Analog Science Fiction & Fact Magazine
"The Alternate View" columns of John G. Cramer
Previous Column Index Page Next Column

"Texas" in Munich, Part 2: Gamma Ray Bursts

by John G. Cramer

Alternate View Column AV-74
Keywords: gamma ray bursts NASA BATSE fireball neutron star merger galactic cosmological cosmology
Published in the October-1995 issue of Analog Science Fiction & Fact Magazine;
This column was written and submitted 3/1/95 and is copyrighted (©1995 by John G. Cramer.
All rights reserved. No part may be reproduced in any form without
the explicit permission of the author.


    This is the second of two columns reporting on the 17th Texas Symposium on Relativistic Astrophysics held on December 12-15, 1994, in Munich, Germany, where I am presently on sabbatical. The "Texas" in the title refers to the site of the first of these meetings, which was held in 1962 in Dallas, Texas shortly after the discovery of quasars. Munich was host to the most recent Texas Symposium because it has become a center of work in x-ray astrophysics, both theoretical and observational, as well as the home base of the European Southern Observatory.

    In this column I will give an overview of the most puzzling phenomenon discussed at the Symposium: gamma ray bursts.

§ The CIA Discovers Gamma Rays from Outer Space! - In 1967, during the Johnson Administration, the United States entered a treaty that banned the testing of nuclear weapons in space. In order to verify that the terms of the treaty were being observed, in the late 1960s the CIA and other intelligence agencies arranged for the Air Force to launch the VELA series of surveillance satellites, which carried omni-directional detectors sensitive to pulses of gamma rays that would have been emitted in great numbers by any space-based tests of nuclear weapons. The VELA satellites never detected a nuclear explosion, or at least never detected one originating in our Solar System.

    However, from the time when the first VELA probe was activated bursts of gamma rays were detected every few days. The intelligence analysts were baffled and suspected satellite malfunctions or some form of deliberate jamming that would hide real tests. Finally, with several satellites in orbit simultaneously it became possible to use the arrival time of the pulses at different locations to determine the direction of the source of the radiation. It was found that the bursts came from a number of directions outside the solar system. The existence of these bursts was kept as a closely guarded secret for several years. Finally, in 1973, a paper was published in Astrophysics Journal Letters describing the observation of gamma ray bursts, and the astrophysicists of the world were confronted with a new phenomenon, which, at the time of the 17th Texas Symposium of Relativistic Astrophysics has still not been adequately explained.

§ The gamma ray burst phenomenon - Gamma ray bursts have an unusual set of features that have proved very difficult to explain. Here are some of them.

(1) Duration - The most enigmatic characteristic of the gamma ray bursts (GRB) are their very short time duration. The burst duration can be as short as 1/1000 of a second, but bursts that continue for a few minutes have been observed, and continued activity in a few cases has been recorded for half an hour. In astrophysics the variation time-constant of an phenomenon times the speed of light is taken as an indication of its source size. The millisecond time constant of GRB thus suggests a source about 300 kilometers in diameter. This is an uncomfortably small size, even for compact astronomical objects like neutron stars and black holes.

(2) Rate of Occurrence - In the past few years, the BATSE experiment on NASA's Compton Gamma Ray Observatory satellite has recorded over 1000 GRB events at a rate of about 1 per day. BATSE has about a 33% detection efficiency, so the true rate of GRB occurrence is about once every 8 hours or 1000 per year. GRB are thus very frequent indeed.

(3) Distribution of Locations - The GRB events recorded by the BATSE experiment show a uniform distribution in the sky, with no preference for the plane of the solar system, the plane of the galaxy, or for any particular direction. Further, there is no evidence of repeating locations. As far as one can tell form the present data, every GRB event is new and unique. The GRB's uniform distribution of locations is now taken by most astrophysicists as an indication that the phenomenon is "cosmological" in origin, i.e., that GRB do not originate in our solar system or galaxy but come from sites that are very far away.

(4) Energy Content - Typically during the detection of a GRB event each square centimeter of detector area receives about 10-5 ergs of energy. If we know (or guess) the distance to the source of the GRB, this detected energy value can be used with the inverse square to calculate the energy released by the source. If the GRB source emitting gamma rays uniformly in all directions were in the outer halo of our galaxy at a distance of 100 kilo-parsecs, its net energy release would be 1041 ergs. If the uniform GRB source was cosmological with a distance of about 3 giga-parsecs, its net energy release would be 1051 ergs. If the source were to "beam" the gamma rays like a lighthouse, these need energies might decrease by a factor of 10 to 100.

(5) Energies of Gamma Rays - Typically the energy distribution of gamma rays in a GRB peaks at around 200 keV (thousand electron-volts) and extends well into the MeV (million electron-volts) region. This spectrum has the wrong shape to be produced by a very hot thermal object. The distribution of the Gamma rays with energies of up to 25 GeV (billion electron-volts) have been observed. These high energy gammas usually occur much later than the main burst, following by several hundreds seconds or more.

