A challenge to all : Read a Scientific Paper

[Hale_Bopp]

I have mentioned before that I would like to see who could read and debate the contents of a scientific paper rather than a press release. I am betting the percentage on this board would be much higher than the general population.

With that in mind, I have selected a paper to read and debate. It is short (less than 4 pages) for a scientific publication and one of the more accessible papers I have read.

The paper announces the possible detection of an "orphan" gamma ray burst afterglow. Normally, gamma ray bursts (grbs) are detected by satellites and follow up observations detect an optical afterglow. In this paper, they argue that a bright optical transient is a grb afterglow for which now gamma rays were detected (no satellite was observing that part of the sky at the time).

I also like this paper because there is no proof of what they found. They outline their methods and reasons for the conclusion. Let's have at it and see what we think here!

The paper can be downloaded as a .pdf file at

http://www.arxiv.org/PS_cache/astro-...11/0111054.pdf

Have at it!

Rob

[GrapesOfWrath]

I can. Here, read this one: Who wants to be a millionaire?

[Donnie B.]

There's one thing I read differently from your description above. I don't think they're saying the GRB (if indeed it was one) was not observed because no gamma ray telescope was pointed in its direction. I think they're saying it wasn't observed because the gamma rays are emitted in a tight beam that didn't happen to be aimed at the pale blue dot.

The most interesting implication is that the source of the burst emits its optical afterglow in a less tightly beamed manner than the actual burst. Thus we may be able to see a lot more of these events if we look for the optical afterglow rather than the gamma rays themselves.

One question comes to mind: are the GRB's that we do observe actually occuring in the time frame we see them, or are we just seeing the beam sweep by, and the burst itself is longer-lasting? (They still must be fairly short-duration, or we would see two or more pulses sometimes as the object rotated "our way" again.)

[Wiley]

On 2002-01-16 07:50, GrapesOfWrath wrote:
I can. Here, read this one: Who wants to be a millionaire?

Sorry GoW, I'm already working on this one.
[img]/phpBB/images/smiles/icon_smile.gif[/img]

<font size=-1>[ This Message was edited by: Wiley on 2002-01-16 17:47 ]</font>

[DoctorDon]

On 2002-01-16 16:18, Donnie B. wrote:
There's one thing I read differently from your description above. I don't think they're saying the GRB (if indeed it was one) was not observed because no gamma ray telescope was pointed in its direction. I think they're saying it wasn't observed because the gamma rays are emitted in a tight beam that didn't happen to be aimed at the pale blue dot.

That's correct. If no Gamma-ray detector
happened to see it, it would be an
"untriggered" GRB. "Orphan" afterglows are
those bursts where the higher-energy
radiation misses the earth.

The most interesting implication is that the source of the burst emits its optical afterglow in a less tightly beamed manner than the actual burst. Thus we may be able to see a lot more of these events if we look for the optical afterglow rather than the gamma rays themselves.

We're trying. No luck so far. It's a tough
job. I'm a co-author on a paper on that very
subject that is nearing completion as we
speak.

One question comes to mind: are the GRB's that we do observe actually occuring in the time frame we see them, or are we just seeing the beam sweep by, and the burst itself is longer-lasting? (They still must be fairly short-duration, or we would see two or more pulses sometimes as the object rotated "our way" again.)

This is what I wrote my phd thesis on...

They don't rotate. They explode. According to the "fireball" model, which is the most
popular model these days, the energy from
the explosion drives the (relatively small)
amount of matter out in a narrow (<~30 deg)
jet at speeds upwards of 99.999995% the speed
of light. It emits no radiation at this
point; all its energy is in the bulk motion
of the ejecta. However, once it collides,
the shock that forms from the collision slows
the matter down, and the bulk flow energy is
converted into turbulence, which then
radiates. So the emission we see as the
burst is actually being emitted quite some
ways from the thing that exploded. When
bursts show multiple peaks and erratic
emission, that is interpreted as being caused
by multiple shells of ejecta catching up and
colliding with each other, while the long-
lived afterglow is believed to come from the
collision of the total ejecta with the stuff
around the star.

