CASU meeting Wednesday 25th July 11:30-13:00 APM Present: MJI, JRL, AKY, SC, JRF, RGM, NAW, JPE, STH Apologies: EGS, MR Agenda: 1. Actions from last meeting 2. Comments on WFAU minutes 3. Hardware update 4. Meetings and telecons 5. Data archives update 6. WFCAM update 7. VISTA update 8. VST update 9. Spectroscopic stuff 10. ESO in-kind stuff 11. AOB/news 1. Actions from last meeting ----------------------------- STH - some progress with VISTA calibration papers, particularly relevant given the circulating draft of the VVV calibration-to-2MASS paper, on the other hand still need to circulate a draft and include the Y- and Z-band Vega zero-point offsets for discussion <<<< JPE - characterising the "radial" extent of diffraction spikes for VISTA JRF pawprints also being worked on by STH + PhD student from the pov of STH variability aspects <<<< EGS - finally managed to rationalise serving up of UVEX data after the MJI U-band recalibration exercise SC - the new switch is being installed even as we type, monitoring web page foibles to follow <<<< MJI - is still keeping an eye out for UCAC4 <<<< but this probably got superceded by the availability of PPMXL SC - contacted Network support who could find nowt wrong with UKLight at the Cambridge end, well ...... other than the fact the connection to WFAU was running like a drain (see later) JRF - was added to the VIRCAM email exploder and is an avid bedtime reader MR - WFCAM data for May was DB-ingested and subsequently released JRL - checked if spikier-than-usual VISTA darks survived the transfer to here but nada visible in QC checks STH - reported on progress with VISTA Y- and Z-band zero-point offsets JPE - reported VIKING Phase3 deliverables were happening JRL - processed an extant night with the "averaged" linearity curve and MJI has it in his in-tray <<<< JRL - acquired APASS v6 and MJI slightly reformatted it and used for June VST illumination correction MJI - updated bandmerging prog for VPHAS data ongoing <<<< 2. WFAU minutes --------------- After a bit of buffing somewhere in Edinburgh transfer speed issues with UKLight vaporised and its now running faster than ever. 3. Hardware update ------------------- An almost action-free month: one of the new G-bit switches finally arrived and is being appropriately worshipped; apm37 was moved from the smaller -> larger computer room to re-balance the heat load; during the UKLight kerfuffle CUDN casually let drop that in about a year's time we would have to get our own GBN circuit for when the UKLight VLAN service gets the chop. As this may coincide with a bit of fibre re-routing through the new AP building all may follow a smooth trajectory, or not. 4. Meetings and telecons ------------------------- There was a VMC skypecon 28th June at which JPE and EGS attended but nothing exciting was recalled. The next skypecon is tomorrow but as that is after today's extended leaving party for JRF the odds of anyone, apart from JPE, calling in are minimal. There was a 4MOST telecon on 28th June at which NAW, MJI and EGS badgered the PO, with some success we think, to simplify and streamline the survey simulation exercise. 5. Data archives update - AAT, ING, WFCAM, UKIRT, VISTA ------------------------ All up to date, much UVEX and Kepler processing occurring and MJI had a very pleasant rain-free observing week for the latter project on the INT. 6. WFCAM update ---------------- Data processing and releases up to date. June release awaits July completion due to small number of nights in June usable for illumination correction computations. There was a slight communication faux pas between the OT and CASU regarding the allowed naming of recipes for sky flipping options e.g. JITTER_FLIP12 1,2 being the detector numbers for alternating target/sky observations had been successfully used in the past. However, the previously satisfied FeII customer managed to find a skyflip recipe name in the OT the pipeline was unaware of, JITTER_FLIPNS and proceeded to use it, with predictable results. After a brief flurry of emails between JAC and CASU this was quickly sorted, the NS is now 32 and all is well. JRL coerced the pipeline to reprocess the (many) affected nights in the correct manner and we think happiness and customer are now reunited. ESO contacted MR regarding raw UKIDSS data deliveries .... finit ? non ? and were informed we will continue to deliver UKIDSS patches data, UDS, calibration files etc. providing the project ID has UKIDSS or CAL in the name. MJI reported that UKIRT consortium promisory contributions from assorted UK University departments were proceeding apace, even including the IoA. Phil Lucas contacted casuhelp about ~10mas systematic differences in astrometry reported by CASUtools software and the WSA listings. After some investigation a list of potential causes of such small differences were drawn up and investigated including all aspects of numerical precision, ZPN inversion differences and so on, but no clear conclusion were drawn. Systematics at this level are well below the likely accuracy of the WCS calibration, so further investigations were put on hold awaiting inspiration from left-field. Steve Warren was worried about intermittent pickup visible during DR10 checks. This manifests at a low level (~+/- 1-2 ADU) and shows 4 quadrant symmetry, particularly apparent on some detectors e.g. #2 in particular and also #4 rather than others. Similar pickup has been seen before so its not clear if this is a low level issue with its own game plan or a more recent phenomenum. More worryingly Steve also reported on a problem with sky subtraction around large galaxies noted in LAS DR10 checks. After a few checks it rapidly became apparent that this was caused by the observing strategy adopted for patching holes due to previous bad data. Rather than observing complete tiles, single or double pawprints followed by jumps to completely different spatial locations seems to have been the norm. Unsurprisingly the pipeline reverted to using the pawskymask method on this patched