Galactic Survey science case

lsjouwerman
Posts: 21
Joined: Mon Jul 18, 2011 3:49 pm

Re: Galactic Survey science case

Postby lsjouwerman » Mon Mar 17, 2014 1:13 pm

Simon, I think you confuse Ka and K band. Note that Ku band includes the 12.2 GHz methanol maser line which is as interesting for SFRs as water masers. I agree that doing methanol (either K or Ka band) would be great, but I disagree to design a multi-khour survey to do new science - you would do that in a pilot. You would want to base the survey on clear science goals or results to get it approved and look between the visibilities later for the new science. Unfortunately the higher bands are less sensitive and require high frequency weather, which are logistical hurdles.

To Laura,
Many of the questions you have have been addressed in the white papers, eg in GUTS (https://science.nrao.edu/science/survey ... _WP_r0.pdf) and the other topics. In the white paper we neglected to address the transients there but that was corrected for in the workshop mash by Betsy (https://science.nrao.edu/science/survey ... SSmash.pdf). The main science case for the higher frequencies is to pick up more of the thermal emission and make the connection to the IR surveys at 1" resolution - something that is lacking so far. Redoing much of the existing non-thermal low-frequency surveys with some extra lines or pulsar searches just because the survey speed is larger there would - in my view - not be the proper justification and requires A-array for 1".

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

Re: Galactic Survey science case

Postby elisabethmills » Mon Mar 17, 2014 3:06 pm

sellings wrote:my personal vote would be for the doing a smaller area (say centred around l = 30) at around 22 GHz since there are more lines, the thermal emission is stronger and with the increased sensitivity we could get the first unbiased census of hyper-compact HII regions, water masers and a range of other interesting star formation tracers (e.g. class I methanol masers at 25 GHz which are relatively common, but for which there is little observational data).


Three questions:

#1- Exactly how small an area would you propose to do at K Band? Keep in mind that anything less than 2000 hours of time can theoretically be proposed for in the context of a large program
#2- Related to this, what do you estimate the legacy value of a small and deep Galactic field to be for the community? Who (especially those outside of the radio community) would continue to use these observations, and for what purpose?
#3- Finally, I wonder whether some of the science goals you list here are not better served with a proposed larger-area Ku-band survey. Hypercompact HII regions are thus far extremely rare, and it seems that a larger area search (which is still at high enough frequencies that these sources are less optically thick) would provide much better statistics. Is K band at 18-26 GHz, with its higher overhead, smaller FOV, and dependence on good weather, really a game changer compared to Ku at 12-18 GHz for continuum studies of these sources?

beuther
Posts: 1
Joined: Tue Mar 18, 2014 7:57 am

Re: Galactic Survey science case

Postby beuther » Tue Mar 18, 2014 8:24 am

Sorry, for chiming in so late. Regarding some more details of the THOR project, please have a look at http://www.mpia.de/thor

The main idea of THOR is to map the plane between longitudes of 15 and 65 degree and latitudes +-1deg in the C array at 15''-20'' resolution in HI, OH and RRL line emission as well as the continuum from 1-2GHz. For the HI data, complementary D-array and GBT observations exist and we have all spatial scales. For the other frequencies, the complementary data do not exist and we work with the C-array data alone.

Regarding extensions of that, one could think of many directions, e.g., extending the longitude/latitude coverage or adding other arrays to have a broader range of scales covered. While all that would certainly be very interesting, I must admit that in the framework of VLASS, I would probably go for the higher frequencies in the Ku band as well. The low frequencies will at some point be done at much greater depth with the SKA, but a Ku band survey could have VLA stand-alone quality also in SKA times.

sellings
Posts: 2
Joined: Sat Mar 15, 2014 3:37 am

Re: Galactic Survey science case

Postby sellings » Wed Mar 19, 2014 6:02 am

elisabethmills wrote:#1- Exactly how small an area would you propose to do at K Band? Keep in mind that anything less than 2000 hours of time can theoretically be proposed for in the context of a large program

Just doing some quick and dirty calculations I estimate that its about 4 hours per square degree per second of integration time and if we consider an RMS of 40 microJy as a desirable sensitivity limit then that would be 75 seconds per pointing or 300 hours per square degree. I think you would want 10+ square degrees to give a representative sample of the plane.

elisabethmills wrote:#2- Related to this, what do you estimate the legacy value of a small and deep Galactic field to be for the community? Who (especially those outside of the radio community) would continue to use these observations, and for what purpose?