So they don't "sweep by" in a "sideways"
sense. They just sweep by like a wave
passing over us. In fact, it's been
suggested that if we want to send a message
to aliens, we should send it right after a
GRB, but in the opposite direction, so that
if an alien species lives along that line
of sight, and happens to be studying GRBs,
they will already be looking at the right
spot on the sky when our message arrives.
Inversely, a civilization further "upstream"
from us might send us a message that way.
But, back to the grbs themselves.

Hence, the long-lived emission from the
external shock is actually highly compressed
in time in our frame, since the emitting
region is barely behind the radiation front.
A year's worth of emission would arrive here
in about ten seconds. For the internal
collisions that cause the erratic grbs
themselves, that's played out more or less in
real time.

Relativity demands that emitters moving at
high speeds beam their radiation in the
direction of motion, and the beaming is
tighter the faster you go. So as it sweeps
up external gas and slows down, it will
beam out to a wider angle. It will also
emit at lower frequencies and be dimmer.
The latter makes them very hard to see.
Theorists are predicting that your best
chance to see orphans would be at something
like 21st magnitude, around 1 day after the
onset of the event. Very hard to do. In
order to get down to 21st, you usually have
to look at a smaller patch of sky, which
reduces your chance of seeing a burst quite
a bit, since they come from random locations.

But that's part of our project's goals...

Yours,

Don Smith
http://www.rotse.net

[Donnie B.]

Thanks, Dr. D. I should have realized we weren't talking about a rotating object (necessarily).

Perhaps this is a bit too arcane for my understanding, but what's the current thinking on why the GRB forms a beam? Your discussion above doesn't indicate why the shock would be produced along a single axis rather than spherically. Is there a straightforward explanation? Or is that question still being debated?

[The Bad Astronomer]

On 2002-01-16 23:29, DoctorDon wrote:

A year's worth of emission would arrive here
in about ten seconds.

You know, that never occurred to me, but it's obvious to me now that you mention it. This really eases up on the emitted energy budget! The total is still the same, but the amount emitted per second drops. So really, the total energy emitted per secdon in the rest frame of the GRB can be, say, thousands of times less than what we actually see on Earth. That's very interesting.

[DoctorDon]

On 2002-01-17 08:47, Donnie B. wrote:
Thanks, Dr. D. I should have realized we weren't talking about a rotating object (necessarily).

Well, the underlying object is very likely
rotating, but the *emission* that we *see*
as a GRB is coming from far away from that
object (at least 10^15 cm), and would not be
rotating.

Perhaps this is a bit too arcane for my understanding, but what's the current thinking on why the GRB forms a beam? Or is that question still being debated?

Oh, very hotly. One idea, which I heard
about through Andrew MacFayden and Stan
Woosley, is the following: if the GRB
progenitor is a collapsing massive star
(akin to a supernova; and there have been
GRBs linked with supernovae), then you
would expect that material in the star's
mantle would tend to fall into the core
more easily along the poles, since there
wouldn't be a centrifugal barrier to overcome.

The density along the poles would therefore
be much less than in the torus around the
equator, so when the explosion (of whatever
origin) happens deep down in the center, it
will be more likely to spew stuff out the
poles because there's less resistence than
there would be along the equator where all
the junk is in the way.

That's one idea being explored.

Yours,

Don Smith

[Valiant Dancer]

On 2002-01-15 21:13, Hale_Bopp wrote:
I have mentioned before that I would like to see who could read and debate the contents of a scientific paper rather than a press release. I am betting the percentage on this board would be much higher than the general population.

With that in mind, I have selected a paper to read and debate. It is short (less than 4 pages) for a scientific publication and one of the more accessible papers I have read.

The paper announces the possible detection of an "orphan" gamma ray burst afterglow. Normally, gamma ray bursts (grbs) are detected by satellites and follow up observations detect an optical afterglow. In this paper, they argue that a bright optical transient is a grb afterglow for which now gamma rays were detected (no satellite was observing that part of the sky at the time).

I also like this paper because there is no proof of what they found. They outline their methods and reasons for the conclusion. Let's have at it and see what we think here!

