Galactic counter proposal

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

Re: Galactic counter proposal

Postby chomiuk » Sun Jun 01, 2014 12:27 pm

rosten wrote:I'm not sure I buy the argument that going for a narrow part of the bulge with deep, high resolution is the best answer, as it is going to favor only a few source populations.

I agree that it focuses on old stellar populations. The argument for focusing on the bulge is simply that GLOSTAR has the rest covered. Not sure if this is a good argument or not :)

rosten wrote:o Cornelia and I had come up with a list of Galactic source types, their spectra, and are in the process of filling out the expected numbers for the sensitivity of the survey (100 muJy in the plane).

Is 100 muJy/beam really the best we can do? It sounded to me like 50 muJy/beam on the plane was still a possibility, and something we should consider fighting for. Could you include 50 muJy rms in your calculations too?

gsivakoff
Posts: 17
Joined: Tue Mar 25, 2014 11:28 am

Re: Galactic counter proposal

Postby gsivakoff » Sun Jun 01, 2014 7:22 pm

chomiuk wrote: rosten wrote:I'm not sure I buy the argument that going for a narrow part of the bulge with deep, high resolution is the best answer, as it is going to favor only a few source populations.


I agree that it focuses on old stellar populations. The argument for focusing on the bulge is simply that GLOSTAR has the rest covered. Not sure if this is a good argument or not :)


While I would love (love, love, love) to get deep images of the bulge, I don't think the VLASS is going to be the right place for this. If we assume my generous 280 square degree bulge (-14 < l < 14, -10 < b 10), that's about 35 hours per ~100 muJy scan using Steve's old numbers. Let's call it 50 hours for T_sys purposes. Getting down to 20 muJy rms would be about 1250 hours (plus overhead, so call it 1500 hours). This is well outside of the realm of a request that makes sense for VLASS given Tony Beasley's comments. For those not on the telecom, I interpreted Tony's comments as suggesting no chunk < 3000 hours. I do think our community could put together a wonderful bulge survey, but not as a part of VLASS.

gsivakoff
Posts: 17
Joined: Tue Mar 25, 2014 11:28 am

Re: Galactic counter proposal

Postby gsivakoff » Sun Jun 01, 2014 7:39 pm

claw wrote:Hi Everyone,
Thanks for putting your thoughts down again. I wanted to update you on the track the discussion has headed lately.
This coming Monday, we are expected to come together on a consensus and start doling out writing assignments. The plan now is for a pure S-band survey (in something close to B-config) like this:

-- All-sky (Dec > -40) to 100 microJy,
-- Wide (~1e4 sq deg aligned with DESI) to 50 microJy, and
-- Deep (10 sq deg) to 1.5 microJy.

This totals up to around 9000 hours, give or take. There is a Galactic option that requires extra justification. If the chairs are convinced, that time will be taken out of the wide tier. The sketch for a Galactic enhancement is a bulge survey in A config (2800 sq deg) to 50 microJy.
The sense I get is that some would rather have more Wide than to enhance the Galactic plane sensitivity. We need to make the case on Monday that there are concrete benefits from a 50microJy S band Galactic survey. Specificity is really what drives the extragalactic cases, so it would help if we could identify Galactic science lost if only covered to 100 microJy (or ~150 microJy considering spillover noise).
This consensus plan has a lot of momentum, so a decision is very likely on Monday. Any new thoughts here would help us make the pitch!

casey


So in other words, we have gone to a survey that is being dictated by the extragalactic science case. For a lot of the Galactic science, a 500muJy rms detection limit is not well motivated. As Shami might point out, this is not deeper than other telescopes have done; I think he mentioned 300 muJy detections as being a meaningful place to consider. My main interest would be for non-pulsar radio emitting compact objects. One that could be a large driver are quiescently accreting black holes.

The majority of quiescently accreting black holes would not be detected at 500 muJy detection limit. My rough numbers for 30 and 50 muJy rms are as follow:
The M22 sources (~50-60 muJy @ 3Mpc) would suggest we'd be looking for sources within 400 pc (150 mJy)/ 140 pc (275 mJy).
This covers roughly 1.5 and 0.5% of the mass in the Milky Way, and assuming 500 QXRBs, I'd predict we detect ~8 and ~2 sources, although identifying them may be difficult
I think these may not be the most radio bright examples of these sources. I suspect there is a similar source that is about 2-3 times as luminous (paper in progress).
This pushes us out to 260 pc and 700 pc (or about 1 and 2.5% of the Milky Way mass), but cuts down on the population. However, if that source is a BH, it could suggest a much, much larger number of QXRBs. Unfortunately, this is work in progress and I won't get the paper done by tomorrow.

Here are some other numbers I've forwarded to Rachel in the past, again looking at 30 and 50 muJy rms, not 100 or 150 muJy rms. (I hope this is a little useful.)

