IPP Progress Report for the week 2012.07.02 - 2012.07.08

(Up to IPP Progress Reports)

Eugene Magnier

  • lots of discussions & investigations with Chris to understand the real impact of the CONST vs IMAGE weighting scheme on the false positives. there was some confusion over versions and recipe keywords which we finally worked out. final conclusion: the CONST weighting enhances the sensitivity to faint non-psf-like artifacts in the data. as a result, the false positive rate is higher even when boosting the variance. We decided to go with CONST weighting anyway and accept the impact on false positives to avoid the faint end bias.
  • pointed Doug & Eddie at MD09 and MD10 data processed with CONST to confirm the faint end bias fix.
  • Roger & Bertrand reported some corrupt files in the released DVO 3pi database (2012.06.06). I updated dvoverify to run in parallel, ran it on the database and found the problem files (and a few more). I repaired the problem files with the backup copy from 2012.05.09.
  • generated statistics and plots on the true and false positives as a function of filter and RA,DEC: PS1_3pi_DVO_20120606
  • worked on finishing the photometry reference ladder paper. sent to co-I's on Friday, to consortium on Jul 9.

Serge Chastel

  • Minor improvements to the databases backup scripts
  • ipp001 databases backup directory cleaning
  • Czar on Wed and Thu
  • Ganglia hack
  • Nebulous monitoring:
    • Each mhpcc storage node has now its own local database (a mysql server is running on each node). ippc11 now holds only a "summary" on those nodes.

Heather Flewelling

Mark Huber

  • czar monday+tuesday
  • ppStack tests continue, mostly finished debugging simtests and plots. when time, notes+plots to be added to the ongoing wiki http://svn.pan-starrs.ifa.hawaii.edu/trac/ipp/wiki/ppStack_testing_201111#TargetPSFforConvolvedStacks
  • MD10.refstack.20120705 selected chip-warp "test" reprocessing with new tag (ipp-20120626) and reftest samples made on stsci nodes after debugging setup. tighter input FWHM constraints from ppStack tests have helped reign in the convolved stack target PSFs but have highlighted a known problem of odd convolution that haven't had the chance to properly trace out in the ppStack tests. stalled until finished tracing out.
  • MD staticsky finished for all fields since fault is intermittent, revert solution for now... skycal problem found, needs to be tuned more appropriately for at least MD.refstack NRAW INST.MAG.RAW in own reduction config group.

Bill Sweeney

  • Returned from long leave
  • Investigated reports of very large numbers of false psphot detections around very saturated stars. Determined that the problem is triggered by the subtraction of models that do not match the sources well from the image before pass 2 source detection. The effect is lessened somewhat with the recipe that the staticsky analysis uses, but more work needs to be done.
  • Arranged for some cleaned up sts warps to be regenerated and distributed for MPG.
  • Worked with Gene to get debug build of IPP working on my Ubuntu desktop.

Chris Waters

  • Finished noisemap/boosting tests, and constructed final boosted noisemap.
  • Launched new SAS processing (label SAS_v7).
  • Launched new LAP processing (label LAP.ThreePi?.20120706).
  • Out one day due to illness, one day due to 4th of July.