The paper can be downloaded as a .pdf file at

http://www.arxiv.org/PS_cache/astro-...11/0111054.pdf

Have at it!

Rob

Ok. I've read it.

Computer programmer synopisis follows:

Something in a fairly far part of the universe (no reference as to how far that I could decypher) went BOOOOOOOMMMMM!!!!!!!! Since the after glow died out relatively quickly, a type 1a Supernovae is ruled out due to the burst die off lasting longer in that case. Accretion disk collisions in a supermassive BH is likewise ruled out due to various causes including duration, energy expended, etc. Therefore, the idea is that it was a GRB afterglow. One that was not dectected because detection devices weren't pointed that way or angle was too great. (The only GRB event recorded near the time in the general vicinity was a unlocalized event March 16th, 1999. Date of the observance was March 20th, 1999)

My head hurts.



Stated problems:
Equipment was not designed to look specifically for GRB's or any other short period object.

Other ideas (mine):
Although the phenomenon they state could very well be the most likely cause, perhaps this is the GRB afterglow from SMBH creation. (if the galaxy from which it eminated is far enough away.) (probably not, but an idea.)

[Donnie B.]

This is really cool, being able to chat with somebody who's really involved in ongoing astronomical research in such a hot area (pardon the pun).

Do we have enough data yet to figure out just how "beamy" these events are? That is, would the signal at the equator be down 20, 50, 100 dB from at the poles? And how narrow is the beam (a related question, of course)?

[Donnie B.]

On 2002-01-17 14:20, Valiant Dancer wrote:
Other ideas (mine):
Although the phenomenon they state could very well be the most likely cause, perhaps this is the GRB afterglow from SMBH creation. (if the galaxy from which it eminated is far enough away.) (probably not, but an idea.)

I think the SMBH idea is less likely because the redshift isn't that large (.38 or so, IIRC) so we're not talking about something at the far end of the cosmos. Unless Keirein and company are right, of course...

[Valiant Dancer]

On 2002-01-17 14:25, Donnie B. wrote:

On 2002-01-17 14:20, Valiant Dancer wrote:
Other ideas (mine):
Although the phenomenon they state could very well be the most likely cause, perhaps this is the GRB afterglow from SMBH creation. (if the galaxy from which it eminated is far enough away.) (probably not, but an idea.)

I think the SMBH idea is less likely because the redshift isn't that large (.38 or so, IIRC) so we're not talking about something at the far end of the cosmos. Unless Keirein and company are right, of course...

Keirein and company are about as likely to be right as I will be likely to convert to fundamentalst Christian. I didn't know that red shift was related to distance. (Remember, I'm a computer geek.) With that new information (for me), I'd have to say that the writers of the paper have presented the most likely explaination for the observations.

[Donnie B.]

Doc D,

If my math is right, a redshift of .385 corresponds to about 1.3 Mpc, or 4.4 million light years. Is that about right?

(I'm using a Hubble constant of 70 km/sec/Mpc; is this value current?)

[DStahl]

Well, I have no background info on GRBs and insufficient math, so most of the paper went over my head. I wrote the following before I read any of the other responses, just so's as to see what came out.

"I love the way information on these mysterious things trickles in: we already know that the speed with which the bursts develop hints at a compact source, and the paper to hand suggests that the gamma rays may be tightly beamed during the first stages of the GRB. If the transient described in the paper was a GRB afterglow, and if the initial gamma ray birst was missed because we weren't in the beam path, then it seems that perhaps because we did see the afterglow that GRBs might have a strong magnetic field creating a tightly focused beam early on but that the field changes rapidly as the GRB evolves."

Doctor Don hints that the source may be less compact than the radiation front makes it seem, ie the evolution of the GRB may be temporally compressed from our viewpoint. I wonder how compact the zone of activity could be and still fit the observations?

And it appears my conjecture about magnetic fields constraining the radiation was wonky. Ah well.

Don Stahl

[The Bad Astronomer]

On 2002-01-17 20:44, Donnie B. wrote:
If my math is right, a redshift of .385 corresponds to about 1.3 Mpc, or 4.4 million light years. Is that about right?

