PDA

View Full Version : Discussion: Recent Blast was Probably a ...



Fraser
2005-Jun-02, 06:14 PM
SUMMARY: When astronomers started analyzing the recent gamma ray burst GRB050509b, they knew right away that they were seeing something very important. This cosmic explosion lasted less than 1/30th of a second, but it provided astronomers with an X-ray afterglow for the first time ever, and pointed them towards a likely cause: the collision of two neutron stars, or a neutron star with a black hole.

View full article (http://www.universetoday.com/am/publish/neutron_star_collision.html)

What do you think about this story? Post your comments below.

antoniseb
2005-Jun-02, 08:57 PM
I'm glad to see some analysis of the observations of this GRB. It's nice to see it match well with some of the previous guesses about the nature of these things.

wstevenbrown
2005-Jun-03, 12:56 AM
:huh: :huh: I'm confused. In the Gehrels article at http://arxiv.org/astro-ph/0505630 , the data very clearly show three spectral lines within range of the instrument, with identical photon count rates, but whose energies (and frequencies) are in the ratio 9 : 3 : 1 . A merger, implosion, or explosion should radiate like a blackbody. This spectrum doesn't obey the Wien-Stefan law. It looks more like the power dissipation for a tuned circuit. Which model predicted this?

Best regards-- Steve

Nereid
2005-Jun-03, 03:41 AM
Steve,

The widely cited Kouveliotou et al (http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1993ApJ...413L.101K&data_type=PDF_HIGH&type=PRINTER&filetype=.pdf) paper (Kouveliotou, C., Meegan, C. A., Fishman, G. J., Bhat, N. P., Briggs, M. S., Koshut, T. M., Paciesas, W. S., & Pendleton, G. N. 1993, ApJ (Letters), 413, 101) set the scene for classifying GRBs into short-hard vs long-soft.

The Gehrels et al paper (pre-print) in your post gives, in Figure 2, the counts by time by energy band. The authors state "A simple power law model fit to 40 ms of data centered on T+23 ms yields a photon index of 1.5±0.4", i.e. a 'classic' hard GRB.

The GRB may very well have radiated like a blackbody; we don't have (in the Gehrels et al. paper) any hint as to what the distribution of energy was across the highest band (100-350 keV).

The Bloom et al. paper on which this story was based give the following references for compact object progenitor models: "the merger of a neutron star binary (NS–NS; Lattimer & Schramm 1976; Paczy´nski 1986, 1991; Narayan et al. 1992; Katz & Canel 1996; Ruffert & Janka 1999; Rosswog & Ramirez-Ruiz 2002; Rosswog et al. 2003) or a black hole–neutron star binary (BH–NS; Eichler et al. 1989; Mochkovitch et al. 1993; Kluzniak & Lee 1998; Bethe & Brown 1999; Popham et al. 1999; Fryer et al. 1999)." Full references are in the preprint (http://www.arxiv.org/PS_cache/astro-ph/pdf/0505/0505480.pdf).

Don Alexander
2005-Jun-06, 12:08 PM
Who wrote this article?? It is actually quite good! ;)

@ Steve & Nereid: GRB spectra are normally described by the so-called Band model (Band et al. 1993), which is a broken power law with a rising and a falling slope and a peak energy. Since hard bursts have their peak energy beyond the maximum energy of BAT, only the falling slope of the power law is visible. The emission is produced by synchroton radiation (relativistic electrons spiralling in a magnetic field) and is decidedly non-thermal, so no blackbody.

Magnetars flares are also ruled out quite strongly. Soft Gamma Repeaters have flaring lifetimes of about 10,000 years, thus they are also associated with high-mass star formation, like long-soft GRBs.

Bersier et al. reported today (via GCN) that they find upper limits for an SN of type 1998bw (the classic long-soft GRB hypernova) between four and six magnitudes under the typical brightness for their observation epochs at a redshift of z = 0.225. This also fits the predictions, as a NS-NS binary merger should not be associated with an SN.

By the way, I must commend Bloom et al. for working incredibly fast (two weeks from event to submission...). Next to the Gehrels et al. paper, a paper from Kulkarni et al. will also appear in Nature (with probably very similar content to Bloom et al.).

