Galactic WG Summary so far (comments please!)

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

Re: Galactic WG Summary so far (comments please!)

Postby smyers » Mon Mar 24, 2014 6:41 pm

All,

I have posted the draft summary of the GWG definition to the GalacticWorkingGroup wiki page:
https://safe.nrao.edu/wiki/bin/view/JVL ... rkingGroup

-Steve

chomiuk
Posts: 12
Joined: Fri Feb 28, 2014 4:16 pm

Re: Galactic WG Summary so far (comments please!)

Postby chomiuk » Tue Mar 25, 2014 11:27 am

Hi Everyone,
I'd like to second everything Shami has said. I'm worried that this discussion is leaving some people behind---people who don't know about this Science Forum, or how far this discussion has gone (or people who just realized how to subscribe to the forum, like myself...duh). It would probably be good to send an email out to the community and let them know what is going on here, and what the current straw man plan is.

In addition, I realize that people like Shami, Tom Maccarone, and myself are advocating for a survey for which no white paper exists (high resolution, lower frequency, focusing on point sources). I was under the impression that there would be a second call for white papers from the 'call for VLASS white papers' site. Is this not true any longer? If so, we can try to put something together less formally to present our case in more detail.

It is certainly not true that there is no one advocating for a lower-frequency survey. I believe C band would yield optimal sensitivity while providing an excellent compromise between thermal and synchrotron emission. I am very uncomfortable with the Ku-band plan that was presented at the telecon yesterday, and I believe we need much more discussion and community consensus before deciding on a VLASS Galactic proposal. Although I do think it's great that we have a default plan to now discuss and fiddle with, and condense conversation around.

I believe it behooves us a Galactic community to merge smoothly, if possible, with the extragalactic and transient components of VLASS. They are certainly not going to be advocating for Ku band. I am worried if we propose something that is completely divergent from the rest of the VLASS plan, our Galactic survey will be easy to cut or diminish. I also believe we need a strong science case to focus on the Galactic plane, and an even stronger science case if we are going to do something wildly different than other VLASS components. I still haven't heard a thorough fleshing out of a Ku band science case, which we simply must carry out if we are going to ask for 3200 hours of VLA time.

Can the GUTS advocates please lay out their specific science goals for Ku band? What science can be accomplished by bridging the gap between 'thermal' freqs and 'synchrotron' freqs? I should say that I work on plenty of thermal sources like symbiotic stars and novae, but I'm not convinced that I'd gain a huge amount more insight by going to 15 GHz rather than 5 GHz. We've heard about some of the spectral line drivers, but the primary science that is going to be accomplished by a survey is continuum science, and I still haven't heard details about why 15 GHz will enable unique insights into specific classes of objects.

I will work with the lower-freq advocates to produce a similar science case with specific applications. And then we can see how many of our science goals could be served by the other survey. How does this sound as a plan? What is our timeline for this, Rachel and Cornelia?

shami
Posts: 7
Joined: Wed Feb 19, 2014 2:24 pm

Re: Galactic WG Summary so far (comments please!)

Postby shami » Tue Mar 25, 2014 1:11 pm

chomiuk wrote:I believe it behooves us a Galactic community to merge smoothly, if possible, with the extragalactic and transient components of VLASS. They are certainly not going to be advocating for Ku band. I am worried if we propose something that is completely divergent from the rest of the VLASS plan, our Galactic survey will be easy to cut or diminish. I also believe we need a strong science case to focus on the Galactic plane, and an even stronger science case if we are going to do something wildly different than other VLASS components.


Yes, exactly. If the Galactic group converges on a different enough survey, it will be easily spun off as a survey that should be proposed separately. And maybe that's the right outcome - a Ku-band survey of the Galactic plane might be sufficiently compelling on its own. But I don't think it is really compatible with the rest of the VLASS proposals.

I am finding this three-stream Galactic/Extragalactic/Transients process frustrating, because the final desired outcome was a single survey proposal, and that will obviously entail some collaborative decisions (not necessarily compromises - I don't think there's a sensible "compromise" between a 2-4 GHz B-array all-sky survey and a Ku-band Galactic plane survey). I'm not sure it works if each group goes off and does their own thing and comes back with orthogonal demands.

(As an aside, our whitepaper (#16 on the list) explicitly calls for a high resolution lower frequency survey of the Galactic plane with a focus on compact sources.)

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

Re: Galactic WG Summary so far (comments please!)

Postby smyers » Tue Mar 25, 2014 1:17 pm

Yes - the White Paper submission is still open! I guess this fact was buried in all the other notices, and I see now that the Call for White Paper page implies we will make an official second call. Guess I will send something to vlass-announce.

But please go ahead and submit something to the White Paper call or vlass@nrao.edu, even a short 1-2 pager.

Regarding Shami's comment - we can easily rearrange the WG's if desired. Once the three groups feel they have had their science discussions and there is a staw overall survey there should be a single WG to discuss this (maybe just have the Programmatic WG do it, or make a ScienceDefinition WG?).

