bandpass solution and apply to data (preflag.aoflagger_bandpass)
  • AOflagging on data-set (preflag.aoflagger_flag)

3. Crosscal (ccal.py)

Goal: Calibrate the data using calibrator observation.
Input: Flagged MS for calibrators, Flagged MS for target.
Output: Calibrated MS (changed in place), calibration tables (*.Df, *.Kcross, *.Xf, *.Bscan, *.G0ph, *.G1ap, *.K)

4. Selfcal (scal.py)

Goal: Self-calibrate the visibilities.

5. Continuum (continuum.py)

Goal: Deconvolve and clean the data to produce continuum images.

6. Line (line_parallel.py)

Goal: Produce a continuum-subtracted line emission cube.

7. Polarisation (polarisation.py)

Goal: Generate Stokes Q,U,V continuum images.

8. Mosaic (mosaic.py)

Goal: Mosaic data from different beams into one large field.

9. Transfer (transfer.py)

Goal: Prepare and transfer final data products to ALTA.

10. Convert (convert.py)

Goal: Convert between CASA/Miriad/other formats.

Clone this wiki locally