Agenda

Minutes in Bold

CASA/JIVE VLBI Coordination Meeting

2019 June 28 @ 1400UTC

Attendance: George, Jay, Ryan, Mark, Des

Logistics

  • Future Meeting dates (usually last Wed of the month):
    • NEXT Meeting: 2019 Aug 21 1400 UTC (postponed from nominal July 31)
    • 2019 Sep 25?
    • 2019 Oct 30?

General Admin

  • CASA 5.6/6.0
    • CASA 5.6 Confluence page
    • Feature freeze: 2019 Jun 07
      • Currently in validation/bug-fixing period, release branch in next few weeks
      • Public CASA 5.6/6.0 release by end of July
      • ALMA/EVLA pipeline release by Oct 1.
  • EHT-related article planned for next newsletter, expected ~alongside CASA 5.6 pipeline release
    • Probably a contribution from Michael Janssen, via Ilse
    • Newsletter scheduled for ~October
  • (Mark) EHT polarization workshop
    • In Bonn, in July; EHT members only. Mark may adapt some ad hoc resolved-calibrator polarization-calibration scripts as tasks.
      • No news. Some internal EHT discussion about role of CASA in further work. Possibly some input to priorities after this workshop.
  • (Mark) Summer student at JIVE investigating scpetral line VLBI with CASA
    • A novice, first learning VLBI in AIPS...
    • Spectral line exposes the single-channel (no delay solve) use case, which will need some attention for 5.7
  • Jay Blanchard now in Socorro, cultivating interest in CASA VLBI (also SRDP)
    • Some discussion of fringefit in recent local CASA User group meetings
      • Well-received, useful discussions, some confusion (mostly because this is largely a VLA crowd less accustomed to VLBI); ambitions for VLBA pipelining.
      • Some VLBI-related validation likely to become available here in Socorro in the coming months: Jay + another new VLBA staff member (Justin Linford, w/ some CASA experience) arriving in the next month or so.
      • Some discussions have begun about potential science involving an ngVLA prototype antenna (sited at VLA) + VLBA. This may eventually generate some CASA/VLBI imperatives, e.g., including managing the conversions necessary to cope with linear feeds on the ngVLA antenna (a la EHT); maybe we should add Ivan Marti-Vidal's 'polcovert' to CASA.
  • (Ryan) CASA Next Generation Infrastructure
    • Trade studies ongoing...
      • Evaluation of several selected technologies ongoing, working toward six-month prototyping period to begin later this year.
      • Rust wrapper around casacore by CfA proving interesting...
      • Trying to wrap of MSv3 study, which will provide schema for CNGI data storage.
  • Other admin?

Development/Features

  • (Mark) Requirements collation ConsolidatedRequirements....Any news? Janssen contribution?
    • Janssen's contribution merged, prioritization pending
  • VLBI Integration branch: CAS-12325
    • Any issues?
    • (Des) Janssen has found a so-far unreproducible bug---analysis of this still pending
  • Items for CASA 5.6/6.0 development
    • (George) CASA 6 VLBI task interface verification
      • (Mark) fringefit task: coordinating with Bob Garwood. OK?
        • Still in progress; merge still desirable/possible.
      • (Mark) fringefit unittest using real data: Status? Done!
    • (George) Calibration changes ~relevant to VLBI:
      • CAS-6747: Relative frequency interpolation for channel-dependent calibration
      • CAS-11552: Updates to multiband delay solves in gaincal for heterogeneous spws
        • SolveDataBuffer::centroidFreq / SDList::aggregateCentroidFreq. (cf SDBList::centrodFreq added for fringefit)
    • (George) Look into correlation-dependent flagging issue (CAS-695)
      • No progress, unfortunately. Defer to 5.7.
    • (Des) Dispersive term
      • (Des) CAS-12493: Extended fringefit CalTable for dispersive term support
        • PR #995 pending. Awaiting final validation? Still TBD, maybe Janssen. NB: affects Mark's new test; will probably merge new version (along with re-merge of master).
        • (User warning since not backward compatible? Attempts to use old will fail comprehensibly.)
      • (Des) Work on actual solve ongoing for CASA 5.7, including user control and an FFT-based option for estimating the dispersive term.
        • Must exist in code by end of year (funding progress report)
    • (Des) CAS-12480: fringefit performance/stability issues traced to neglecting some terms in fit accumulations; now on track to validate for 5.6
      • PR #984 merged June 4
    • (Des) CAS-xxxxx? fringefit weighting control implemented
      • Status?
        • Defer to 5.7. Need good use case for testing/validation. Philosophy of this mechanism "interesting".
    • (Des) CAS-12501 Iteration control (niter) for fringefit
      • PR #994 merged June 26
    • (Des) Baseline stacking
    • (Des) Wide bandwidth
      • Must exist in code by end of year (funding progress report)
    • (George) perscan interpolation option? (5.7)
    • (George) consider overlapping solution intervals? (5.7)
    • (Walter) Advanced VLBA Amplitude calibration a la Walker memo from a few years ago
      • Bandpass normalization flexibility: voltage vs. power

Verification/Validation/Documentation

  • (Jay) Testing of fringe fitting with a model... status?
    • (George) Confer with Kumar on improving warnings for convolved model images or with wrong units
      • TBD
      • Des has python functions that perform conversions of CC tables into image grid or ComponentList. We should adopt these in CASA proper---George will coordinate with Kumar on this (for 5.7).
  • (Mark/George) Status of recent casacore changes?
    • SIGMA/WEIGHT storage manager fix
    • FITSIDI Single-band case
    • Not urgent for 5.6, but need to confirm when this is merged to CASA master.
      • This has been merged into 5.6 now.

AOB?

-- GeorgeMoellenbrock - 2019-06-28
Topic revision: r2 - 2019-07-08, MarkKettenis
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback