Gamma-Ray Bursts
The Brightest Explosions Since the Big Bang
H. A. Bethe Lecture
March 6, 2002
Electromagnetic radiation – of which ordinary visible light is an
example – can be characterized by its wavelength. Gamma-rays have
the shortest wavelengths of all and are very penetrating. They pass easily
through a block of wood but not through the Earth’s atmosphere.
Consequently cosmic sources of gamma-rays can only be seen by
satellites, rockets, and high flying balloons that go above most of
the Earth’s atmosphere.
Velar – to watch
Nuclear Test Ban Treaty, 1963
First Vela satellite pair launched 1963
The Vela 5 satellites were placed in orbit by the Advanced
Research Projects of the DoD and the AEC. Launched on
May 23, 1969 into high earth orbit (118,000 km), this pair
of satellites and their predecessors, Vela 4, discovered the
first gamma-ray bursts. The discovery was announced
by Klebesadel, Strong, and Olson (ApJ, 182, 85) in 1973.
First Gamma-Ray Burst
The Vela 5 satellites functioned from July, 1969 to April, 1979
and detected a total of 73 gamma-ray bursts in the energy
range 150 – 750 keV (n.b, radiation is sometimes defined by its
energy measure in electron volts. 0.3 to 30 keV approximately is
x-rays. Greater than 30 keV is gamma-rays)
Typical durations are
20 seconds but there is
wide variation both in timestructure and duration.
Some last only hundredths
of a second. Others last
thousands of seconds.
A Cosmic Gamma-Ray Burst, GRB for short, is a brief,
bright flash of gamma-rays lasting typically about 20
seconds that comes from an unpredictable location in
the sky.
Some, in gamma-rays, are as bright as the planet Venus.
Most are as bright as the visible stars. It is only because of
the Earth’s atmosphere and the fact that our eyes are not
sensitive to gamma-rays that keeps us from seeing them
frequently.
With appropriate instrumentation, we see about one of
these per day at the Earth. They seem never to repeat from
the same source.
A problem in 1973 was – and still is – that gamma-ray detectors
(CsI and NaI) do not give much information on directect.

Yes, there are gamma-rays
but I don’t know where they
came from.....
but I do know when ...
One can get information on directionality from timing
if there are more than one detectors.
1
2
If “1” hears the sound later than “2” the sound is somewhere
to the right.
• Interstellar warfare
• Primordial black hole evaporation
• Flares on nearby stars
uncertainty in distance –
• Distant supernovae
a factor of one billion.
• Neutron star quakes
• Comets falling on neutron stars
• Comet anti-comet annihilation
• Thermonuclear explosions on neutron stars
• Name your own ....
In 1993 there were 135 models
nb. 27, 42, 105, 126
Most of them involved neutron stars in our own Galaxy
(quakes, comets falling, thermonuclear runaways, etc.)
The expected distribution on the sky if Galactic neutron
stars were responsible should center around the Galactic equator.
Plane of the Milky
Way Galaxy
So basically we wandered in the
wilderness for twenty years ...
(1973 – 1993).
Compton Gamma-Ray Observatory
April 5, 1991 – June 4, 2000
BATSE Module
Observed
Expected if haven’t
reached any edge yet
log number
of sources
log sensitivity
This posed a problem for the models that had GRBs in our
own Galaxy ..
*
*
*
*
*
*
*
*
*
*
*
*
*
**
*
*
X
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
We are not at the center
of our galaxy so we
should see more bursts
towards the center than
in the opposite direction,
Isotropy could mean three things:
• Very nearby bursts centered on the Earth – e.g.,
the Oort cloud of comets
• A very extended spherical halo around the Galaxy –
much bigger than the distance from here to the
center of the Galaxy
• Bursts very, very far away – billions of light years
What we really needed was source identifications.
High Energy Transient Explorer (HETE-1)
HETE-1
conceived July, 1981, Santa Cruz
born Nov 4, 1996
died Nov 5, 1996
BeppoSax (1996-2002)
(Italian-Dutch X-ray astronomy mission)
MEC S and LECS (medium and low
energy x-ray sensors, 1 arc min positions)
(2-30 keV; 20x20 degree FOV
angular resolution 5 arc min)
The scintillator anti-coincidence shields of the Phoswich detector
are able to detect gamma-rays 60-600 keV and get crude angular
information
BeppoSax GRB 970228 (discovered with WFC)
Feb 28, 1997 (8 hr after GRB using MECS)
March 3, 1997 (fainter by 20)
Each square is about 6 arc min or 1/5 the moon’s diameter
GRB 970228
William Hershel Telescope
Isaac Newton Telescope
Groot, Galama, von Paradijs, et al IAUC 6584, March 12, 1997
Wagner, Foltz, and Hewet IAUC 6581 using the MMT
get a crude spectrum of the host galaxy. Estimate z = 0.5
March 10, 1997
Later ....