I think that it would be of broad benefit to the broader star formation community identifying target regions for ALMA, helping with SED modelling for lots of Galactic sources, particularly those with thermal spectra and providing general complementary data. The real utility would depend on the lines that could be co-observed with the continuum (see below).

elisabethmills wrote:#3- Finally, I wonder whether some of the science goals you list here are not better served with a proposed larger-area Ku-band survey. Hypercompact HII regions are thus far extremely rare, and it seems that a larger area search (which is still at high enough frequencies that these sources are less optically thick) would provide much better statistics. Is K band at 18-26 GHz, with its higher overhead, smaller FOV, and dependence on good weather, really a game changer compared to Ku at 12-18 GHz for continuum studies of these sources?

For the continuum alone probably not, I agree that the better system performance at Ku band outweighs the arguments for higher frequency. 22 GHz is only superior if you can do a reasonable number of spectral lines simultaneously with the continuum, as the K-band selection is much better than those at Ku band. I do think that there are more deeply embedded sources which can be detected by sensitive K-band observations (not just hyper-compact HII regions) than is generally appreciated, although I admit that it requires some work to plausible argue that that is the case rather than just going on my gut feeling.

elisabethmills wrote:Ku (14-18 GHz) band offers 12 GHz methanol masers and absorption of the 14 GHz formaldehyde lines against the continuum, but if you want a (relatively) strong ISM tracer, the next best bets aside from ammonia are HC3N 1-0 at X band, and CH in S-band (the latter frequency not really being ideal for detecting thermal sources in the plane).

With the Menten et al. survey at C band including a sensitive search for 6.7 GHz methanol there is no point looking for 12 GHz masers in a blind survey. There are ~1000 6.7 GHz methanol masers known to date and in all cases the integrated intensity of the 6.7 GHz emission is greater than the 12 GHz, and the peak intensity is nearly always greater. So I wouldn't expect to detect any sources not previously found from that and other surveys, so a targeted followup for 12 GHz masers source be efficient.

So if it isn't practical to do 4 or more spectral lines along with 8 GHz continuum bands then it probably isn't worth going to that 22 GHz.

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

Re: Galactic Survey science case

Postby rosten » Thu Mar 20, 2014 10:33 am

This is great discussion; thanks everyone! I see the discussion converging on two possibilities: a small, 10 square degree survey at K band, but which might only be practicable if one can do 4 spectral lines with 8 GHz continuum bands in one go. Does anyone have experience with observing at this frequency who can answer that question? The other possibility would be a wide survey at Ku band, trying to catch continuum plus some lines, although it seems like there is debate about the utility of those lines (e.g. methanol) compared to lower frequency transitions. We need to be thinking about legacy value, and the overlap of the wider Galactic plane at Ku band with surveys at other wavelengths seems like a strong selling point.

At least one of the initial white papers mentioned the Galactic center -- is there a strong argument to treat this part of the galaxy differently (either deeper or a wider range in b)? And what about extensions to star forming regions that aren't that far out of the plane? I haven't heard any discussion on these last bits.

tjmaccarone
Posts: 2
Joined: Thu Mar 20, 2014 9:49 pm

Re: Galactic Survey science case

Postby tjmaccarone » Thu Mar 20, 2014 11:01 pm

Hi, everyone,

Sorry for being late to the discussion. I do want to echo some of Laura's comments. I realize that there is a transient working group which covers some of the compact object science, but there is a lot of crucial compact object science that is not based on transient behavior. The survey Lorant proposed could do a fair bit of the compact object work, but there are some additional shortcomings of that project for compact object work that weren't raised there. These issues would motivate some combination of deeper data, or use of lower frequencies, or both.