Black hole X-ray binaries (all are thought to be transient. thought some have had decade long outbursts)

Galactic

L_X>1E36 erg/s in outburst sources should be no problem at either 150 or 275 mJy detections.
E.g., H1743-322 was routinely above 1 mJy during its hard state in its 2009 outburst.
Typical outburst times of 2-4 weeks over an entire outburst are reasonable.
Given that there are ~ 6-10 of these a year, I'll assume 5/year and a month outburst in the VLASS footprint.
In a one pass per year survey design, we're looking at detecting about 2-3 of these over five years.
In a one/fifth survey area per year design, we'll detect them well when on Galactic areas

Quiescent black hole X-ray binaries may be an interesting source population.
The M22 sources (~50-60 muJy @ 3Mpc) would suggest we'd be looking for sources within 400 pc (150 mJy)/ 140 pc (275 mJy).
This covers roughly 1.5 and 0.5% of the mass in the Milky Way, and assuming 500 QXRBs, I'd predict we detect ~8 and ~2 sources, although identifying them may be difficult
I think these may not be the most radio bright examples of these sources. I suspect there is a similar source that is about 2-3 times as luminous (paper in progress).
This pushes us out to 260 pc and 700 pc (or about 1 and 2.5% of the Milky Way mass), but cuts down on the population.

Extragalactic

L_X>1E39 in outburst will be a problem for the 275 mJy level.
The M31 microquasar was ~300-400 muJy, placing most galaxies as being too far away.
We gain a little at the 150 mJy level, but as this is for Galactic areas, it is a non-driver for this.
These are likely to be rare, but very interesting sources.

Neutron Star X-ray Binaries (most are persistent, although some are transient)

Galactic

L_X>1E36 erg/s neutron star X-ray binaries in outburst
Near the Galactic Centre these may be clearly detectable at ~ 200 - 400 muJy (e.g., AC 211 in M15 at 10.3 kpc, although this is a persistent source that is likely bright and seen edge on).
Aql X-1 (~5 kpc) may be a better comparison. The 150 limit would have detected these for about a week each in the rise and decay state, while the 275mJy would miss the decay state.
In a one pass per year survey design, we won't detect these sources (assuming 500 muJy detection limit)
In a one/fifth survey area per year design, we won't detect them as well as BH XRBs.

Extragalactic
Unlikely in the extreme!!

Cataclysmic Variables

Galactic

Nearby Dwarf Novae
A good number to use is that SS Cyg in peak was 1-3 mJy at ~ 114 pc, and more typically ~ 200 muJy during outburst.
But its outburst last about a week. So, I imagine we're looking at single-pass senitivities...
At 500 mJy rms detection, I imagine we may get Dwarf Novae out to ~ 150 pc.
The number of nearby DNe would have to be calculated, but I think I could ask Boris Gaensicke for his input.

SNR

Galactic

These will need to have a pretty high surface density to be detected.
I'm going to be thinking about simple box tapers and their sensitivity for B array.
My guess is we're looking at about a factor of two sensitivity cut for a factor of 4 bigger beam.

Extragalactic

These may be an interesting population that were never mentioned in the Galactic Working group, but I think of as being more strongly linked scientifically to "Galactic" science.
Based on Laura Chomiuk 2009 paper, these will be detected in the Local Group; I'm just not sure what we'll learn that is new.

claw
Posts: 26
Joined: Mon Mar 03, 2014 7:18 pm

Re: Galactic counter proposal

Postby claw » Sun Jun 01, 2014 8:48 pm

Hey Everyone,
I finished my estimate of the extra time needed to get uniform sensitivity at negative declinations. The principle effect I consider here is larger Tsys from spillover at low elevations. This does not consider confusion or sky Tsys effects, which are generally small except in the inner degree or so (I think).
This calculation is at the bottom of the VLASS Calculator notebook at http://goo.gl/jHEPdQ. The tl;dr:
-- Tsys scaling requires us to ramp up integration time to 3x nominal at -40 Dec (NVSS ramped up to 2x, so we're in rough agreement).
-- For sky below Dec=0, the total integration time scales up by a factor of about 1.6.

This would not be quite the right number for a plane-shaped survey region, since I assume we cover all RA at a given Dec. But the simplest argument that we need to make is for uniformity (after all, NVSS did it!) and this is the scale factor for negative Decs.

casey

claw
Posts: 26
Joined: Mon Mar 03, 2014 7:18 pm

Re: Galactic counter proposal

Postby claw » Mon Jun 02, 2014 1:00 am

And one more refinement to the calculation above...
Including the scaling at high and low Declination, plus integrating over the whole sky down to Dec=-40, the enhancement needed to get uniform sensitivity is 14%. 12% of that is at low Dec, 2% is at high Dec (starting around +80 deg).

casey

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

Re: Galactic counter proposal

Postby shami » Mon Jun 02, 2014 8:14 am

For archival purposes: this is the substance of two emails I sent out on June 1st.