PS.: This post was quite a bit longer before my browser crashed... To lazy now to write everything again... :angry:

jbloomastro
2005-Jun-06, 01:52 PM
Originally posted by Don Alexander@Jun 6 2005, 12:08 PM

Who wrote this article?? It is actually quite good! ;)


The paper is http://arxiv.org/abs/astro-ph/0505480, written by a number of us at Berkeley, Santa Cruz, IAS, and PSU amongst other places -- too bad that universetoday did not link to this paper. While the image in the article is from the NASA (as credited) Swift satellite, it is a figure generated by my colleague and co-author Dave Pooley for our paper.

Thanks everyone for the comments and discussions so far.

Josh

Fraser
2005-Jun-06, 04:08 PM
I just went in and fixed the article to link to your paper a couple of times. Thanks for stopping by Josh.

Nereid actually wrote the article Don. :-)

wstevenbrown
2005-Jun-06, 05:18 PM
All of the energy estimates are in terms of isotropic radiation. The observations show the broken power law, suggestive of synchrotron radiation, and more than a little bit suggestive of beaming. None of the authors of either article give beaming more than a passing reference, and no discussion. What gives? This thing fairly shouts of a narrowly collimated jet. Steve

Nereid
2005-Jun-06, 11:08 PM
Steve,

I think you'll find that all researchers in the GRB field are acutely aware of the beaming vs isotropic issues. That you don't see it more explicitly in these three papers may be more due to the 'common understanding' of the folk likely to read them than the intent of the authors.

In what way do you think their presentation of this issue impacts their conclusions re the likely progenitor, environment, and host?

Nereid

wstevenbrown
2005-Jun-07, 03:52 AM
Nereid:
Let me preface these remarks with the observation that, among authors of scientific papers, there seem to be two sorts:
1) Authors who expound enough to interest and inform their audience, who carefully list their assumptions, present their reasoning in an orderly fashion, and cite the prior work on which their work is based. The more thoughtful among them synopsize the key concepts from the citations to show why they are relevant.

2) Jargon-slingers who only attempt to impress the narrow audience within their
specialty, who seek to confound rather than inform, and devalue the
opinions, any attempt at comprehension, and any possible contributions from
anyone less narrowly specialized than themselves. They live by the motto:
“If you can’t blind ‘em with your brilliance, baffle ‘em with your bull****.”

I have very little patience with the latter. Their encrypted publications are no different from code-speak e-mails, which ought to have been kept private. I confess that I am myself a little thick, a little slow, with a tendency toward tunnel vision. It’s curable, tho, and as RP Feynman used to say, “Life’s too short.”

To return to your original question: If the observed radiation is from a collimated beam, it changes the energy budget. If the presumed jet only irradiates a small fraction of the total sphere, we only need account for an amount of energy equal to the isotropic amount multiplied by the inverse of that fraction. With reduced energy needs, the event could be a less-energetic cause closer to the observer.

If, on the other hand, the authors adduce a cause which ought or must be isotropic, they need to clarify why the radiation does not follow the Wien-Stefan Law. IOW, to explain why only some preferred frequencies are emitted. If broader frequency ranges are emitted, they need to explain why they are not observed. If all the other frequencies go down the rabbit-hole (into the presumed BH), even a vague, handwaving description would be nice.

I find the claimed distance too convenient. What eliminates foreground objects from consideration? That line of sight could contain an undetermined number of dwarf galaxies, LSB galaxies, even ejected binaries in intergalactic space. I understand the probabilistic argument based on population syntheses—but the sample space of hard-short GRB’s with constrained distances is too small to generalize from. The discrepancy between the GRB location and the X-ray location projects as many years of travel time between the two events. Are the error bars not as claimed?

Too often, statements containing the formula “is consistent with” are self-serving attempts to prove some particular thesis. They should, at minimum, also contain reasons why no other extant model will serve, or the frank admission that alternative explanations are possible.

I found your article stimulating, Nereid—enough so to seek out the sources and try to learn more. With intensely frustrating results. Best regards-- Steve

Nereid
2005-Jun-07, 08:25 AM
You may have been a little too harsh Steve - GRBs are intensely interesting, and especially the short-hard ones. Folk who work in this area may well feel pressure to get something out quickly, or are just too excited to take the time to write a rich article accessible to a very wide audience.

Too, there's always shorthand, in any group, that is opaque (to varying degrees) to outsiders.

If the opportunity arises, perhaps I could write another story, to bring out how deeply the beaming-isotropic issue is considered?

In the meantime, here (http://gcn.gsfc.nasa.gov/gcn/other/050603.gcn3) is a compilation of GCNs on GRB050603. #3520 has explicit mention of this issue, and references a well-known, widely-used technique for estimating the extent of the beaming. (Just a random sample).

And thanks for the kind words.

Don Alexander
2005-Jun-07, 10:14 AM
Hello, again, every!

@ Josh Bloom:

Hello, Josh, as Fraser states one post later, I was of course asking about the author of the Universe Today article. I read your paper within hours of its publication on astro-ph and liked it a lot.

If your into the technical language, two more articles appeared today on astro-ph:

http://arxiv.org/abs/astro-ph/0506104
http://arxiv.org/abs/astro-ph/0506123

@ Steven:



Jargon-slingers...


I personally do not like articles that rehash everything that came before and where the abstract + introduction amounts to half the article. These articles are written by scientists for the scientific community, and that is a narrow audience. Maybe, within a year or so, a more popular article will appear in, say, Scientific American.



If the observed radiation is from a collimated beam, it changes the energy budget.


This is quite correct, but beaming always reduces the energy budget. The recent observations of the SGR 1806-20 hyperflare have, if I remeber correctly, shown (via the discovery of an expanding spherical radio nebula) that magnetar hyperflares do not produce ultrarelativistic ejecta and thus isotropic energy release is a good measure.



they need to clarify why the radiation does not follow the Wien-Stefan Law


Like I mentioned before, this is non-thermal synchroton radiation, and not a blackbody at all. Furthermore, on nomenclature: The basic law of blackbody radiation is Planck's law. The high-energy part is Wien's law, and the low-energy part is the Rayleigh-Jeans law. Integration over all wavebands and the complete sphere gives the Stefan-Boltzmann law. I've never heard of a "Wien-Stefan law".



If broader frequency ranges are emitted, they need to explain why they are not observed.


How about the fact that they were simply too faint? This was the faintest GRB ever localized with Swift, and progenitor models of short-hard GRBs predict extremely faint afterglows.



What eliminates foreground objects from consideration?


This is actually an interesting question, especially concerning neutron star binaries in intergalactic space. Recent results from the Local Group have shown that ultra low surface brightness dwarf galaxies are very common. We have taken wide images of the field with the Tautenburg Schmidt telescope, and there are no large galaxies nearby. What one could do is the following: Measure the redshifts of all galaxies within, say, 1 degree of the XRT localization. Derive the projected distances from these galaxies. Use distance and luminosity to estimate mass of galaxies, compute escape velocity. Compare with models of asymmetric SN (kick velocities) and neutron star binaries (stability of the system against kicks). Derive time needed to travel projected distance from galaxy at maximum (stability)/minimum (escape velocity) velocity. I think you will find that no visible galaxies except G1 can be the host within the age of the universe.



The discrepancy between the GRB location and the X-ray location


Which discrepancy? The XRT error circle lies fully within the BAT location... You seem to be misunderstanding something here.



the formula “is consistent with”


Which normally means that two measurements with error bars overlap within these error bars.

wstevenbrown
2005-Jun-07, 02:39 PM
Slightly better discussion:

http://arxiv.org/astro-ph/0506104

Much better discussion:

http://arxiv.org/astro-ph/0506123

It’s all about the assumption set, and the probabilistic, dependent modeling. Don’t bet the farm on any of the present conjectures—they would be easily falsified by a single example of repeat behavior in spatial coincidence with a previous short-hard gamma burst (SGR behavior, short-soft bursting, etc.).

Several alternate scenarios come to mind, foremost among which is merging of a really hot Jupiter or stripped stellar core with a NS. Same kind of orbital decay as AM CVn stars, with merger accelerated at the end by magnetic coupling. Because of the energy budget, this would necessarily be more nearby than the presumed host of GRB050509b.

I’m not proposing this as THE model for this event. We need only keep track of our assumptions. S

Postscript: Just read the two preceding posts-- shoulda known you guys would be on the case. Thanks for additional info, and for putting up with my gripes.

Don A, thanks for terminology correction-- my paraphrasing memory sometimes gets me in trouble that way.

Nereid-- I would LOVE to see your take on isotropy vs. beaming, in a feature-length article. I think a full discussion would be very timely, and your comments are always lucid and well-informed. As several have noted, the data are coming in at a furious rate, and it would be very helpful to the interested amateur community to know what's being talked about, at a level deeper than wow-gee-whiz. S