1) Sensitivity to pulsars.

I think probably nearly everyone here realizes that pulsars are extremely steep spectrum sources, and that essentially none will be detected at 12 GHz, and just figures that that is why we have the GBT. There are a few major advantages to finding pulsars with the VLA. The first is positioning, which can be done by building up timing solutions with the GBT, but is difficult. The second is sensitivity to pulsars in environments which have very high foreground free electron densities, and hence have their pulses dispersed away or scattered out from being detectable via timing. The third is that binary pulsars with unusual orbits present computational problems for pulsar search algorithms. By unusual orbits, what I mean are short period eccentric orbits. These are the systems most likely to be double compact objects with relatively short (relative to the Hubble time) timescales before they merge.

2) Sensitivity to black hole X-ray binaries in quiescence

At the present time, we know of about 20 dynamically confirmed stellar mass black holes. None of these eclipses. The mass distribution of the black holes is a crucial indicator of just how the explosion mechanism in supernovae works. The lack of eclipsers means that we do not have the most precise mass estimates possible for any black hole candidates, since the inclination angles right now need to come from ellipsoidal modulations of the donor stars. Because the ellipsoidal modulations are diluted by poorly understood disk emission in quiescence, most of the measurements have large poorly understood systematics, which eclipsing systems would not. At the present time there appears to be a mass gap between the neutron stars and the lowest mass black holes, and while the qualitative finding is probably robust, the quantitative details are extremely important for understanding why core collapse supernovae blow up.

There are also other major selection effects on the population of black hole binaries that can be reduced, if not entirely eliminated by radio detections in quiescence. For example, the orbital period distribution of black hole X-ray binaries is very different from that of cataclysmic variables. A recent paper by Gillian Knevitt showed that this is likely to result from the selection effects imposed by all sky X-ray transient monitors. Surveys for quiescent objects will still have some bias toward longer period systems because they tend to have higher mass transfer rates, but the bias will be much weaker.

The black hole X-ray binaries in quiescence should stand out as flat spectrum sources. Some AGN do this as well, but if we also have good multi-wavelength coverage, it should be straightforward to cull the list to something manageable for spectroscopic follow-up. On the other hand, if we do not have the radio data, the black hole X-ray binaries end up lost in a sea of cataclysmic variables. It takes excellent optical data (i.e. multi-epoch spectroscopy, which is expensive for such bright targets, and shouldn't be done with low hit rates), or merely a radio detection in quiescence, to separate out these possibilities.

Lorant's survey would be sensitive to these objects out to only about 700 pc, if we take the nearby, well-studied object A0620-00 as a guide. Within that volume, sticking only to the part of the sky projected against the Galactic Plane, we probably will not detect any new black hole binaries. The 10 square degree field of view would likely only detect a good number of sources if it were aimed at specific regions where they are expected to be found.

What I think would help a lot for the compact object science, and would partially deal with the pulsar issue, as well, would be to focus a deeper, lower frequency, A-config tier of the survey. Even if this were restricted e.g. to cover only Galactic fields with Chandra data already in the archive, it would be of major value.

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

Re: Galactic Survey science case

Postby elisabethmills » Fri Mar 21, 2014 1:07 am

In regards to studies of compact objects, I have a couple of questions and comments

First, there is some talk of a wide or all-sky component of VLASS in the extragalactic group which would potentially be at S-band, A or B configuration, and having an RMS sensitivity of some tens of microJy (35-50, likely less sensitive for all-sky). Would this be better optimized for this science?

Second, as in the extragalactic discussion, I am curious whether there are any other proposed experiments (EMU?) with SKA pathfinders which will be optimized for this science. One strength of doing a Ku survey is of course that it exploits a unique capability of the VLA.

I also do not want to neglect Sanjay's proposed L-band survey (Bhatnagar et al. white paper) in these discussions, though its design (D+C configurations) was optimized for extended emission sources and not for the point sources mentioned here.