data which does indeed cause problems around large objects given the small number of dithers used. JRL and MJI spent a while trying to figure out a post-processing fix for this but rapidly came to the conclusion that this was not going to be practical. The sky estimation strategy for WFCAM has been very successful and evolved through many iterations, so global changes at this late stage are probably not a good idea. Science-wise the only solution is to reprocess ALL patched LAS (and any other) observations like this using a specifically tailored sky estimation strategy. Tests on one night (20120104) where this problem happened with montonous regularity yield much improved results. The downside is that there is an unknown, as yet, number of nights where this reprocessing needs to be done. 7. VISTA update ---------------- Processing and data releases are proceeding well after the sticking recalcitrant March 23rd and 30th data, among several others, were booted out of the door. Data to end of May (bar one other sticky night 20120527 awaiting EGS's return) have been released to PIs and June is ready for its photometry updates prior to grouting. Eckhard reported back that one catalogue, v20120229_00110_st_cat.fits, mysteriously had lots of NaNs and Infs in extension 14, equally mysterious was they all vanished when it was regenerated - d'er! There was some discussion about the Terrapix downloads of vast quantities of VIKING data. Conclusion, they don't have enough to do and passwords have a habit of being misused. Mario Soto had recently sent round a draft VVV photometric calibration ->to->2MASS paper that had excited much discussion (via email) among the parties involved. Just goes to prove that getting our VISTA calibration paper(s) out as soon as possible is good news. Phase3 deliveries to ESO for VVV and VHS are proceeding apace. VMC is also in the queue. A pity the changes to the FITS keywords demanded by ESO are going to cause much confusion further downstream. RGM had raised some interesting aspects of astrometric comparisons with 2MASS of assorted tiles in recent emails. These revealed systematic, mainly radial, patterns of +/-100mas in some but not all comparisions. There was some suggestion of changes in the patterns mid-2010. These changes may correspond to VISTA interventions when assorted shimming of the focal plane was carried out. It was certainly the case that in the first year or so of operations the astrometric distortions of the focal plane may have changed but should have settled down after the last major intervention Spring 2011. It was also the case that sometime after one of these interventions the ZPN distortion coefficients were updated (20101201->) to better reflect the updated geometry. MJI is investigating and will report back at the next <<<< meeting. Meanwhile, as noted previously anyway, for accurate astrometry, particularly proper motion work, use the stacked pawprints. 8. VST update -------------- VST data receipt, ingest and processing all proceeding smoothly. Data up to end of June has been released to sundry interested parties. After a query from a user about how do I read the compressed images in "Montage" ? the usual brief note on Rice compression was added to the data processing page. Good to see new users ! MJI has been a reluctant participant in astrowise telecons but did at least learn from a recent set of KIDs v1.0 release notes that a video board was replaced in OmegaCam early-June in an attempt to improve the gain stability of the detectors. Good of ESO to let everybody know and you can bet that information isn't in the FITS headers. Video board changes usually imply gain changes implying further still that a comparison of pre- and post-board change flats would be good news. <<<< Independent checks of the illumination-corrected ATLAS catalogues by Nigel Metcalfe in Durham showed they do what it says on the tin. Even the gain unstable detectors were spotted. There is an ongoing request from Richard Hook for more PR images from VST. Some of the VPHAS observations have potential but the short exposure times and the lack of dithers may make post-colour image manipulation more challenging than normal. 9. Spectroscopic stuff ----------------------- GaiaESO pipeline comparison test results look generally very good and there has even been a recent release of stacked spectra. All quiet for now on the 4MOST data management front. The feedback from the internal interim WEAVE review was good. NAW is busy Gantt'ing and xl'ing managerial information. 10. ESO in-kind stuff --------------------- JRL is busy dismantling the why's and wherefore's of the ESO-supplied HAWKI pipeline/workflow prior to MJI and STH helping out with the assessment and feedback report to ESO. 11. AOB/news ------------ MJI reminded folk that the registration to the ESO workshop "Science from the Next Generation Imaging and Spectroscopic Surveys" was closed on 24th July, so interested parties had better get a move on. ...... and finally, JRF is leaving at the end of the month so a CASU-style bread and beer tasting ensued after the meeting, preceeded by Margarita cocktails and a bottle of some interesting, if weird, Pamplemousse-infused vodka, and postceded by an evening BBQ - hence the late delivery of these minutes hic ! Continuing actions ------------------ STH - progress evermore the VISTA calibration papers writeups, plus VISTA Y and Z Vega offsets and circulate draft papers before next CASU meeting JPE - continue to characterise radial extent of diffraction spikes for a STH range of values of NDIT*DIT for pawprints, and thence tiles SC - continue investigation of computer montoring web page foibles when the new switch is installed MJI - is still keeping an eye out for UCAC4 MJI - empty in-tray and look at "averaged" linearity data MJI - update bandmerging prog to cope with weird VPHAS observing pattern New actions ----------- MJI - investigate astrometric distortion patterns seen in tiles JRL - did the VST video board change alter the internal gains ?