Second Virtual Meeting 2020, October 2nd/October 6th via zoom

Telecon Agenda:

• Implementation note for radio data in the VO - we would like to document the current efforts to implement radio data in the VO. If you have an implementation and can write a paragraph or two describing it we will assemble these into a document to post to the IVOA site. (see below a skeleton of the note)

• Planning a future workshop on VO implementation, targeted to software developers from radio astronomy archives.

• Common needs for the various types of radio data in the VO environment (visibility data, single dish spectroscopic & imaging data, pulsar data), for example, description of data provenance, typical use of frequency vs wavelength etc. (see below some slides)

Radio Interest Group virtual meeting 2nd Oct 2020 (18 participants)

——————————————————————————————————

Implementation note for radio data in the VO:

Requested input from people/organizations who have successfully made radio data available through the VO, or plan to do so in the near future. 1-2 paragraphs on services offered, lessons learned, any other comments. Volunteers included James Dempsey, Mark Kettanis, Baptiste Cecconi, Juane Santander-Vela, Paul Harrison and Yan Grange. Francois has written a skeleton document, and the inputs will be organized according to type of data and services offered. Input is requested on about a one month timescale to have a draft document ready for the IVOA interop in November.

Workshop for developers:

Scheduling unclear, should concentrate though on training software engineers to install and use existing packages of VO software by e.g. CADC, CDS etc as coding from scratch is hard and documentation of standards is insufficient in places. Also important is to correctly map the metadata from the observatory’s data products to the VO ObsCore and any necessary extensions to enable meaningful search results in the VO. This would be a good topic for the radio data BOF at ADASS.

Common needs of radio data in the VO:

ASTRON has HIPS running on the TGSS ADR, looking at adding other datasets.

ASTRON also working on representing uv-coverage (ellipticity and filling factor of uvplane). Unclear how to represent largest angular scale yet (depends on interpretation of ObsCore s_resolution_min, or whether another keyword is needed as an extension).

Provenance: IVOA provenance data model is inspiring work in ASTRON.

One outstanding problem is representing FOV when it is dependent on frequency across a large fractional bandwidth - extreme case is NenuFAR (10-80MHz) but also affects VLA (e.g. S-band 2-4GHz). Could break the data up into sub-bands, but artificial. Really need a function to generate the coverage as a function of frequency for a given search.

Radio Interest Group virtual meeting 6th Oct 2020 (11 participants)

——————————————————————————————————

(Same agenda as above)

Implementation note:

Severin Gaudet, Peter Teuben and Alessandra Zanichellli agreed to write contributions for the implementation note. These will list services offered (or plans for them), together with “storytelling”/lessons learned putting radio data into the VO.

Developers workshop:

Discuss at ADASS BoF. Hard to set up VO services from scratch - may be better to focus on harvesting/mapping metadata, then running services through a large datacenter. May need to extend the mandate of big data centers to allow/encourage this.

Common needs for radio data in the VO:

Francois found an earlier paper (2010) by Anita Richards describing how radio data could be incorporated into the VO - https://wiki.ivoa.net/internal/IVOA/SiaInterface/Anita-InterferometryVO.pdf In particular, this discusses how to represent uv-coverage

Francois presented slides and gave a summary of the discussion from the Oct 2nd meeting, including the discussion of needing extensions to ObsCore to properly represent radio data.

Severin gave a summary of ALMA’s progress - astroquery.alma is now available, with a TAP interface to the ALMA Archive. A Datalink interface (with pyVO underneath) will be rolled out shortly (so data can be extracted). A SODA service (cutouts) is also in the works.

There was some further discussion on representing uv coverage in the VO - parameters of the uv-distribution such as quartiles or percentiles of the uv distribution that could be easily searched on. The full distribution could be available via datalink. Vincenzo discussed how it would be useful if the uv distribution could be generated dynamically based on selected spectral windows from a single observation, or even from separate observations combined. Alessandra described the Italian efforts, both for interferometry data, but also including pulsar data and OTF mosaicked Single Dish data. Also noted that dynamic calculation of metadata and plots would be useful for beam-formed data.