(I'm using a Hubble constant of 70 km/sec/Mpc; is this value current?)

No, your distance too low. 0.385 means (very roughly) that it is moving at 0.385 times the speed of light (this assumption fails at higher redshift, but it's a place to start). So 0.385 x 300,000 kilometers per second means it is moving at 115,000 kps. Divide by 70 gives you 1650 megaparsecs, or 1.65 gigaparsecs. That's 1.65 billion parsecs.

Remember, the Andromeda Galaxy is about 0.8 million parsecs away, and it is the closest big spiral. It is actually close enough to show a blueshift. A redshift of even 0.1 is a pretty distant object.

[DoctorDon]

On 2002-01-18 00:04, DStahl wrote:
I wonder how compact the zone of activity could be and still fit the observations?

The shortest timescale for variability in
GRBs is about 0.1 s, so the emitting region
can't be bigger than 3x10^9 cm.
Yours,

Don Smith

[Hale_Bopp]

Great to see a discussion going on...just what I hoped would happen!

Anyway, I seem to recall seeing a histogram of the lenght of grbs. If I remember correctly, there were two peaks corresponding to short grbs (~1 sec) and long grbs (~30 sec to 1 mind) and the speculation was that different mechanisms may be responsible for each. I will see if I can find a reference on that, or if our resident GRB expert, Dr. Don, cares to comment [img]/phpBB/images/smiles/icon_smile.gif[/img]

Rob

[Donnie B.]

Dang it! I just redid my calculation exactly as I did last night, and got the same answer as the BA: 1650 Mpc.

It must have been my calculator...

[Lowers his eyes sheepishly]

So that works out to be 5.38 billion LY... somewhere between 1/3 and 1/2 the age of the universe.

Not exactly right next door...

[DoctorDon]

On 2002-01-18 09:20, Donnie B. wrote:
Dang it! I just redid my calculation exactly as I did last night, and got the same answer as the BA: 1650 Mpc.

Actually, if you use a model with an accelerating
expansion, you get 2.08 gigaparsec.
(Using H0=70 km/s/Mpc, Omega_m=0.3, and
Omega_lamda=0.7)

And yes, the GRB distribution is bimodal.


Yours,

Don Smith

[DoctorDon]

I read the paper. I thought it was a very
good paper; very responsible and clear. I'm
not convinced that it was a GRB afterglow;
as they say, there wasn't much in the way
of a decay, and to say "it happened to be
in a plateau state" seems to me to be too
much special pleading, especially since those
plateau states are pretty rare. It certainly
*could* be a GRB, and they do a good job of
discussing alternate hypotheses, but my gut
instinct is probably not.

I doubt if we'll ever know. Now that the
thing has faded away, it's too late to gather
more information to definitively rule out
most of the alternate hypotheses. As they
point out, further monitoring can tell if the
thing flares again (which would rule it out
as a GRB right there), but in terms of
studying *this* flare -- it's over.

However, what the paper does, which is
crucial, is show that it is *possible*
to detect a GRB without a trigger from a
satellite. This is a huge boost for our
project, since we will be conducting large-
scale sky surveys for GRBs when we're not
actively responding to a trigger, and there
were not a few voices saying we wouldn't
find anything. This shows (if it's a GRB)
that they are there to be found, and our
project will get the word out very quickly
for other astronomers to study the thing
with many different instruments while it's
still bright.

So it's an important paper for us, even if
only tangentially. :-)

Yours,

Don Smith

[ToSeek]

Yes, my overall impression of the paper was not that they'd proven anything, but that they found something they thought other people should keep an eye out for.

[Wiley]

Thanks, Hale Bopp. Good Idea.

While I don't know enough to meaningfully contribute to the discussion, I learned a good bit decyphering the paper and from reading the current discussion. The only way to understand the current literature is get in, read, and discuss.

[Hale_Bopp]

Okay, the answer is in...this was not a GRB afterglow. It flared again and someone was watching!

http://arxiv.org/PS_cache/astro-ph/pdf/0202/0202354.pdf

Another tecnhical paper, but probably can be understood by a good number on this board.

It appears to be an unusual, highly variable AGN.

Rob