CASU meeting Wednesday May 25th 11:30-12:45 CASU meeting area. Present: MJI, AKY, JRL, STH, MR, SC, JCR Apologies: NAW, JPE, RGM, EGS Agenda: 1. Actions from last meeting 2. Comments on WFAU minutes 3. Hardware update 4. Meetings and telecons 5. Data archives update 6. Optical/NIR processing 7. WFCAM update 8. VISTA update 9. AOB Minutes ======= 1. Actions from the last meeting -------------------------------- AKY - is still comparing Henry's deep stack with our equivalent <<<< STH - has put 2MASS JHKs -> i-band predictor information on web pages AKY - has finished off the flat-field properties investigations and will now complete the report <<<< JRL - is still investigating using "rolling" master linearity corrections <<<< All - thought about propagating pawprint QC provenance info to tile headers as part of v1.2 upgrades. Probably best implemented at the grouting stage. JRL will create an example for further discussion <<<< JRL - fixed the 20101225 # of images faux pas SC - checked out the problems with the Gaia monitoring web pages SC - progress ongoing with Ubuntu/Debian upgrade for Blades <<<< JRL - has set up VISTA P87 directories and web page MJI - produced a new header keyword to record the number of available JRL 2MASS astrometric standards in the detector footprint -> NUMSTDS TBU as part of checks for sensible astrometric convergence EGS - check how often the WCS convergence problem seems to have occured for MJI VISTA and WFCAM by comparing # of matched stds etc.. for each detector Ongoing, but for WFCAM the alerts for GPS data being trialled are stdcrms > 0.2 numbrms < 0.25 "average" numbrms (viz max,min clipped "averages") stdcrms > 2.0 "average" stdcrms <<<< MJI - has finished off nebulising the select DR8 GPS data MJI - has written and sent off UKIRT Board report (as did STH) EGS - found out the VVV tile without "tmp" in the provenenance has a few friends. Circumstantial evidence suggests it only occurs when tiles are reprocessed outside of pipeline using CASUtools. EGS and MJI don't like the "tmp" aspect and suggest removing it for v1.2 upgrade and adding an extra keyword to the effect tile catalogue made from mosaiced nebulised pawprints MJI - has contacted VVV team to get examples of possible astrometric faux pas, but so faux no pas 2. Comments on WFAU Minutes --------------------------- They seem to be getting less frequent. 3. Hardware Update ------------------ Nothing has been bought / broken / fixed / improved etc... 4. Meetings and Telecons ------------------------ Upcoming: 31/05/11: The VMC Skypecon has been moved to Tuesday 31st May. 06/06/11: JRL will attend the "VISTA Survey Operations end-to-end Review" at ESO on June 6th. 08/06/11: The next VISTA IOT telecon will be at 14:00 on June 8th. It would be good to have the zpts done by then once we have the next data disk (JRL, AKY). <<<< 18/07/11: The Galactic Plane Survey meeting takes place at U. Herts. It runs for a week progressing from NIR to optical. 23-28/05/11: Cambridge beer festival, Jesus Green <<<< Downgoing: The ESO in-kind software contribution saga continues to rock'n roll. At least ESO+CASU have reached agreement on their part. However, ESO and STFC lawyers are now on the case -d'oh! After that the next (last?) hurdle is the ESO council meeting in early June. 5. Data archives update ----------------------- MJI (re)discovered a bunch of problems with the ING raw data archive of INT WFC observations in the period 1999 - 2000 i.e. pre-internet transfer days. This includes missing files, files mangled in various creative ways etc.... Que pasa ? MJI, JRL and EGS will convene a scrum when EGS gets back <<<< 6. Optical/NIR processing ------------------------- There have been many more requests for INT camera processing. The VST is acquiring some on-sky data some of which is being used for PR poipoises. It would be nice to get hold of some of the data to shakedown our VST pipeline. JRL will politely (sic!) attempt to achieve this either before during or after the meeting at ESO on the 6th. <<<< 7. WFCAM update --------------- May has been a relatively quiet month so far due to inclement (%$!*&) weather. Processing, checking and release of data is on track. A slight faux pas with fixes for the occasional missing header items for camera#2 was spotted via SJW's early access to processed data. We blame the fixed CD2_1 being only wrong by a factor of 1000, though oddly enough the data processed sin problema through the pipeline and somehow passed all the QC checks. This required MR to revisit all the affected nights (16th March to 18th April) and fix the fixes. While he was so happily engaged MJI took over the processing to keep his eye in. The intermittment headers took kindly to the camera#2 system being rebooted on the 18th April and haven't happened since (last checked up to May 26th). MJI noted that on the UKIRT web pages there is an item about an extension of UKIRT survey operations for another year viz 2012-2013. Mike Read spotted a good'un, the raw data for the 27th January 2011 had been processed twice and one copy dumped in the 30th instead of the correct data. MJI admitted guilt and can only conclude he accidently copied the raw data from the 27th to the 30th processing directory by mistake and carried on regardless. This is the Occam's Razor minimum cockup route theory. Mike checked the rest of the WFAU suppository but found nothing else quite like that. Assorted GPS reprocessing in readiness for GPS-DR8 is ongoing including: nebulising - done; further refining some frames with imperfect astrometry - partly done; and photometric recalibration for all? GPS frames - still TBD but will enable better zp solutions in high extinction regions and also better aperture corrections in these crowded fields. <<<< 8.VISTA update -------------- VISTA restarted taking data on 6th May and the first disk with data up to 12th May arrived recently. Unfortunately this only has a few flatfield and linearity sequences and certainly not enough to begin processing. The next disk shipment is expected over the next few days. When there is enough data we need to do a comparison between the zpts from the silver coated mirror -v- the aluminised mirror (and also check for any colour term changes). Interestingly although we have well-tabulated information for the previous silver coating there is virtually no information available on the web about the expected reflectance properties of aluminised mirrors as a function of wavelength. After a bit more rooting around, it turned out that UKIRT has aluminised mirrors and PCH had managed to find an ancient table of aluminium reflectance -v- wavelength for the 2006 paper (Hass G., Hadley L., 1963 from a handbook) Based on that we would expect the JHK NIR end to have similar reflectance to silver, Y to be 5% worse and Z 10% worse i.e. 0.1 mag difference in zpt for Y and 0.2 mag worse for Z. AKY has some experience with the properties of aluminised mirrors and it was noted that she'd be very popular if she put this info online. <<<< Prior to VISTA restarting EGS had updated the zpt trend to reveal an alarming downturn in the throughput since the beginning of this year casu.ast.cam.ac.uk/surveys-projects/vista/technical/photometric-properties EGS has received a 3-page gripefest from ESO about issues with the deliverables, although admittedly some of these are related to PI documentation. MJI, EGS and JRL will go through the relevant bits and see what, if anything, needs doing about this for future deliveries. Apparently ESO will "fix" any minor issues with the current lot of delivered files. <<<< JRL has been working on improvements to the VIDEO Skymaking Saga. Unfortunately he missed the Cannes deadline, but after feedback from the Jury about which version they prefer then all of the VIDEO data will need to be re-processed. JRL noted that Eckhardt had discovered a bunch of "broken" catalogues. This has happened before ...... this has happened before ...... this has happened before ...... and was fixed, or so it seemed. Action on JRL, AKY and EGS to FOWTF happened and contact relevant PIs if there are any important faux pas consequences. <<<< There was then a brief discussion of the possible v1.2 upgrades (listed below) plus timescale for implementing the more invasive of them bearing in mind the tradeoff between too many upgrades and hanging Chavs. Possible v1.2 upgrades ---------------------- minor tweaks to zpt estimation particuarly in regions of high extinction at some point re-run all the tile catalogues with the latest photometric calibration updates which also includes more accurate footprint selection and better rejection of saturated standards extend illumination correction tables by a column either end implement an acceptable solution to pawskymask viz-a-viz deep stacking (only affects VIDEO data so far but may impact PI programmes later) will require reprocessing all v1.1. VIDEO data record the potential no. of astrometric standards in each image footprint in the headers to aid in diagnosing possible astrometric solution faux pas (watch out for NUMSTDS ...... alternatives NUMNUTS ......) the tiles could inherit QC summary information from the pawprints to ease in diagnosing suspect data (still debating this one since in principle all the information is there via the provenance inheritance) put a description of the colour equations and a How To use the photometric information in the headers (cf. MegaCam) [NB. this may be confusing problem if ESO muck up the exptime keywords] remove the "tmp/" bit of provenance keywords and add keyword/value noting that the tile catalogue was generated from nebulised images with parameters used specified therein RADECSYS keywords - primary stays same i.e. FK5 since it relates to the telescope pointing. Should we add this keyword to the SHDUs and set it to have value ICRS ?? -> tests suggest (mostly) harmless 9. CASUHELPSs ------------- No corkers this month. 10. AOB ------- MJI still needs to fix the APM catalogue server -> recompile binaries <<<< Continuing Actions: ------------------- AKY - compare Henry's deep stack with our equivalent JRL - investigate benefits of a "rolling" master linearity correction AKY - finish off flat-field properties investigations, add a few more words of wisdom and make report available on web pages All - more cogitation viz inheriting pawprint QC info to tile headers SC - progress Ubuntu/Debian upgrade for Blades EGS - check how often the WCS convergence problem seems to have occured for MJI WFCAM & VISTA by comparing # of matched stds for each detector etc .... New Actions: ------------ JRL - assuming disks arrive in time generate zpt info before June 8th IOT AKY ALL - drink beer at Cambridge Beer Festival MJI - ancient INGARCH problems - que pasa ? JRL EGS JRL - sneakily attempt to get some VST data during 6th June meeting MJI - implement WFCAM astrometric fixes as revealed by GPS tests EGS - chew over the ESO gripes about the delivereds MJI JRL MJI - recompile the APM catalogue server binaries on something ...... JRL - get to bottom of VISTA broken catalogue saga and fix the bergers EGS AKY