HETE-2 Observation of Two Gamma-Ray Bursts at z > 3

Abstract

GRB 020124 and GRB 030323 constitute half the sample of gamma-ray bursts (GRBs) with a measured redshift greater than 3. This paper presents the temporal and spectral properties of these two GRBs detected and localized with the High Energy Transient Explorer 2 (HETE-2). While they have nearly identical redshifts (z = 3.20 for GRB 020124 and z = 3.37 for GRB 030323), these two GRBs span about an order of magnitude in fluence, thus sampling distinct regions of the GRB luminosity function. The properties of these two bursts are compared with those of the bulk of the GRB population detected by HETE-2. We also discuss the energetics of GRB 020124 and GRB 030323 and show that they are compatible with the Ep-Eiso relation discovered by Amati et al. Finally, we compute the maximum redshifts at which these bursts could have been detected by HETE-2, and we address various issues connected with the detection and localization of high-z GRBs.

Full text

Page 1

arXiv:astro-ph/0502494v1 23 Feb 2005
HETE-2 Observation of two gamma-ray bursts at z > 3
J.-L. Atteia,3N. Kawai,4,5R. Vanderspek,1G. Pizzichini,16G. R. Ricker,1C. Barraud,3
M. Boer,18J. Braga,14N. Butler,1T. Cline,13G. B. Crew,1J.-P. Dezalay,11
T. Q. Donaghy,2J. Doty,1E. E. Fenimore,10M. Galassi,10C. Graziani,2K. Hurley,6
J. G. Jernigan,6D. Q. Lamb,2A. Levine,1R. Manchanda,15F. Martel,1M. Matsuoka,8
E. Morgan,1Y. Nakagawa,9J.-F. Olive,11G. Prigozhin,1T. Sakamoto,4,5,13R. Sato,4
Y. Shirasaki,5,7M. Suzuki,4K. Takagishi,17T. Tamagawa,5K. Torii,19J. Villasenor,1
S. E. Woosley,12M. Yamauchi,17and A. Yoshida5,9

Page 2

– 2 –
ABSTRACT
GRB 020124 and GRB 030323 constitute half the sample of gamma-ray bursts
with a measured redshift greater than 3. This paper presents the temporal and
spectral properties of these two gamma-ray bursts detected and localized with
1Center for Space Research, Massachusetts Institute of Technology, 70 Vassar Street, Cambridge, MA,
02139.
2Department of Astronomy and Astrophysics, University of Chicago, 5640 South Ellis Avenue, Chicago,
IL 60637.
3Laboratoire d’Astrophysique, Observatoire Midi-Pyr´ en´ ees, 14 Ave. E. Belin, 31400 Toulouse, France.
4Department of Physics, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo 152-8551,
Japan.
5RIKEN (Institute of Physical and Chemical Research), 2-1 Hirosawa, Wako, Saitama 351-0198, Japan.
6University of California at Berkeley, Space Sciences Laboratory, Berkeley, CA, 94720-7450.
7National Astronomical Observatory, Osawa 2-21-1, Mitaka, Tokyo 181-8588 Japan.
8Tsukuba Space Center, National Space Development Agency of Japan, Tsukuba, Ibaraki, 305-8505,
Japan.
9Department of Physics, Aoyama Gakuin University, Chitosedai 6-16-1 Setagaya-ku, Tokyo 157-8572,
Japan.
10Los Alamos National Laboratory, P.O. Box 1663, Los Alamos, NM, 87545.
11Centre d’Etude Spatiale des Rayonnements, Observatoire Midi-Pyr´ en´ ees, 9 Ave. du Colonel Roche,
31028 Toulouse Cedex 4, France.
12Department of Astronomy and Astrophysics, University of California at Santa Cruz, 477 Clark Kerr
Hall, Santa Cruz, CA 95064.
13NASA Goddard Space Flight Center, Greenbelt, MD, 20771.
14Instituto Nacional de Pesquisas Espaciais, Avenida Dos Astronautas 1758, S˜ ao Jos´ e dos Campos 12227-
010, Brazil.
15Department of Astronomy and Astrophysics, Tata Institute of Fundamental Research, Homi Bhabha
Road, Mumbai, 400 005, India.
16INAF/IASF Sezione di Bologna, via Piero Gobetti 101, 40129 Bologna, Italy.
17Faculty of engineering, Miyazaki University, Gakuen Kibanadai Nishi, Miyazaki 889-2192, Japan.
18Observatoire de Haute Provence, 04870 St. Michel l’Observatoire, France.
19Department of Earth and Space Science, Graduate School of Science, Osaka University, 1-1
Machikaneyama, Toyonaka, Osaka 560-0043, Japan

