IPP Progress Report for the week 2012.04.30 - 2012.05.04

(Up to IPP Progress Reports)

Eugene Magnier

  • mostly satisfied with relphot, but there seem to be some times when one or more of the remote jobs fail to apply the measured zero points back to the database. So far, re-running the zero point application has ben successful, but I'll need to address this.
  • I generated a number of images to visualize the statistics of the photometry quality. One set showed the difference between the mean magnitudes of stars and the synthetic magnitude for that star. I expected to see a pattern reflecting the dependence of the synthetic photometry on Galactic latitude. Instead, I found patches which had a grid pattern on the sky, with bands of ~6 degree height in DEC and a few degrees in width. This is the same as the 2MASS scan pattern, so I suspect this is showing systematic variations in the 2MASS photometry.

Serge Chastel

Heather Flewelling

  • psps
    • meeting with conrad about psps and czartool (overview of dxlayer, and I've been granted access to psps wikis and computers). I've also been reading the psps wiki
  • wikiwork
    • started a wiki on dvodbs. It's not very awesome yet but at least includes where the frequently asked about dvodbs live.
    • made a wiki on how to ssh to ipp machines.
    • writing up basic information for ipp processing (both for newcomers to ipp and for my aas poster)
  • megacam
    • major cleanup of megacam:
      • the raw exposures (%o) were set to state 'raw' so that they are not queued again.
      • the new exposures were ingested, and processed through warp using eam_branches/ipp-20110505 with the label u.MD__.20120502. I may move them to u.MD20110602 to make it simpler in the future for queueing. Both use the same revision/branch of ipp and the same recipes.
    • investigations of how to proceed (the quickest) to get nice stacks for JT:
      • plotting histograms of sigma_ra, if I exclude sigma_ra > 1, I exclude ~ 250 exposures out of 1900. These clearly have bad astrometry, and I need to investigate this. However, I'm tired of megacam.
      • turning off convolution - the configs related to convolution were never tweaked for megacam - some exposures are dropped from stacking. JT doesn't want convoluted stacks anyways so I'm going to turn that off for megacam.
  • jtrp
    • I queued up the MD nightly stacks that were processed in the previoud month as a V2 tesselation for John.
  • ippmonitor
    • cleaned up/ reorganized ippMonitor.
    • dvodb menu now works properly ( has addRun/minidvodb/mergedvodb, and summaries for dvodb processing which I hope to evolve into a czar like monitor)
    • we now have a czar menu with czar related tools (czartool pages, nigh summary, mask stats, mops/exposure status)
  • czar
    • nothing to note - bad weather so I queued up jtrp.
    • dvodbs are still building - low on space on ipp005

Roy Henderson

  • left Hawaii this week, so lost three days to prep and travel. Will make up time over coming month.
  • PSVO:
    • fixed bug preventing user working offline using Derby Db
    • added warning popup if requested PSPS catalog is not available
    • some refactoring for clarity
    • introduced config file:
      • class to encapsulate config, which parses file and provides getters
      • re-factored all code to use config
  • long explanatory email regarding PS1SC mailing lists management
  • some help with the switchover of mailing list stuff to ps1scn machine
  • updated PSPS news page with details of new SAS with 'better' objects from IPP

Mark Huber

  • jumping through hoops to get web services allowed to desktop for photpipe and other test data products with help from Gavin.
  • a little work with Peter in his S/N calculations with diffims for MOPS.
  • ppStack tests - no progress on cleaning up 7TB of test data on ipp059
  • ppSub tests - http://svn.pan-starrs.ifa.hawaii.edu/trac/ipp/wiki/ppSub_testing_201201
    • going around in circles a bit on these ad hoc test examples. re-runs of the diffims can produce different detections on the residuals.
    • setup test run of dophot run versus psphot on IPP diffim.

Bill Sweeney

Chris Waters

  • Added crosstalk masking for bright star saturation bleeds. These seem to follow the established rules, but crosstalk into all cells in the same OTA/cell row as the source star.
  • Finished work on video dark/mask switching. This appears to work as intended, applying the correct detrend based on the existence of video cells on the image.
  • Created and registered videomask detrends for all masks currently in use.
  • Continued work on reflection ghosts by studying a selection of images with multiple ghosts. There seem to be systematic offsets between the model position (and therefore the masks) and the true position of the ghost images. I'm still working on pinning a better model down.