Report from CASU ================ Grant funding ------------- This was finally resolved in September. The rolling grant submitted May 2007 and due to commence April 2008, has now been finally issued as one slimmed down rolling grant together with a so-called VISTA exploitation grant to cover the slimming costs. The overall grant start date was moved to April 2009 and notionally runs 3 years to end of March 2012 with some caveats. In practice for WFCAM this means that we now have enough equipment money to maintain our processing kit and provide enough disk storage for the raw (including tape backup) and processed data up to end of March 2012. Due to knock-on delays in staff recruitment and hiring out staff to other projects we are still short-staffed overall which continues to limit the effort available for WFCAM reprocessing and detailed investigation of problems/issues. However, we anticipate having enough staff in post to continue normal WFCAM processing operations at least until March 2012. Data transfers -------------- These are still running smoothly through completion of 09A and the first few months of 09B. As we have noted before, the raw WFCAM data automatically arrives in Cambridge and is checked, MEF'd, OMP'd and ingested within 24 hours of it being taken. The nightly processing typically runs a few days behind this. Checking the products, and bundling up the monthly photometry updates takes the rest of the time. The raw WFCAM data archive in Cambridge is also up to date, as are the transfers of all raw UKIDSS and calibration data to the ESO archive. The current state-of-play including transfers and processing status can be found at http://casu.ast.cam.ac.uk/surveys-projects/wfcam/data-processing/ The total of raw WFCAM data received for semesters 05A-09A and the first few months of 09B totals around 160 Tbytes (~40 Tbytes of actual RAID disk storage with Rice-tile compression, equivalent to ~8 million 2kx2k science images). Processed data has risen to around 260 Tbytes (~65 Tbytes compressed). Cassegrain UKIRT archive ------------------------ The Cassegrain UKIRT archive has been undergoing an update, in parallel, to correct some missing FITS headers and some missing parts of datasets. Due to problems with reading our copies of the original DLT archive tapes this has involved JAC resending us some of the recalcitrant data and has meant the whole process is taking far longer than anticipated. The current version of the UKIRT archive has been available throughout this process. When completed we will make a new offline backup on LTO4 tapes. WFCAM processing ---------------- Semester 09A processing is complete. Semester 09B started on July 27th and processing of that continued seamlessy on from 09A. Occasional small problems are fixed as they occur, e.g. a few MSBs have had incorrect RECIPE keywords but advance warning from JAC has enabled these to be corrected before processing begins. After the recent standown a new problem with pickup, particularly on detector #3, was noticed, reported back and fixed. There are still occasional holdups in processing whilst waiting for suitable new twilight sequences to be taken but these are not causing any significant problems. One aspect worth raising is the lack of recent non-linearity measures to check the linearity or otherwise of the detectors. While we appreciate that issues with illumination of the new dome screen have caused some delay, it has been several years since the last set of suitable measurements were taken. As of 20091001 processed data is being compressed using the latest CFITSIO Rice compression software. The changes are only relevant for compression of 16-bit integer data, i.e. the confidence maps. Both CFITSIO and DS9 seamlessly interpret old and new versions. The 16 bit-specific compression is slightly more efficient and produces slightly smaller (by a few %) compressed files. Pipeline updates ---------------- There have been very few changes recently. The most notable is that the UDS requirement for using the object-masking pawprint sky subtraction method has been incorporated directly into the pipeline's automatic allocation of sky subtraction strategies. This removes the general need for reprocessing UDS data from 09A onwards. WFCAM reprocessing ------------------ The reprocessing queue has been whittled down to some nebulous GPS data for 08B (for DR6) and the remaining UDS request for J,H-band data from 05B-07B. Since the last Board report we have finally laid to rest the Petrosian et al. fix and applied it to all processed data prior to 08B (for 08B onwards the pipeline had already had the fix installed). The relevent code was sent to WFAU to do the same with the catalogues at their end, and to form the basis for their direct DB update. DXS tests --------- In conjunction with Alastair Edge we have carried out some tests with the pawprint sky method on DXS data, the default is to use the tile method. A pair of 08A DXS datasets were selected for the tests. Using pawsky on this data makes little difference to the visual appearance of the pipeline produced skies/science frames nor to statistical measures of pixel noise properties. After stacking one deep DSX K-band set using both methods conclusions are: there is very little depth difference between using pawsky and tilesky methods on DXS data. The only significant difference is a slightly blotchier background appearance (which you can optionally process out using e.g. the nebulosity filter) and which has no discernible effect on the local rms noise and hence limiting depth attainable; the individual MSB outputs and the deep DXS stacks are very close to photon noise limited as judged both from direct rms pixel noise measurements and from randomly placing 2 arcsec diameter apertures over the frames. Comparing noise in aperture with that expected from rms pixel noise shows they are the same, in apparent contrast to results for the deeper UDS stacks.