Page 3

– 3 –
HETE-2. While they have nearly identical redshifts (z=3.20 for GRB 020124,
and z=3.37 for GRB 030323), these two GRBs span about an order of magnitude
in fluence, thus sampling distinct regions of the GRB luminosity function. The
properties of these two bursts are compared with those of the bulk of the GRB
population detected by HETE-2. We also discuss the energetics of GRB 020124
and GRB 030323 and show that they are compatible with the Ep- Eisorelation
discovered by Amati et al. (2002). Finally, we compute the maximum redshifts at
which these bursts could have been detected by HETE-2 and we address various
issues connected with the detection and localization of high-z GRBs.
Subject headings: gamma rays: bursts (GRB 020124, GRB 030323)
1.Introduction
High redshift gamma-ray bursts (GRBs) may become useful beacons for the study of
the young universe. Gamma-ray bursts are expected to be visible out to very large redshifts
(z=10-20, Lamb & Reichart 2000), if indeed they are generated there, and offer the possibility
to probe the interstellar medium along the line of sight and to address important cosmological
issues like the evolution of the star formation rate.
Before one can undertake such studies, however, it is important to understand the
intrinsic properties of gamma-ray bursts at high redshift. There are only four GRBs with a
measured redshift larger than z=3: GRB 971214 at z=3.42 (Kulkarni et al. 1998), which was
localized by BeppoSAX; GRB 000131 at z=4.5 (Andersen et al. 2000), which was localized
with the IPN; and GRB 020124 at z=3.20 (Hjorth et al. 2003) and GRB 030323 at z=3.37
(Vreeswijk et al. 2004), which were localized by HETE-2.
This paper describes the HETE-2 observations of GRB 020124 and GRB 030323. After
a quick summary of the localization history of these bursts (section 2), we present their
spectral and temporal properties in section 3 and compare them with those of the bulk of
the GRB population. Their energetics are discussed in section 4, where it is noted that GRB
030323 is the first GRB detected at large redshift which does not belong to the bright end
of the GRB luminosity distribution. In this section we also show that HETE-2 could have
detected and localized GRB 020124 at a redshift of z=6.4 at least, and we discuss some
necessary conditions for the detection of soft, faint GRBs at high redshift. Section 5 looks
into the observations of the afterglows and host galaxies of these bursts.

Page 4

– 4 –
2.Detection, localization, and optical afterglow identification
2.1.GRB 020124
On 2002 January 24, at 10:41:15.15 UT (38475.15 SOD), the High Energy Transient
Explorer satellite HETE-2 (hereafter HETE) detected GRB 020124, a moderately bright
GRB. No flight localization was issued by the satellite, but the analysis of data on the
ground resulted in a coarse localization distributed to the GCN11.4 hours after the GRB,
and in a refined position distributed to the GCN 10.7 hours after the GRB (Ricker et al.
2002). The refined position was a circle of radius 12’, centered at RA = 09h 32m 49s, Dec
= -11◦27’ 35” (J2000). The optical afterglow was reported 28 hours after the GRB, (in
data taken 13.5 hours after the GRB) at the position: RA = 09h 32m 50.8s, Dec = -11◦31’
11”, well within the refined error box (Price et al. 2002). An IPN annulus was also obtained
for this burst, which was fully consistent with the WXM error box (Hurley et al. 2002).
Fig. 1a shows the projection of the WXM error box on the sky, the IPN annulus, and
the position of the optical afterglow. At the time of its identification, the afterglow had a
magnitude R=18.5 (Price et al. 2002). Details of the identification of the optical afterglow
and its evolution can be found in Berger et al. (2002). The afterglow spectrum recorded
with ISAAC on the VLT-Antu is analysed in Hjorth et al. (2003), who report a redshift
z=3.198.
2.2.GRB 030323
On 2003 March 23, at 21:56:57.60 UT (79017.60 SOD), HETE detected GRB 030323,
a faint GRB. As for GRB 020124 no flight localization was issued by the satellite, but the
analysis of data on the ground resulted in a WXM localization distributed to the GCN 5.0
hours after the GRB, and in an SXC localization distributed to the GCN 7.5 hours after the
GRB (Graziani et al. 2003). The WXM position was a circle of radius 18’, centered at RA
= 11h 06m 54s, Dec = -21◦51’ 00” (J2000). The SXC position was a trapezoid with an area
of 71 arcmin, fully included within the WXM error box. The center of the SXC error box
was RA = 11h 06m 06s, Dec = -21◦54’ 20” (J2000). The optical afterglow was reported 22
hours after the GRB, (in data taken 9.6 hours after the GRB), at the position: RA = 11h
06m 09.38s, Dec = -21◦46’ 13.3” , close to the boundary of the SXC error box, but within it
(Gilmore et al. 2003). Fig. 1b shows the projection of the WXM and SXC error boxes on the
sky, and the position of the optical afterglow. At the time of its identification, the afterglow
1Gamma-Ray Burst Coordinate Network, see http://gcn.gsfc.nasa.gov/

