N A W G ======= NRAO Algorithms Working Group Meeting - DRAFT MINUTES Date: 2005-2-09 (Wednesday) Time: 1300 MST = 1500 EST Video Hub: CV-conf (others call in) Rooms: SOC317/CV311/GB241/TUCN505 NOTE: Again, I apologize for the problems encountered in getting the visuals to all the sites (in particular GB with only one TV in the room). In the future, I will try to make the talk viewgraphs available on the web beforehand and also to make sure we know how to work the cameras right! Thank you for your perseverence and patience... ======= Minutes: Send additions and corrections to smyers 1. Organizational issues o Call for topics - anyone who wishes to speak in upcoming meetings (next time is 9 Mar) or know of someone they wish to nominate to talk, or just have an idea for an interesting topic to discuss should let me know! o I intend to continue to broadcast the info on scistaff unless people object. I think these issues are of sufficient interest to the scientific staff to make this worthwhile, rather than setting up yet another mailman list. If you know of anyone else who is not on scistaff who might be interested, I can include them in future mailings. 2. Project-related (ALMA, EVLA, GBT, VLBA, &c.) algorithms o LWA issues and plans - A group of us in Socorro (Frazer, Sanjay, Kumar, Urvashi, and myself) have been discussing issues related to low-freqency imaging and calibration for the LWA project (the NRL and Southwest Consortium array project). Frazer reports that there is likely to be an ad out for postdocs to work with us (and UNM) on LWA im/cal problems, possibly as many as 4! Stay tuned for developments in this area, and if there are others in the observatory interested in participating in LWA discussions or related work (e.g. ionospheric problems, wide-band wide-field imaging) let us know. 3. General algorithmic issues o Astronomy Science Data Model (ASDM) update - Ed discussed the recent meeting held here in Socorro last week with the ALMA SDM group (Francois Viallefond, Michel Caillat, Ed Fomalont, John Benson, Alan Farris, Joe McMullin, and others) regarding the ASDM and changes needed to make this useful for EVLA. It sounds like most issues were resolved, though there were still some unsupported features (such as baseline-dependent integration times) which may or may not matter. The GBT single-dish data model was not yet available, and so did not enter this discussion. 4. ***Main Event*** - Presentations and Discussion Items ******************************************************************* o Bill Cotton - single dish image processing and calibration of GBT data ******************************************************************* Viewgraphs available at: http://www.aoc.nrao.edu/~smyers/nawg/presentations/ NAWG20050209-Cotton-GBTOTF.pdf ******************************************************************* Discussion points: - Jeff mentioned that there were some multi-beam systems that used cross-correlation between pixels to remove atmospheric and gain effects (any references?). It was noted that the GBT system does not (and will not) support synchronous outputs which would allow this. - Bill thought that there was a factor of 2 loss in sensitivity in the "matched filter" convolved OTF image above the level for just an ON-OFF position-switched difference image (above and beyond that due to spending more time off source), for example for a point source. Others disagreed (Jay and Jeff for example) pointing out the successes that the 12-m OTF mapping had, which did not show any loss. It was not clear to me (STM) that people were talking about the same reference (e.g. versus a straight integration w/o any switching at all, or an ON-OFF 2-position) either. There were also some questions about the proper (lossless) resampling and deconvolution kernels to use. - Bill pointed out that there is an interaction between the scan baseline removal (which effectively removes low-order signal variations in the scan) and reconstruction of diffuse structure. Note that CMB experiments have been dealing with this problem also, and that this should be mitigated somewhat by having a (complex) interlocking scan pattern with many crossings, though you still end up downweighting the large-scale spatial modes. - Bill pointed out that the scanning (with large ~20s turnaround overheads on 10s scans) was sub-optimal. It was not clear whether some other modes were available, such as the ability to take data during the turnarounds and to use other scan patterns (like a lissajous or spirograph pattern). Scan efficiency will be important, particularly for PennArray. - The PennArray will have 64 beams (at 2f lambda spacing) and thus all other things being equal should provide much better performance that the single-beam tests reported here, as there is essentially only one sky and atmosphere to take out. Note that development of the mapping and calibration algorithms to deal with PennArray is very important, so it was good to see the results presented today! NOTE: let me know if I missed any points or got them wrong. Also I can include any further discussion on the NAWG page. COMMENTS: (hliszt) "In reading about the discussion of OTF, when Bill talks about OTF and others cite GBT or 12m examples, Bill is talking about continuum and the others are talking about spectral line. The availability of signal free baseline regions in spectral line data allows good data to be had in situations where continuum mapping would be impossible or compromised. This is also why Bill's and Jeff's reminiscence might differ on such things as the influence of baseline removal on the recovery of diffuse structure. In the line it could be recovered under situations where the continuum structure couldn't. All the tests I did on the rms in maps from the 12m showed as good agreement between a priori and a posteriori rms as could be expected (i.e. quite good) given the diceyness of the overall calibration at the 12m. To my knowledge I am the only person to have published OTF continuum images from the 12m (90 and 140 GHz images of Cas A). It was done by dual beam switching but examination of the data showed that the same results could have been obtained without the beam-switching. In general, OTF maps from the 12m simply don't look like maps made from pointed observations. They clearly recover fourier components which the pointed maps obscure. Its striking to anyone who's made pointed maps how different the OTF data are and one would never go back to maps made from individual pointings given the choice. At the 12m there was fully a > 10:1 gain in efficiency (independent beam areas sampled/unit observing time) going from pointed observations to OTF." ******************************************************************* 5. Relevant reading: o Paper & Memos: - Beyond the isoplanatic patch in the VLA Low-frequency Sky Survey (W.D. Cotton et al.) Abstract: http://link.aip.org/link/%3FPSISDG/5489/180/1 - Solving for the Antenna Based Pointing Errors (Bhatnagar, Cornwell & Golap) http://www.aoc.nrao.edu/evla/geninfo/memoseries/evlamemo84.pdf - EVLA and SKA Computing Costs for Wide Field Imaging (Cornwell) http://www.aoc.nrao.edu/evla/geninfo/memoseries/evlamemo77.pdf - Effects of Atmospheric Emission Fluctuations and Gain Fluctuations on Continuum Total Power Observations with ALMA (Holdaway) http://www.mma.nrao.edu/memos/html-memos/abstracts/abs490.html http://www.mma.nrao.edu/memos/html-memos/alma490/memo490.pdf o Post-processing requirements: - ALMA Offline Data Processing Requirements http://www.alma.nrao.edu/development/computing/docs/joint/0018/ALMAoffline-sw-18.pdf - EVLA Offline Data Processing Requirements http://www.aoc.nrao.edu/~smyers/evla/evla_offline.pdf o GBT: - GBT Software Review Agenda & Presentations http://wiki.gb.nrao.edu/bin/view/Software/ReviewAgenda2004 - GBT Standard Observing Modes http://wiki.gb.nrao.edu/bin/view/Observing/StandardObservingModes 6. Upcoming meetings and deadlines: o 2005 Mar 1-3 RadioNET meeting (Jodrell Bank, UK) o 2005 Mar 9 NAWG meeting Agendas, minutes, and presentations available on the NAWG page: http://www.aoc.nrao.edu/~smyers/nawg/