Skip to content

Latest commit

 

History

History
58 lines (42 loc) · 4.19 KB

README.md

File metadata and controls

58 lines (42 loc) · 4.19 KB

AstKAT

What is this?

This routine empirically determines the astrometric error for the time domain with radio interferometric data (Originally Designed for MeerKAT observation). See the files:

  • explanation.pdf — for a Chapter describing the routine
  • (COMING SOON) example.py — for an example of how to use the output to quantify astrometric errors/plot

Running it is easy: put all your images in the image directory and then cd into the scripts directory and enter ./run_all_scripts.sh. The output called observation_properties.json should have everything you need/

The Controlling Parameters are in the config.ini file

[TARGET]

  • name — Name of the source/target (used for naming of output files plots)

[POSITIONS]

  • target_ra — Right ascension of target: in astropy SkyCoord format "hms"
  • target_dec — Declination of target: in astropy SkyCoord format "dms"
  • phase_center_ra — Right ascension of phase/pointing centre: in astropy SkyCoord format "hms"
  • phase_center_dec — Declination of of phase/pointing centre: in astropy SkyCoord format "dms"

[PYBDSF]

  • snr_threshold — minimum signal-to-nouse (SNR) ratio for PyBDSF to identify the source (Default = 3.5). Make it ~0.5 less than the desired minimum source SNR
  • fix_to_beam — decided whether PyBDSF fits all of the Gaussian components to be the shape of the PSF (Default = True)
  • output_file_type — a type of output catalogue (Default='srl'), `source list' see https://pybdsf.readthedocs.io/en/latest/ for more information

[THRESHOLDS]

  • variability_threshold — maximum separation between the min/max flux before the source is rejected (Default = 2.0)
  • flux_threshold — maximum (fractional) deviation between the island and peak fluxes for a source to be considered a point source (Default = 0.1)
  • size_threshold — maximum (fractional) deviation between the source shape and PSF shape for a source to be considered a point source (Default = 0.1)
  • target_threshold — the maximum distance in arcseconds for PyBDSF source to be identified as the target (Default 20 arcseconds)
  • epoch_min — minimum absolute number of epochs that a source is detected for it to be included in the catalogue; i.e., if a source is missing from > epoch_min number of observations, it is excluded (Default = 100; NOTE: for a small number of epochs, the bootstrapping will fail for some sources * unless epoch_min is set to 0)
  • epoch_fraction — minimum relative number of epochs a source for a source to include in the catalogue; i.e., if a source is missing from > epoch_fraction * (total number of observations), it is excluded (Default = 0.25)
  • match_threshold — the maximum distance; this is expressed in units of fractional PSF; i.e., point sources are matched if they are within match_threshold * (FWHM BMAJ) (Default=0.33)
  • snr_threshold — minimum (median) signal to noise of a source (Default = 4.0)
  • ref_index — reference index to perform catalogue matching (Default = 0, the first eepoch)

[MCMC]

  • convergence_threshold — Convergence parameter for MCMC fitting, defined as the average difference between the current and last iteration in units of standard deviation (default = 0.1)
  • n_bootstrap — number of bootstrap iterations (Default = 5000; lower this if your computer is running out of memory)
  • phase_offset_limit — maximum distance from phase centre for a source to be used in MCMC fitting (Default = 0.3 -- The inner ~25% of the MeerKAT L-band beam)
  • n_iteration — number of concurrent good iterations before the routine stops (Default = 5)
  • min_snr — minimum signal to noise for a source to be used in MCMC fitting (Default = 0.0 -- No minimum)
  • max_snr — maximum signal to noise for a source to be used in MCMC fitting (Default=1e10 -- No Maximum)

Test suite

The test suite includes two extra scripts to simulate images:

  • ``Simulate_Images_Flat_Noise.py'' — this will make images and scatter the positions, enforcing the assumed astrometric error; this is just to make sure things work.
  • ``Simulate_Images_Flat_Noise.py'' — this will make images with (correlated) Gaussian noise, applying a user-defined $B$ offset (EXPERIMENTAL as of 2024 July 1)