Dear All, This is the summary of Terapix activities over the period from Feb. 12 to March 5 11 2009. The meeting started on March 5, at 14:30, room 281. The summary is a bit short, due do vacations and other committements and teaching duties that slew down processing activities. Please let me know if points are missing, wrong or need clarifications. Amicalement, Yannick {{I. T0005 and T0006: Yannick, Mathais S.}} {{II. Processing MegaCam Yannick, Patrick}} {{III. Processing WIRCam : Patrick, Yuliana, Henryi, Yannick}} WIRCam data processing (Ibata's data set - IBATA_08BF13): led by Yuliana, with help from Patrick & Yannick - Q.C.: visual inspection of the J-band images & redo qualityFITS for failed images - non-zero median background values (qualityFITS index.html & ds9 visual inspection): sky-subtraction pass with constant value - deriving astrometry with SCAMP: problems with the astrometric accuracy. High S/N Internal Astrometric Sigma values twice larger than the total Internal Astrometric Sigma values and numerous clumps in the 2-D Internal Error Plot. Tests performed: (1) uncertainties due to the small overlapping region: SCAMP the two fields separately (2) uncertainties due to bad image quality (seeing): SCAMP different sub-samples of images (3) uncertainties due to objects pre-selection: change the high S/N interval (SCAMP SN_THRESHOLDS parameter) - 10.0,100.0 (default) - 10.0,40.0 - 40.0,100.0 NOTE: Changing upper limit only has no effect on the Internal Astrometric Sigma values, since it does not trigger rejection. Rejection i possible by changing the lower limit. (4) uncertainties due to object mismatch: change SCAMP CROSSIR_RADIUS - 2.0" (default) - 1.5" - 1.0" - 0.5" (Best: CROSSIR_RADIUS=0.5") (5) uncertainties due to saturated objects: run SCAMP with FLAGS_MASK=0x00f0 (default) & IMAFLAGS_MASK=0x00ff - get saturation masks from *s.fits (pre-process but not sky-subtracted) images - redo qualityFITS with saturation masks - redo SCAMP with FLAGS_MASK & IMAFLAGS_MASK turned on - resampling and stacking images with SWarp: problems with the background of the stack - both large and small scale variations -> redo sky subtraction (on-going) - Processing of data 07BC23 by Henry (PI: Mohaddesseh Azimlu). The initial processing has been carried out and all the p images stacked. The data consists of five fields each observed in JHK filters. The fields are very dense stellar fields with diffuse emission. Unfortu- nately, the observations contain no 'off' fields. After some interactions with the PI (who sent us the observing proposal) it has been decided to carry out the sky subtraction using the provided data. Deadline: end of next week. - Processing of WIRDS data. Rich Bielby has completed a preliminary JHK stack of the D1-field containing all data. Number counts have been generated for Ks and H data. Chris Willott has carried out high-level verification of this data and everything checks out. Currently, rich is carrying quality control on the input images; rejecting bad stacks and re-doing the stacks with more stringent seeing / data quality cuts. We aim to release this data to the WIRDS collaboration by the end of the month. Olivier Ilbert has started looking a preliminary JHK catalogue generated by Rich. {{IV. YOUPI development: Mathias M., Greg.}} The image selector has been improved with two interesting features largely discussed at the last demo about Youpi image tagging capability: - One can now click on any image name in an image selection to display contextual information about this image. Display fields are image name, related instrument, run, channel, server-side data path, QSOStatus, alpha, delta, exptime, flat, mask, reg, ingestion related data (name, start, end, owner) - One can build an image selection without using available search criteria but instead by uploading a plain text file that describe the selection. It could allow to easily build an image selection of T004 images an then tag the selection with the appropriate tag. The syntax is very simple, each line must match the pattern IMAGENAME[, MD5SUM]. The MD5SUM field is optional. When not provided, the first occurence of the image is selected. A valid plain text file sample may be: 704303p, 0aadeec2151c565729d9b0074f63d920 704304p 704305p 704306p, a8c2dd51054919ba27b8ebef5c6b6950 704307p, 6bb0ca8e6e7db791d87e2465b428868a 704308p, 9f3319720ee24ee787ea2dc1b3105fa5 704309p, 7da7a930f8e60d71c68cdb214bba209b 704310p, 1f9495900d3d297553ba77564d1bf390 704311p 704312p, 7a7df40c157259edaefd8af92be08ce7 704895p, a99f15133f23738cf68fe94865144d46 ... A presentation will be made during the meeting. {{V. Developpement processing: Yuliana, Patrick, Fred}} - Yuliana is developing tool for deriving seeing statistics with bad quality image rejection based on kappa-sigma cliping (interactive option) and a seeing limit (interactive option). Results automatically written in an image rejection list. - Yuliana worked on the Condor/qualityFITS failures problem (with Fred) Results so far: (1) in all cases checked the problem occurs during the PSFEx run: - warning about unknown input keywords : PSF_NAME, SAMPLE_MAXELONG, (PSF_NSUPER) - output FITS-file name wrong: E.g. 1017489p_02_protopsf_1017489p_02_psfex.fits instead of 1017489p_01_protopsf.fits (2) it doesn't depend on image (all images successfully redone) (3) it doesn't depend on the node (failed on different nodes) (4) PSREx version is the same on all nodes TBD: To be further clarified with Emmanuel {{VI. Hardware/system/cluster Fred:}} {{VII. Divers (All):}} - Yannick presented the ESA/NASA dark energy space mission and the DADA Phase 0 project to the Terapix, Planck, Cosmologie et Grandes Structures and Galaxies teams as well as to the Head of IAP. Both the IAP Director and the Teams were supportive and agrred yannick could contact the PI of EUCLID to discuss how IAP could be more involved in the AO that will be issued in March. Alexandre Refregier's feed back was positive and Yannick will phone to Alexandre on Wednedsay MArch 11.