Laura Chomiuk said:
**Pulsars** Shami et al.'s white paper on pulsars/related exotica requests 5 sigma sensitivity of 100 muJy (How certain is the team that this is a good number? How well can it be backed up?). An rms sensitivity of 20 muJy/beam would really only be possible over a relatively small area, say with a focus on the Galactic bulge (how much area around the Galactic bulge would the pulsar people really need?) 30 muJy/beam may be possible over a much larger swath of the Galactic plane. Also requests multiple epochs.

We laid out the science case for pulsars in our white paper in some detail, but here's a very terse recapitulation:

- Pulsars are interesting to find for many reasons - among others, all PTA gravitational wave detection estimates assume that we're going to continue finding "good" millisecond pulsars at a steady rate.

- Blind (singe dish) surveys are very successful, but have some specific weak spots:
- Highly (pulse) scattered sources can only be detected at higher frequencies, but time domain surveys are prohibitively expensive (smaller beams); imaging surveys are minimally affected by angular scatter broadening.
- Highly accelerated sources (NS-BH or NS-NS binaries) are very hard to detect, especially at very short orbital periods (<1 hr, say) in standard surveys. A candidate selected from imaging can be followed up specifically. (See Jim's nice figure in the white paper)
- Very interesting intermittent objects will often be missed by one-pass time domain surveys. Even for imaging surveys, two or more passes would be very interesting for these objects, among other "slow" transients.
- RFI reduces sensitivity in visible and not-so-visible ways (crowding out real pulsars with fake candidates).

Joe Lazio pointed out that we weren't being very quantitative so far in bold predictions of the numbers we would detect. I did numbers for a pilot field (10 sq deg) in the plane: "Above 100 microJy, about 10^3/deg^2 extragalactic sources are expected (Hopkins et al. 2003; Bondi et al. 2008) compared to ~8 pulsars/deg^2 out of a total ~200/deg^2 within the volume extending through the Galactic disk (Lorimer et al. 2006a). The total number of expected pulsars expected in our 10 deg^2 survey fields is ~80 compared to 28 known objects (none of which are MSPs; the fastest has P_spin = 45 ms)."

This was for a specific pilot field and an L-band survey. At S-band our targets are fainter, but we can probably do better at identifying point sources in images. We'd be happy enough with 30 microJy if it meant getting a much larger area of the Galactic plane.

However...

Even if we were to just converge on a simple uniform all-sky survey, a nominally uniform survey does *worse* on the Galactic plane due to Tsys increase as well as source confusion. To approach true uniform sensitivity, there should be a better nominal sensitivity in the Galactic plane, and preferably higher resolution as well.

So a *minimal* request might be to add A or BnA array observations of the plane, especially at lower Dec, to boost sensitivity and add higher resolution. This would be just a "let's not have the Galactic plane do worse than the rest of the sky" request.

(Casey estimates 2-3x extra integration on up to 14% of the survey area - see his post.)

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

Re: Galactic counter proposal

Postby jlazio » Mon Jun 02, 2014 4:24 pm

Thinking ahead to the actual proposal, what do we need? Using a couple of examples from ones that people have posted, I could imagine a structure such as the following.

Section: Galactic Science Enabled by the VLASS

The Galactic component would consist of H hours focused on the area ... in the A configuration to obtain approximately 1" resolution.

Pulsars
We expect to detect N (~ 10,000?) pulsars, of which 10--30 will be "exotic" objects such as hierarchical triples, sub-millisecond pulsars, or tight binaries.
[Paragraph or two summarizing the reasons why a high resolution survey enables the identification of unique objects, which comes from the Chatterjee et al. white paper.]


Symbiotic Stars
We expect to detect N symbiotic stars, which are potential Type Ia SNe progenitors, and for which current surveys are biased because they are typically identified by optical emission lines, which are subject to significant obscuration through the Galactic plane.
[Paragraph or two describing the yield of symbiotic stars, to what distance, etc. Laura has already written?]


Novae
We will conduct an unbiased survey to a distance of D kpc, and anticipate detecting N novae. Radio emission from novae tracks the mass loss from the progenitors, but detecting distant ones is biased because of obscuration effects in the Galactic plane.
[Laura has already written a couple of such paragraphs?]

gsivakoff
Posts: 17
Joined: Tue Mar 25, 2014 11:28 am

Re: Galactic counter proposal

Postby gsivakoff » Tue Jun 03, 2014 3:46 pm

I'm willing to write up something about quiescent black hole X-ray binaries. I'll see if Tom Macarrone is willing to help.

jlazio wrote:Section: Galactic Science Enabled by the VLASS

The Galactic component would consist of H hours focused on the area ... in the A configuration to obtain approximately 1" resolution.


I also think we should think about Galactic components enabled by the other survey tiers.


Return to “Galactic Working Group”

Who is online

Users browsing this forum: No registered users and 1 guest