So, what specific design recommendations would you make for a survey of compact objects (minimum sensitivity, resolution and areal coverage, optimal frequency)?

lsjouwerman
Posts: 21
Joined: Mon Jul 18, 2011 3:49 pm

Re: Galactic Survey science case

Postby lsjouwerman » Fri Mar 21, 2014 11:33 am

I personally strongly dislike to use the VLASS opportunity for yet another Galactic survey at a synchrotron-biased observing frequency by limiting the scope to the single needs of a smaller group. The absolute strength and legacy value are in the higer frequency ranges and serves the whole non-pulsar Galactic community to obtain something that is absolutely unique and otherwise unavailable - and probably will never be considered again. Note that Ku band is the winner in instantaneous continuum sensitivity of all VLA bands. The low frequency pulsars and sensitivity for X ray binaries (at low freq) can be done with other instruments (eg SKA), so I would argue to use the VLASS opportunity to propose for something that is not in the pipeline or possible anywhere else.

tjmaccarone
Posts: 2
Joined: Thu Mar 20, 2014 9:49 pm

Re: Galactic Survey science case

Postby tjmaccarone » Fri Mar 21, 2014 3:52 pm

elisabethmills wrote:In regards to studies of compact objects, I have a couple of questions and comments

First, there is some talk of a wide or all-sky component of VLASS in the extragalactic group which would potentially be at S-band, A or B configuration, and having an RMS sensitivity of some tens of microJy (35-50, likely less sensitive for all-sky). Would this be better optimized for this science?

Second, as in the extragalactic discussion, I am curious whether there are any other proposed experiments (EMU?) with SKA pathfinders which will be optimized for this science. One strength of doing a Ku survey is of course that it exploits a unique capability of the VLA.

I also do not want to neglect Sanjay's proposed L-band survey (Bhatnagar et al. white paper) in these discussions, though its design (D+C configurations) was optimized for extended emission sources and not for the point sources mentioned here.

So, what specific design recommendations would you make for a survey of compact objects (minimum sensitivity, resolution and areal coverage, optimal frequency)?


OK -- for the pulsars, the answer is "the lower the better" for the optimal frequency, and there is such an enormous range in pulsar luminosity that, at least my opinion, would be that this shouldn't drive anything other than having a low frequency component to the survey if the pulsar work is deemed to be important.

For the X-ray binaries -- there is not really an optimal frequency. The non-transients are flat spectrum objects (maybe slightly inverted in some cases, but not enough to really matter much). So the optimization is on solid angle covered at a particular flux density. Things will be a factor of a few faster at low frequency (whether 1-2 GHz or 4-6 GHz) to cover the same solid angle to the same depth as at 12 GHz, The survey Lorant has proposed would be a little too shallow to detect much. It would detect objects like V404 Cyg, the brightest quiescent black hole X-ray binary, out to about 1 kpc. We expect ~a few black hole X-ray binaries within that volume and for none of them to be that bright.

Regarding other projects besides the VLA -- EMU's angular resolution is inadequate both for counterpart identification and for dealing with confusion in regions like the Galactic Center and inner Bulge. MeerKAT may eventually do some of this, but none of the key projects are set up for it.

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

Re: Galactic Survey science case

Postby shami » Mon Mar 24, 2014 5:28 pm

tjmaccarone wrote:OK -- for the pulsars, the answer is "the lower the better" for the optimal frequency, and there is such an enormous range in pulsar luminosity that, at least my opinion, would be that this shouldn't drive anything other than having a low frequency component to the survey if the pulsar work is deemed to be important.


I left a comment about this on the other thread as well - there's a strong case for a lower frequency, higher resolution survey of the Galactic plane for pulsars and radio transients, as laid out in the original VLASS white papers ( https://science.nrao.edu/science/survey ... ite-papers ). Independent of transient or neutron star science, "looking up a counterpart to my optical / IR / X-ray source" is a very common use case for a legacy survey, and the NVSS has neither the sensitivity, nor the resolution, to be effective in the plane for this purpose.


Return to “Galactic Working Group”

Who is online

Users browsing this forum: No registered users and 1 guest