-s

Note: the upshot of the SSG co-chair telecon yesterday is that we will continue discussions another week and the WGs will have a second draft next Monday.

chomiuk
Posts: 12
Joined: Fri Feb 28, 2014 4:16 pm

Re: Galactic WG Summary so far (comments please!)

Postby chomiuk » Tue Mar 25, 2014 2:10 pm

Here is an attempt to lay out particular source classes of interest, and how they might appear (*rough* estimate of spectral index, if the sources tend to be point-like or angularly extended). I stress that this is not a scientific justification, but I did try to be fairly inclusive in considering possible source classes of interest in the Milky Way (let me know if you think of others!).

The final row shows a figure-of-merit for each class of sources; higher numbers are better. The figure of merit was simply calculated as the primary beam area times the source brightness divided by an rms sensitivity achievable in a set amount of time. Figure of merit was then normalized to 15 GHz observations of flat spectrum sources.
Slide1.png

What you see is that in practically all situations, C band gives a much higher figure of merit than Ku band. It is only at the extreme of optically thick thermal sources that Ku competes with C band for efficiency. I assume by targeting thermal, we'd like to be inclusive of both optically thick and optically thin sources, right? Then C band really wins, just because of the larger primary beam (it is true that Ku band is more sensitive, but only marginally so---by ~10% or so, according to the exposure time calculator).

I have observed that very few optically thick thermal sources are truly characterized by nu^2 spectra. nu^1 or nu^1.5 are much more common, and of course for stars it's nu^0.6. So even in the case of optically thick-ish thermal sources, C band is the winner over Ku band.

So I ask again---what are the scientific problems that simply must be tackled at Ku band, given that C band gives a comparable or higher continuum figure-of-merit for all plausible spectral indices?

jlazio
Posts: 8
Joined: Thu Mar 20, 2014 11:13 pm

Re: Galactic WG Summary so far (comments please!)

Postby jlazio » Tue Mar 25, 2014 7:48 pm

Following up on Laura's post, I've been trying to figure out the answer to her question as well. It seems to me that there are two answers:

1. CORNISH exists. Thus, a 5 GHz Galactic plane component of the VLASS would have to argue that the increase in sky area and depth is sufficient to justify something like a "super-CORNISH."

2. Lines. The yield of lines near 15 GHz is much higher than at 5 GHz. What is less clear to me is that "more lines" equals "more or better science." Both 15 GHz and 5 GHz pick up a methanol maser, though with potentially different implications about the underlying massive star formation. Do the other lines produce a sufficient enough return to tip the scales in favor of 15 GHz rather than 5 GHz?

rosten
Posts: 19
Joined: Fri Feb 14, 2014 3:40 pm

Re: Galactic WG Summary so far (comments please!)

Postby rosten » Tue Mar 25, 2014 10:12 pm

Hi folks,

A note to add to our discussion: please take a look at the preliminary surveys being laid out by the other groups; they are available here:

https://safe.nrao.edu/wiki/bin/view/JVLA/TransientsVariabilityWorkingGroup

and

https://safe.nrao.edu/wiki/bin/view/JVLA/ExtragalacticWorkingGroup

Please look in particular at the extragalactic working group's survey definition. The extragalactic group especially is advocating for a 3 tiered approach, one of which is an all-sky survey at S band, with an rms sensitivity of 100 microJy/beam, at high resolution. I have attached the matrix describing this tier so you can quickly take a look. How does that fit with some of the discussions going on in this forum about a lower frequency survey? Can those parameters fulfill the science goals that you are advocating, or would they be achievable with some adjusting? Gordon Richards, the co-chair of that group, contacted me and Cornelia to see if there was momentum for discussing common ground, so I'd like to get the ball rolling.
Attachments
vlass_exgalwide.tiff

chomiuk
Posts: 12
Joined: Fri Feb 28, 2014 4:16 pm

Re: Galactic WG Summary so far (comments please!)

Postby chomiuk » Wed Mar 26, 2014 9:50 am

Great points, Joe.

Re Rachel and Joe's concerns about a Super-CORNISH: I think S band would be absolutely fine for the majority of Galactic sources, only coming up weaker than Ku or C band for nu^2 sources. See my new and improved plot below!
Slide2.gif

I am of the opinion that aligning with the extragalactic survey could accomplish much of the Galactic science we've been discussing. Of course, it's still more targeted towards point-like stellar sources than extended nebulae, but that seems to be the way our discussion has been converging for a long time.

If we are going to be primarily motivated by lines in our choice of observing frequency, we need to make it clear that we'll be going deep enough to detect interesting objects. As a person who does mostly continuum, my naive view is that we probably can't go deep enough to do all that much interesting line science, and I bet I'm not alone with this belief. So if line science is going to be central to our strategy, we need to show the skeptics that we can do great science with a wide-field survey! (again, specific science example, exposure time calculations, etc).

elisabethmills
Posts: 13
Joined: Tue Mar 11, 2014 3:58 pm

Re: Galactic WG Summary so far (comments please!)

Postby elisabethmills » Thu Mar 27, 2014 1:43 am

I can't contribute much because this is not my area of science expertise, and the little I know really just pertains to an interest in star formation, but I will at least try to give a brief defense of the reasons to go to higher frequencies in order to optimize the science done with thermal sources.