Spectrum of the host galaxy of GRB 970228 obtained at the Keck 2 Telescope.
Prominent emission lines of oxygen and neon are indicated and show that the
galaxy is located at a redshift of z = 0.695. (Bloom, Djorgovski, and Kulkarni
(2001), ApJ, 554, 678. See also GCN 289, May 3, 1999.
From the red shift a distance can be inferred – billions
of light years. Far, far outside our galaxy.
From the distance and brightness an energy can be
inferred.
1.6 x 1052 erg in gamma rays alone
This is 13 times as much energy as the sun will radiate in
its ten billion year lifetime, but emitted in gamma-rays
in less than a minute. It is 2000 times as much as a
really bright supernova radiates in several months.
GRB 990123
This shows the light curve as seen by BATSE starting at 9:46:56 UT on
January 23, 1999. The burst was seen simultaneously by the WFC on BeppoSax.
The position was promptly determined to a few degrees by BATSE and shortly
thereafter to 5 arc min by BeppoSax.
The Robotic Optical Transient Search Telescope (ROTSE-1)
by Akerlof et al, located at Los Alamos, images a piece of the sky
16.5 degrees across. It is able to slew anywhere in the sky in about 10 s.
It was notified of GRB 990123 within 4 s of the onset and was taking data
22 s into the event. The first three ROTSE points are shown superimposed
on the BATSE light curve.
m = 11.82
22 s, 5 s exposure
m = 13.22
259 s, 75 s exposure
m = 8.95
47 s, 5 s exposure
m = 14.0
447 s, 75 s exposure
m =10.08
72 s, 5 s exposure
m = 14.53
612 s, 75 s exposure
And Palomar finds a light where no light was before.
This ID was made 3 hours after the burst following a 5AM
“wake-up” call to Palomar from Italy
GRB 990123
BeppoSax observations of
GRB 990123 6 to 33 hours
after the burst and 34 to 64
hours after the burst.
The x-ray source is clearly
fading. Each grid box is
12 x 15 arc min. The error in
position is then about 1.5 arc
min.
Two HST images of GRB 990123. The image on the left was taken
February 8, 1999, the one on the right March 23, 1999. Each picture
is 3.2 arc seconds on a side. Three orbits of HST time were used for
the first picture; two for the second – hence the somewhat reduced
exposure.
The spectrum of host galaxy (Kelson et al, IAUC 7096) taken
using the Keck Telescopes gives a redshift of 1.61.
Given the known brightness of the burst (in gamma-rays) this
distance implies an energy of over several times 1054 erg.
About the mass of the sun turned into pure energy.
Had this burst occurred on the far side of our Galaxy,
at a distance of 60,000 light years, it would have been
as bright – in gamma-rays – as the sun. This is ten billion
times brighter than a supernova and equivalent to seeing a
one hundred million trillion trillion megaton explosion.
As of February, 2002, 129 GRBs had been
promptly localized by various satellites
(starting in 1997).
X-ray afterglows had been detected from 40 and
optical afterglows from 28.
Red shifts (distances) were determined for about 21.
Typical values are z ~1. The largest is 3.42 (GRB 971214)
corresponding to emission when the universe was
about 15% its present age.
The typical energy is 1053 erg or about
5% of the mass of the sun turned to pure
energy according to E = mc2
But are the energies required really that great?
Earth
If the energy were
beamed to 0.1% of the
sky, then the total
energy could be
1000 times less
Earth
Nothing seen down here
But then there would be a lot of bursts that we do not see
for every one that we do see. About 1000 in fact.
Earth
Quasar 3C 175 as seen in the radio
Quasar 3C273 as seen by the
Chandra x-ray Observatory
Microquasar GPS 1915
in our own Galaxy – time sequence
Artist’s conception of SS433
based on observations
It also turns out that in order to get the spectrum
and time scales for GRBs correct the beam must
be “relativistic”, that is it must travel at very close
to the speed of light.
Otherwise the bursts would be much softer and
last much longer.
