Welcome to the VLA Sky Survey Forum!

smyers@nrao.edu
Posts: 2
Joined: Wed Jan 04, 2012 12:25 pm

Welcome to the VLA Sky Survey Forum!

Postby smyers@nrao.edu » Mon Sep 09, 2013 4:20 pm

Welcome to the Jansky VLA Sky Survey (VLASS) Science Forum!

This forum is for discussion of survey science goals, techniques, development areas, and overall design, and other aspects related to the definition, observing, and data processing for a new generation VLA radio sky survey.. See https://science.nrao.edu/science/surveys/vlass for details on the VLASS initiative.

All users that wish to contribute to the forum discussions are required to register via http://my.nrao.edu. See the Terms of Use post for guidelines on use of this forum.

We look forward to your participation in the VLASS!

-Prof. Stefi A. Baum, RIT
Dr. Steven T. Myers, NRAO

gerryharp
Posts: 2
Joined: Mon Sep 30, 2013 3:25 pm

Improve VLASS with Companion Telescopes

Postby gerryharp » Mon Sep 30, 2013 3:54 pm

Hi

We all know that the VLA is a fantastic interferometer and is very well suited to capture of the short and medium angular scale structure of the sky. Yet there is always the "zero-baseline problem," where one cannot measure the longest length scales on the sky without single-dish measurements. In the UV plane, there is a "hole" near zero wavenumber where no fourier components of the sky can be measured using antenna cross correlations.

1) If the autocorrelations (cross correlation of each antenna with itself) of the VLA antennas were used carefully, it might be possible to use them to estimate the longest length scales to make a complete sky survey. Has anyone studied the feasiblity of such analysis? The data come for "free." My guess is that the telescope electronics are not designed to produce sufficient quality single dish data to augment the cross-correlation data, but it might. For one thing, we can be certain that the autocorrelation data is from exactly the same time and region on the sky.

2) Unfortunately, the zero-baseline hole at the center of the UV plane is always larger than the region of UV plane covered by a single VLA dish. So another approach to make the VLASS complete is to use a companion single-dish telescope, say GBT, and run simultaneously with the VLA. The two telescopes should be pointed in the same directions at all times, so that any time-variable sources will be captured by both telescopes. The only problem is that GBT has a smaller FOV than VLA hence it might miss some transient signals. Still this is a worthwhile approach for static and transient sources on longer time scales.

I hope this post will start a discussion. There are some science goals, such as intergalactic HI that would benefit greatly from getting ALL the fourier components.

Gerry Harp
Director of Center for SETI Research
SETI Institute

smyers
Posts: 29
Joined: Mon Sep 09, 2013 4:43 pm

Re: Welcome to the VLA Sky Survey Forum!

Postby smyers » Mon Sep 30, 2013 5:08 pm

Gerry,

Those are some excellent points for discussion! We have not yet tried using the VLA auto-correlations for complementing the cross-correlations for imaging, due to some of the issues you raised (like the fact that the VLA electronics has not been optimized for stability for total power, nor is there a beam-switching provision beyond moving the telescope). There is also the constraint that due to a correlator design choice we made, so by default you do not get the full set of auto-correlations for all antennas along with all the cross-correlations (I think you get around half the antennas). You can get these by trading resources (trading off polarization) or by using extra resources - we have been discussing possibilities such as setting it up to record the full auto-correlations on a much faster timescale than the cross-correlations (useful for transient searches, as well as helping with noise issues such as 1/f). This would be something fun to test out when enabled! But in any event there might be something clever we can do with the fact that we get auto-correlation data in "total power" for N (=25-27 or maybe 12 in the default case) antennas where the noise is mostly independent (except for common-mode RFI etc.) for the two polarizations R and L (as well as the RL and LR auto-correlations) - could one could use the cross-correlation of the auto-correlations or something similar to get more robust imaging information?

Another thing to consider is that many scenarios for a VLASS involve On-the-Fly (OTF) scanning, where we scan the antennas in a given (linear) trajectory across the sky while correlating the data - thus there is in principle continuous data from the auto-correlations more analogous to what a single-dish might do for optimal scanning. I wonder if use of OTF with fast auto-correlations in parallel with slower cross-correlations might be useful for imaging larger scale emission such as you described.

