Configurations Overview VLBI Opportunities Bryan Butler NRAO U

Configurations Overview & VLBI Opportunities Bryan Butler – NRAO U. S. Radio/Millimeter/Submillimeter Science Futures II U. S. RMS Science Futures II AUG Aug 3, 2016

ng. VLA Configurations One of the most important design decisions for a synthesis array is the configuration of the antennas. Along with sheer number of antennas, it determines or affects: – Resolution – Imaging quality – Brightness temperature sensitivity – Calibration strategies And there are many cost and operational implications which are associated with decisions on antenna configuration(s). AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations The Good: – Whatever size and number of antennas we arrive at in the end for ng. VLA, the number will be large (of order hundreds), which allows quite a bit of flexibility. – There has been a tremendous amount of time and effort put to this problem in the past (e. g. , 39 memos in MMA/ALMA memo series alone – maybe as many for SKA). AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations It’s fairly straightforward to make a toy configuration which has Gaussian distribution of baselines, good u-v coverage, and other desirable properties (c. f. Boone 2001, 2002; Conway 1998, 2000). AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Or other configurations which have similar desirable properties (Clark & Brisken 2015; based on Conway 2000). AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations The Bad: – Despite how much work has gone into, for example, optimizing antenna configurations for imaging, that optimization depends strongly on the science (notably the size and spatial structure of sources). For a general purpose instrument, and especially one that may not be reconfigurable, difficult decisions (and compromises) will have to be made. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations The Ugly: – Even if we could come up with an acceptable “optimal” configuration, the practicalities of where we can sensibly place antennas (due to availability, accessibility, power, fiber, etc. ) will almost certainly outweigh the desire to have that ideal configuration, at least for the longer baselines. This is because the scale which we’re currently considering (hundreds of km) is larger than local geologic/geographic scales. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Because of these practical considerations, and the underlying science, it is instructive to think of the ng. VLA as consisting of three parts: – A short baseline component, with a large number of antennas out to 1’s of km (co-located with current VLA). – An intermediate baseline component, with antennas distributed on the plains of San Agustin, out to 10’s of km. – A long baseline component, with antennas out to 100’s of km. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Short Baseline Component • Main driver is sensitivity to large-scale structures (high brightness temperature sensitivity). • Many antennas out to few km diameter. Say 1 km maximum baseline, with 18 m antennas, then 10% filling factor would be of order 100 antennas. • Maybe some fraction of these in an even more compact core (20 antennas in 200 m is almost 50% filled)? AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Short Baseline Component There is still a hole in the u-v plane on the scale of the shortest spacing between antennas – how to fill it? • Some antennas total power? • Array of smaller antennas? • Larger antenna? • Some combination? Does it even need to be filled for the most important science? AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Intermediate Baseline Component • Bridges the gap between the short and long baselines; needed for good imaging on intermediate scales. • Ideally want “smooth” distribution of baselines connecting to short baseline component (some might argue Gaussian, some might argue power law). • Antennas distributed on the plains of San Agustin, out to 10’s of km. • Land access already becoming an issue, at this scale. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Intermediate Baseline Component AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Intermediate Baseline Component AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Long Baseline Component • Need long baselines for resolution, of course. But cannot only have a few antennas or imaging will suffer. • Ideally want “smooth” distribution of baselines connecting to intermediate baseline component (some might argue Gaussian, some might argue power law). • Antennas distributed across state of NM (AZ and TX also? ) and maybe into Mexico, out to 100’s of km. For scale, note that the state of NM is of order 550 km across. • Land availability, access (topography, roads), power, fiber now a serious issue. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Long Baseline Component Easy to plop down a 300 km diameter circle on a map, but difficult to find good locations for antennas. Topography is an issue. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Long Baseline Component As is land availability and access (roads). AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Long Baseline Component And power. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Long Baseline Component And access to fiber. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Long Baseline Component Frazer Owen and Eric Greisen have started to address some of this. But many uncertainties remain. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Longer Baseline Component • We have components on 1’s, 10’s, and 100’s of km – the next step (1000’s of km) is the territory of VLBI. • In fact, the shortest baseline of the VLBA (LA-PT) is 236 km; shorter than the few hundred km maximum baseline we’re discussing for ng. VLA. • We should retain the capability of using the longest baselines of ng. VLA along with the VLBA, and even adding some ng. VLA antennas (or stations) on somewhat longer baselines (many hundreds of km, at least). • A more ambitious possibility – absorb the VLBA antennas into ng. VLA. The VLBA antennas (or replacements thereof by ng. VLA antennas or stations) then simply become the longest baseline component of ng. VLA. • Software becomes more complex (scheduling, observing techniques, correlation, data reduction, etc. ), but the scientific U. S. RMS Science II payoff can be significant, and the. Futures operations gain should not AUG 3, 2016

ng. VLA Configurations Implications for Phase Calibration Techniques of phase calibration may have implications on configurations of antennas: • The paired-antenna or calibration/reference array technique of phase calibration requires multiple antennas (though not necessarily all identical) per location. • This might be appropriate for the long baseline component, but perhaps not for the intermediate, and probably not for the short. • See Dave Woody’s talk. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Reconfigurability • The possibility of moving antennas, at least in the short baseline component and possibly even in the intermediate (but probably not in the long) is intriguing. Reconfigurability certainly allows more flexibility to a broader range of science. • The difficulty is cost – for roads, transporters, and antennas. • See Jim Condon’s talk. AUG 3, 2016 U. S. RMS Science Futures II

ng. VLA Configurations Open Questions • What is the size and number of antennas? • What is the longest baseline? • How much relative collecting area in the main three components (short, intermediate, and long baselines)? • Do we need to fill the u-v hole, and if so how? • What are the implications for phase calibration (or vice versa)? • Should it be possible to reconfigure the short and/or intermediate baseline components? • For the long baseline component, is it possible to get reasonable antenna locations given issues of land availability, access, power, fiber, etc. ? • How much integration with VLBA should be pursued? AUG 3, 2016 U. S. RMS Science Futures II

www. nrao. edu science. nrao. edu public. nrao. edu The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. AUG 3, 2016 U. S. RMS Science Futures II