--Detecting optically thick HII emission sources which have not been systematically surveyed in our Galaxy.

To say that an S-band survey will be great for everything except exactly these sources for which radio surveys are least complete and we have the least knowledge seems to me to be missing an opportunity.
One specific class of these objects are hypercompact HII regions, whose population is poorly understood (at present,only a few tens are known) leaving a gap in our knowledge of this evolutionary stage of massive stars.
CORNISH had the sensitivity to detect all ultracompact HII regions in the region of the Galaxy it surveyed (here the plot shows different HII regions at 20 kpc distance), but not hypercompact HII regions:
Image
GUTS would be uniquely sensitive to hypercompact HII regions within 20 kpc; other surveys proposed and in progress would fall short (admittedly, I do not know what the spectral type of the HII regions shown in this plot is; if it is for average O stars, or the faintest B stars, so there could be some leeway). I also show a plot of an even earlier stage of massive stellar evolution, a nearby massive YSO:
Image
GUTS would marginally detect it at 3 sigma, other proposed surveys (including a survey sensitive to 100 uJy sources at S or L band in Shami's white paper) would not.

-- High frequency constraints for SEDs/ spectral index fitting. Presumably a survey is not just about detecting sources, but the information gained by detecting them at a particular frequency. Ku band fluxes should be a useful constraint for SED modeling of thermal sources, both for optically thick sources where it will help constrain the turnover, and for dust-dominated sources, at which these frequencies mark the transition where thermal dust emission begins to contribute. I only note that this would be new and unique information, I cannot say from experience how important it is.

-- Optimal sensitivity to and resolution of recombination lines. A continuum detection alone contains limited information; for thermal sources detecting recombination line emission gives you constraints on source kinematics and distance. Though there are slightly fewer lines in the band at higher frequencies, the line/continuum ratio increases at higher frequencies for most compact HII regions, which have N_e between 10^3 and 10^4, and log[emission measures] < 7.0 (plots of line intensity for densities 10^3 and 10^4 are shown here, as well as curves for different log[EM]):
Image

The lines are also less pressure broadened at higher frequencies:
Image
enabling more precise kinematics. This would then yield richer comparison data for non-radio observers looking for information on source counterparts in a VLASS catalog.

-- Stronger recombination lines would also mean one would be more sensitive to a larger/more distant population of HII regions for mapping Galactic structure. This is also an argument for observing a maser line (which is also being done in C-band with the arguably more optimal 6.7 GHz methanol maser).

-- Finally, if one does observe lines, at higher frequencies, there is less pressure on the correlator to observe more lines with higher resolution and over a sufficient velocity range, Though there are ~30 recombination lines available at S band, I am not sure how many one can realistically observe at a good resolution while still getting 8 GHz of continuum ---edit: sorry, just 2 GHz at S band, 4 GHZ at C band. It looks like at S band you could get ~24 of them, dual pol, at 1-2 km/s resolution, or all of them at 3-4 km/s resolution. So, this should not really be a concern. Stacking lines would be ~1.7 times more efficient at S than at Ku band, but the intrinsic brightness of the lines may still favor Ku. This also assumes that one does not need extra resolution on the continuum windows for better RFI excision, and that no recombination lines are lost to RFI, which I presume is not the case.

So, given that GLOSTAR:
Image
and THOR are in progress, my question is--- what are the science questions that cannot be answered from the completion of these surveys (and what makes a low-frequency survey of the plane, unlike these surveys, not best done as PI science?)

Finally of course, it is hard to make these arguments without knowing exactly what is being proposed as an alternative. I have yet to see the full numbers for the area, frequency, resolution and sensitivity of a high-resolution, low frequency survey. Given that is also not my science area, I could use some more details on the science drivers (other than that low frequency would be very useful, and one might find a juicy population of pulsars) and how the non-radio community will use it.

Personally, I would hope that one need not be so driven by the desire to create a monolithic survey that one cannot optimize the science unique to the Galactic plane (e.g., kinematics from lines, detecting new thermal sources, and the nearly impossible ISM tracers) but I recognize that my views are in the minority, and like everyone, influenced by my own science interests :)
Last edited by elisabethmills on Thu Mar 27, 2014 3:33 pm, edited 1 time in total.

rosten
Posts: 19
Joined: Fri Feb 14, 2014 3:40 pm

Re: Galactic WG Summary so far (comments please!)

Postby rosten » Thu Mar 27, 2014 1:58 pm

Hi Betsy,

Thanks for the detailed information. You can find the parameters of the all-sky high resolution low frequency survey being proposed in the extragalactic working group discussion forum, or by following the link I provided in my previous post. I also cut & pasted the matrix of frequency, area, etc. I don't think the idea is to create a single monolithic survey for VLASS, but each tier to the survey does need to be justified (and will be sent out for review), so to the extent that there is synergy with what other groups are proposing we should capitalize on that as it will strengthen the case.

That said, the fact that multiple tiers are being discussed in no way obviates the idea of having multiple components to the VLASS, especially one with unique science in the Galactic plane.


Return to “Galactic Working Group”

Who is online

Users browsing this forum: No registered users and 1 guest