It is a property of matter moving close to the speed
of light that it emits its radiation in a small angle along its
direction of motion. The angle is inversely proportional to the
Lorentz factor
1

,
E.g.,  100 v  0.99995 c
2
2
1 v / c
 1 / 
  10
v  0.995 c
This offers a way of measuring the beaming angle. As the
beam runs into interstellar matter it slows down.
Measurements give
an opening angle of
about 5 degrees.
Frail et al. Nature, (2001), for 17 GRBs
with known redshifts and afterglow
light curves.
Using the angles inferred from
this sort of analysis of the
afterglows, Frail et al find that
an inverse correlation exists
between the apparent energy of
the burst and this angle.
Correcting for the fact that the
burst is beamed to a small part of
the sky, they find a typical
energy in gamma-rays is
5 x 1050 erg. For a reasonable
conversion efficiency between
explosion energy and gamma-rays
(20%), the total jet energy is about
2 x 1051, not so different from
an ordinary supernova.
Requirements on the Central Engine
and its Immediate Surroundings
• Provide adequate energy to material moving close to
the speed of light (2 x 1051 erg)
• Collimate the emergent beam to approximately 5 degrees
• Last approximately 10 s
• Make bursts in star forming regions
Merging Neutron Stars
Merging neutron star black hole pairs
Strengths: a) Known event
b) Plenty of angular momentum
c) Rapid time scale
d) High energy
e) Well developed numerical models
Weaknesses: a) Outside star forming regions
b) Beaming and energy may be inadequate
for long bursts
But this model may still be good for a class of bursts called
the “short hard” bursts for which we have no counterpart information
yet.
The Collapsar Model (aka the “Hypernova”)
Usually massive stars make supernovae. Their iron core
collapses to a neutron star and the energy released explodes
the rest of the star.
But what if the explosion fizzled? What if the iron core
collapsed to an object too massive to be a neutron star –
a black hole.
A star without rotation would then simply disappear....
But what if the star had too much rotation to all go
down the (tiny) black hole?
If supernovae are the observational signal that a
neutron star has been born, what is the event that
signals the birth of a black hole?
In the vicinity of the rotational
axis of the black hole, by a
variety of possible processes,
energy is deposited.
The exact mechanism for
extracting this energy either
from the disk or the rotation
of the black hole is fascinating
physics, but is not crucial
to the outcome, so long as the
energy is not contaminated by
too much matter.
7.6 s after core collapse; high viscosity case.
SN 1998bw/GRB 980425
NTT image (May 1, 1998) of SN 1998bw in
the barred spiral galaxy ESO 184-G82
[Galama et al, A&A S, 138, 465, (1999)]
1) Were the two events the same thing?
2) Was GRB 980425 an "ordinary" GRB
seen off-axis?
WFC error box (8') for GRB 980425
and two NFI x-ray sources. The IPN
error arc is also shown.
A 1053 erg event situated 30,000 light years away
(distance from here to the Galactic center) would give
as much energy to the earth in 10 seconds as the sun –
equivalent to a 200 megaton explosion.
Does it matter having an extra sun in the sky for
10 seconds?
Probably not. This is spread all over the surface of
the earth and the heat capacity of the Earth’s atmosphere
is very high. Gamma-rays would deposit their energy
about 30 km up. Some bad nitrogen chemistry would happen.
Noticeable yes, deadly to all living things – No.
Biological Hazards of Gamma-Ray Bursts
Distance
(kpc)
Events
/10 by
10
100 – 1000
1
1 – 10
0.1
0.01 – 0.1
Megatons
200
20,000
two million
Results
Some ozone damage, EMP
acid rain
Ozone gone, acid rain, blindness
2nd and 3rd degree burns*
Shock waves, flash incineration,
tidal waves, radioactivity (14C)
End of life as we know it.
* Depends on uncertain efficiency for conversion of energetic electrons
to optical light
Milagrito detection of GRB 970417a
Atkins et al. (2000), Astrophysical Journal Letters, 533, L119
Large watertight detector near Los Alamos.
Threshold about 100 GeV (gamma-rays
with 100,000 times more energy and shorter
wavelength than “ordinary” GRB energies)
Fluence of this burst in BATSE was
1.5 x 10-7 erg cm-2.
100 times more energy in TeV radiation??
Gamma-Ray Large Area Space Telescope (GLAST)
Sensitive to GRBs in the
energy range 20 MeV to 300 GeV
Scheduled for launch by NASA
2005
detector follows the paths of
electron-positron pairs created
in foils.
Will the high energy emission
of GRBs be their dominant emission?
HETE-2
October 6, 2000
Last night ...