Page 5

– 5 –
had a magnitude Rc=18.7 (Gilmore et al. 2003). A temporal and spectral analysis of the
optical afterglow can be found in Vreeswijk et al. (2004), who report a redshift z=3.372.
3. The prompt emission
3.1.Light curves and temporal properties
The light curves of GRB 020124 and GRB 030323 are displayed in Fig. 2. GRB 020124
and GRB 030323 belong to the class of long GRBs with durations T90= 49.4 ±1.3 sec (GRB
020124), and T90= 15.0 ±2.6 sec (GRB 030323) in the energy range 6-400 keV. Table 1 gives
the durations of these two bursts, and their one sigma errors, in various energy bands. In
the FREGATE data, GRB 020124 exhibits little duration shortening with the energy with
T90(resp. T50) varying from 51.4 ± 1.4 sec (resp. 26.0 ± 2.7 sec) in the energy band 6-15
keV to 43.0 ± 6.1 sec (resp. 19.2 ± 1.8 sec) in the energy range 85-400 keV. The situation
is more confusing in the WXM 2−25 keV energy band (T90and T50seem to follow distinct
trends), possibly due to the lower signal to noise ratio of the burst in this instrument. The
light curve of GRB 020124 appears very spiky: at least 9 individual spikes can be identified
in it.
GRB 030323 is significantly fainter than GRB 020124, and we could only divide the
energy range into two subranges : in 6-30 keV we measure T90= 12.8 ± 2.5 sec and T50=
6.6 ± 1.5 sec, and in 30-400 keV we measure T90= 12.2 ± 3.6 sec and T50= 5.2 ± 1.6 sec.
With only two energy bands, it is difficult to say whether GRB 030323 exhibits significant
duration shortening with energy.
3.2.Spectra
In this section we investigate the average spectral properties of GRB 020124 and GRB
030323. The joint WXM+FREGATE spectra have been fit with a powerlaw times ex-
ponential model (PLE), n(E) ∝ E−αexp(−E/eobs
which satisfactorily fits most GRB spectra (Band et al. 1993).2This parametrization al-
lows us to compare these bursts with the GRBs detected by BATSE (Band et al. 1993;
Preece et al. 2000), BeppoSAX (Frontera et al. 2000; Amati et al. 2002), and HETE (Barraud et al. 2003
0), and with a Band function (GRBM),
2In this paper we use capital letters for intrinsic spectral parameters (at the source), and lower case letters
for the observed spectral parameters.

Page 6

– 6 –
Sakamoto et al. 2004b). The spectral parameters for the burst-averaged spectra of GRB
020124 and GRB 030323 are given in Table 2. The emission properties of GRB 020124 and
GRB 030323 are given in Table 3. Emission properties depend on the model used in the
spectral analysis. In Table 3 we report the numbers given in Sakamoto et al. (2004b), which
are based on a PLE fit for GRB 020124 and on a simple powerlaw fit for GRB 030323. It
should be noted that a spectral analysis based on the best fit Band function (given in Table
2) gives emission parameters which differ by less than 10% from the values given in Table 3.
The discussion in section 4, which requires the measure of ’bolometric’ fluences is based on
the best fit Band functions given in Table 2.
3.2.1. GRB 020124
With a fluence of 8.1 × 10−6erg cm−2in the energy range 2 - 400 keV, GRB 020124
belongs to the brightest third of HETE GRBs. Its ’softness’3is 0.32, placing it at the
boundary between GRBs and X-Ray Rich GRBs. The peak energy, eobs
constrained to be about 90 keV. The difference between this value and the value of 133 keV
quoted in Barraud et al. (2003) is due to the inclusion of the WXM data covering the range
2-25 keV in this refined analysis.
p
, is relatively well
3.2.2.GRB 030323
With a fluence of 1.2 × 10−6erg cm−2in the energy range 2 - 400 keV, GRB 030323
belongs to the faintest 20% of HETE GRBs. Its ’softness’ is 0.38, making it an X-Ray Rich
GRB. The eobs
p
value of ∼ 60 keV is not well constrained and could be as low as 20 keV or
as high as 200 keV.
4.Distance and energetics
Knowing the redshifts of GRB 020124 and GRB 030323 allows us to compute their
intrinsic spectral parameters: Ep, the peak energy of the νFν spectrum; Eiso, the isotropic-
equivalent radiated energy (in the energy range 1-10000 keV), and Nγthe isotropic-equivalent
3Following (Sakamoto et al. 2004a), we define the softness as the ratio Sx/Sγ,where Sxis the fluence in
the range 2-30 keV, and Sγis the fluence in the range 30-400 keV. Using this definition X-Ray Flashes have
a softness greater than 1 and X-Ray Rich GRBs have a softness in the range 0.33 to 1.