(6) Time Structure - The time profile of GRB varies enormously from burst to burst and over time scales from milli-seconds to minutes. Examples of time structures that are smooth, multi-spiked, short, long, complex, and quasi-periodic can be found among the catalog of GRB. One common time profile is the "FRED", an acronym that stands for fast-rise and exponential decay. There is, however, no evidence for any regular periodic time structures like those that might be expected from a rapidly rotating neutron star or merging pair of neutron stars (see below).

(7) Counterparts of GRB - Ever since their discovery and particularly since BATSE has gone into operation there have been many attempts to correlate the occurrence of GRB with excursions in the output of objects observed with radio telescopes, infra-red telescopes, optical telescopes, cosmic ray detector arrays, and even neutrino detectors. So far, there is no convincing evidence of any counterparts of GRB. For those GRB that are the best localized, there is also no particular correlation with stars or galaxies in the same angular "detection box".

(8) Intensity cutoff of GRB - The BATSE detector has sensitivity to GRB that are considerably weaker than the "average" GRB detected. Analysis of the observed GRB shows that there are considerably fewer weak GRB that would be expected if these were distributed uniformly in space and time. This suggests and "evolution" of the phenomenon, in other words that in earlier stages of the universe there were fewer GRB events than there are now.

§ GRB at the Texas Symposium - At the Texas Symposium Dr. Gerald Fishman, the leader of the BATSE project, gave an invited talk that gave a nice overview of the detector and its results. The latest BATSE catalog includes 1121 GRB events, and that number is growing at the rate of one per day. There are about 2000 papers in the scientific literature on gamma ray bursts. A recent review article has provided a list of over 100 theoretical models, all of which in different (and usually mutually exclusive) ways attempt to explain the GRB phenomenon.

    In the afternoon following Fishman's talk was a special session of papers devoted to the gamma ray burst phenomenon and recent results, many of which I have summarized above. At this session there was also considerable discussion of theoretical models.

    There are three generic sites where theories have attempted to place GRB sources, in the Oort cloud of our solar system, in the halo of our galaxy, and at cosmological distances. The Oort cloud models are attractive because they require a much smaller energy release to explain the observations. However, they have been largely excluded by the BATSE measurement of uniform location distributions and by timing measurements that place the GRB source at a distance of at least 25 AU (Earth-Sun radii) away.

    There are still some advocates who would place the GRB in the halo of our galaxy, attributing GRB to phenomena like comets falling on neutron stars or events involving very magnetic neutron stars. These models also have problems with the BATSE measurement of uniform location distributions. and would require very energetic ejection of neutron stars from the galactic plane when they are formed in supernovas.

    The "standard model" for explaining GRB would place them at cosmological distances and attribute them to the merger of a binary pair of neutron stars that are spinning down to collision as they lose energy through gravitational radiation. Such objects are known to exist and have been carefully studied. It is estimated that there should be about one such merger event per galaxy per million years, which implies about 100 such events per year should occur within the 3 Gpc radius suggested by the BATSE intensity cutoff. This is only a factor of 10 lower than the actual observed event rate, which is considered pretty good by astrophysics standards. The other thing the neutron star merger model has going for it is that the total energy release of a pair of merging neutron stars is estimated to be about 1054 ergs, most of which goes into the emission of a burst of neutrinos. It is argued that if 0.1% of this energy is channeled into electromagnetic energy, it corresponds to the energy release needed for the GRB source.

    The problems with this model are:
(1) it would normally produce a very thermal distribution of gamma rays that is difficult to avoid,
(2) there is no evidence in the BATSE data of any periodic time structure such as might be expected from a binary neutron star spin-down and merger,
(3) if even a small amount of normal matter contaminates the electromagnetic fireball which is expected to result from the neutron star merger, the fireball output is converted to energetic protons instead of gamma rays,
(4) there are severe problems in accommodating the millisecond time scale, and
(5) there are even more severe problems in accounting for the 25 GeV gamma rays that have been observed as a part of some GRB events.

    Fortunately, this model will become testable in the near future. When the LIGO gravity wave detector goes into operation in a few years, it should be able to detect the gravity waves generated in a binary neutron star spin-down and merger. These should correlate in time and spatial direction with GRB, of the neutron star model is valid.

    My guess, however, is that GRB are NOT neutron star mergers, but something far more unexpected and exotic. I think Nature is sending us a message that arrives once every 8 hours telling us that the Universe is a far stranger place than we had imagined.

SF Novels by John Cramer:  my two hard SF novels, Twistor and Einstein's Bridge, are newly released as eBooks by Book View Cafe and are available at : .

AV Columns Online: Electronic reprints of about 177 "The Alternate View" columns by John G. Cramer, previously published in Analog , are available online at:

Previous Column Index Page Next Column

Exit to the Analog Logo website.

This page was created by John G. Cramer on 7/12/96.