Obviously use of companion telescope(s) such as GBT and/or others would be an excellent idea to pursue. Something that came out of the Radio-LSST workshop last May was the idea of "co-observing" with multiple (multi-wavelength) telescopes/arrays. I think it is a good idea in any event to pre-publish (and provide in real-time) the observing schedule and pointing positions during the survey to enable this.

I hope we get more discussion on this topic :)

-Steve Myers

gerryharp
Posts: 2
Joined: Mon Sep 30, 2013 3:25 pm

Simultaneous observations with collaborating telescopes.

Postby gerryharp » Sun Oct 20, 2013 6:19 am

Hi Steve and all
We gotta get this discussion off the ground! C'mon people! Join in.

Forgive me for a shameless plug, but another radio telescope that could dramatically improve VLASS scientific outcomes is your little brother up north, the Allen Telescope Array. The ATA uses 42 (6m) dishes, each smaller than VLA dishes (25 m) resulting in an unprecedented field of view of 3.5 degrees diameter at 1 GHz, scaling down at higher frequencies. Thanks to a generous donation by Franklin Antonio of Qualcomm ($3.5 M), all the receivers at the ATA are being upgraded with a new design achieving ~30K system temperatures below 5 GHz, ~50 K Tsys up to 13 GHz, and useful sensitivity up to 18 GHz. This is a million times better than the original ATA feeds where Tsys at 8 GHz was ~200K. Jack Welch figured out what was wrong with the orginal design and has patented the corrections that make his dual-pol 4-octave receiver a real game changer.

For example, with the right back-end RF and digital electronics, we can simultaneously study 4 different frequency ranges from 0.9 to 10.5 GHz (and to 18 GHz with an upgrade of analog fiber links). This makes the ATA the perfect instrument for dramatic studies of so-called fast radio bursts (FRBs) due to rare extragalactic events recently discovered at Parkes and lately confirmed at Arecibo. The ATA will have adequate sensitivity to discover ~1 FRB per day based on recent estimates of appearance rate. This is as good, or better, than can be accomplished at VLA looking for the same bursts. Although the VLA has greater sensitivity (collecting area) than the ATA, the VLA field of view covers only 5% of the ATA field of view. On balance, the ATA is a better instrument than VLA for performance of a FRB survey especially because we can devote thousands of hours per year to the FRB survey while in the context of VLASS, the VLA can devote <10% of this observation time to the same project.

My suggestion to you (all), is that if the VLASS science is to include an exploration of fast radio bursts, a very hot topic right now, then it would be better to subcontract the FRB survey to the ATA, which is better suited to the survey and has observing time available to perform a more thorough survey than could ever be scheduled at VLA. This frees up more time on the VLA for doing what it does best, making the highest resolution and highest fidelity radio images on the planet.

I will be very interested to hear different people's opinions about this novel suggestion. VLASS should leverage other US instruments to produce the best science ultimately improving the "bang for the buck." This sort of thing hasn't been done before hence initial reactions might be, "That is impossible." Lets move beyond this obstacle and ask ourselves if science outcomes of collaborating with multiple telescopes would be worth the pain of multi-site collaboration.

Thanks!

Gerry Harp
Director, Center for SETI Research (& ATA)
SETI Institute

chal.nikkal
Posts: 1
Joined: Sun Jan 11, 2015 12:24 pm

Re: Welcome to the VLA Sky Survey Forum!

Postby chal.nikkal » Sun Jan 11, 2015 12:26 pm

You can get these by trading resources (trading off polarization) or by using extra resources - we have been discussing possibilities such as setting it up to record the full auto-correlations on a much faster timescale than the cross-correlations (useful for transient searches, as well as helping with noise issues such as 1/f). This would be something fun to test out when enabled! But in any event there might be something clever we can do with the fact that we get auto-correlation data in "total power" for N (=25-27 or maybe 12 in the default case) antennas where the noise is mostly independent (except for common-mode RFI etc.) for the two polarizations R and L (as well as the RL and LR auto-correlations) - could one could use the cross-correlation of the auto-correlations or something similar to get more robust imaging information??


Return to “VLA Sky Survey”

Who is online

Users browsing this forum: No registered users and 1 guest