IPP Progress Report for the week YYYY.MM.DD - YYYY.MM.DD

(Up to IPP Progress Reports)

Eugene Magnier

I've largely been working on comments from Doug F and Paul P on the photometric ladder paper. Doug pointed out the higher scatter in my figures compared to the ubercal version, for the same regions. I tracked this down to small number statistics (smaller areas used in my plots), failure to exclude the bad detections (bad flags), and failure to exclude enough of the galaxies (thus larger scatter at high |b|). I'm still not completely satisfied, and am grabbing samples of the LSDatabase from Eddie to do more direct comparisons before I submit the paper.

I've started to work on the very bright stars in the deep stacks. These have too many false positives in halos due to poor subtraction of the model flux. I have set up code to measure and subtract a radial profile (with circular symmetry). The result looks good at the first pass, but I need to be careful about edge cases.

Serge Chastel

  • Czar on Tue till Fri
  • Moving the monitoring activities to the new ippm account
  • MOPS WS diffs
  • ipp002 cleaning

Heather Flewelling

Mark Huber

  • QUB UK-liaison visit
    • general group discussions
    • full MD skycal copy brought over/setup until normal distribution ready
    • M31 data discussions
    • playing with the detectability server
    • examples of diffim stat cases passing cuts
    • discussion of full reprocessing of MD05 nightly stacks, alternative refstacks, MD10 refstack swap
    • basic ppStack, ppSub local test runs
  • U.Edinburgh UK-liaison visit
    • brief coffee talk, some general discussions with the groups interests
    • postage stamp server exercises for unconvolved stacks
    • alternative download script for limited data fetching and not mirroring, will focus on SAS2,MD09 ri stack products
    • investigating stack artifacts and corresponding input warps, added python code for pixel histograms

Bill Sweeney

  • Changed the algorithm psphotStack uses to choose objects to for forced photometry. Now it will be done only if the number of good detections for the object is >= 2 or if the only deteection was in y band (controllable by recipe)
  • Did some analysis of the source matching in psphotStack.
  • Implemented some changes to the detection efficiency measurements that avoids errors for inputs with a small number of pixels.
  • Ran a sample of 167 LAP skycells through the current version of psphotStack. All but 3 succeeded which is much better than we have ever done before with these data. Fixed a bug that allowed one of the others to succeed. The remaining two have psf matching problems that need to be investigated.
  • Modified psphotStack to remove forced photometry sources for which an aperture flux could not be measured. This happens if all of the pixels are masked. This should reduce the number of useless stack detections in PSPS significantly.
  • A user submitted some detectabiltiy query requests to the postage stamp server. They did not work. I introduced a trivial bug back in February that caused all detectability queries to fail. No one noticed. Fixed that and some of his requests ran into problems. He also reports that it is difficult to match up the results to each of his requests. We need to tweak the interface a little. I plan to work on that next week.

Chris Waters

  • Vacation
  • Visit to IU to present PanSTARRs status/information talk.