wiki:DC3aProductionRunAnalysis
Last modified 10 years ago Last modified on 06/23/2009 02:40:04 PM

Analysis of Output from DC3a Production Runs

CFHT images are being been processed using the DC3a alert production pipelines. Information on the production runs is supplied on DC3a processing.

Scientists at IPAC and elsewhere are encouraged to analyze the output image data and to provide feedback to the applications group. See the list below for suggestions of features to search for and quantities to measure. This list is a work in progress, so do return to this page periodically. In addition to using this applications-oriented list, please also study the images from a science/SDQA point of view; for example, some image quality use cases are described on SDQA use cases. Note that Jeonghee Rho has posted an example of her analysis on analysis report.

Useful analysis activities suggested by Andy Becker:

  • Check that the variance is reasonably valued (e.g. not 1e8)
  • Check that edge mask bits are set
  • Measure residuals around bright stars (divide by the sqrt of the variance for significance in sigma)
  • Look for dipoles in the residuals, suggestive of Wcs mis-registration
  • Make sure cosmic rays are not in the image (more Isr than diffim)
    • Compare exposure 0 and exposure 1 (we put cosmics in by-hand in exposure 1)
    • Blinking the e0 and e1 images pre-Isr will reveal all the cosmics
    • Blinking the images post-Isr should show all of them masked and interpolated