Page 7

– 7 –
photon number in the same energy range.
with Ωm = 0.3, ΩΛ = 0.7, and h0 = 0.65. For GRB 020124, we get Ep = 390+70
Eiso= 25 ± 6 × 1052erg, and Nγ= 34+20
270+600
In the following we adopt a flat cosmology
−120keV,
−10× 1058photons. For GRB 030323, we get Ep=
−2.5× 1058photons.
−180keV, Eiso= 3.2 ± 1 × 1052erg, and Nγ= 5.4+6
We note that GRB 020124 is intrinsically bright, similar to GRB 971214, which had
Eiso= 3 × 1053erg (Kulkarni et al. 1998; Dal Fiume et al. 2000; Amati et al. 2002); and
GRB 000131 which had Eiso= 11×1053erg (Andersen et al. 2000). The isotropic-equivalent
energy of GRB 030323 is however about an order of magnitude fainter, a fact that demon-
strates the ability of HETE to detect and localize high-redshift GRBs that are not at the
bright end of the GRB luminosity function.
We have checked whether GRB 020124 and GRB 030323 are compatible with the em-
pirical Eiso- Eprelation discovered by Amati et al. (2002), which can be expressed as E0.5
Ep∼ 1 (where Eisois measured in units of 1050erg and Epin keV; see also Lamb et al. 2005)
This is indeed the case with E0.5
GRB 030323 is not well determined and we cannot draw any conclusion based on this burst
(although for the sake of completeness we should mention that the best fit values give E0.5
Ep= 0.65). The spectral parameters of GRB 020124 are well determined, and the fact that
this burst follows the Eiso- Eprelation closely could indicate that this relation has little or
no evolution with the redshift.
iso/
iso/ Ep= 1.3 for GRB 020124. For GRB 030323 the Epof
iso/
4.1.The maximum distances at which GRB 020124 and GRB 030323 could
be localized
Fig. 4 shows the position of GRB 020124 and GRB 030323 (the two large triangles, GRB
020124 is the rightmost large triangle) in a fluence-softness diagram, among the population
of GRBs detected with HETE (Barraud et al. 2004). This figure also shows the positions
that these bursts would have in the same diagram if they had occured at redshifts 1, 2, 5,
10, and 20. The tracks of GRB 020124 and GRB 030323 in the figure are computed by
taking into account the spectral redshift of the bursts, and assuming that we can measure
the total fluence, even when the burst is at a high redshift. From this figure we can see that,
at redshift 1, GRB 020124 would have been one of the brightest GRBs detected by HETE,
while GRB 030323 would have been in the middle of the fluence distribution. Regarding
their hardness, we note that at a redshift of unity, GRB 020124 and GRB 030323 would
both have been unambiguously classified as ’GRBs’, rather than XRR-GRBs or XRFs.
Figure 4 can also be used to discuss the maximum redshift at which HETE could have

Page 8

– 8 –
detected GRB 020124 and GRB 030323. Based on this figure, we see that GRB 030323 is very
close to the boundary of the GRB population detected and localized with HETE. In contrast,
the fluence of GRB 020124 appears to remain well within the distribution of localized GRBs
up to redshift z ∼ 15. In reality, the fluence is not the best intensity indicator for GRB
detection because the trigger algorithm is essentially based on the search for excess counts
in short time intervals. In order to assess more precisely the maximum redshift at which
HETE could have detected GRB 020124, we have performed a detailed analysis including
the following steps:
- Compute the trigger threshold (number of counts) of FREGATE at the time of GRB
020124, on the 1.3 sec and on the 5.2 sec trigger timescales.
- Estimate the signal to noise ratio of GRB 020124 for its detection by FREGATE.
- Determine the redshift at which the counts from GRB 020124 would reach the trigger
threshold of FREGATE, taking into account the effects of the distance, of the time dilation
and of the spectral redshift of the photons.
This analysis shows that a burst 3.2 times fainter than GRB 020124 would have triggered
FREGATE in the 7-30 keV energy range in a time window of 5.2 sec. We thus conclude that
GRB 020124 could have been detected by FREGATE up to a redshift z=6.4. The trigger
scheme of HETE includes many more trigger possibilities than the simple trigger palette of
FREGATE, especially some using longer timescales for triggering. It is thus reasonable to
assume that GRB 020124 could have been detected by HETE up to redshift 7-8. At this
redshift HETE could have localized GRB 020124 because its localization capabilities are
more dependent on the fluence of the burst than on its peak flux, and because the fluence
decreases more slowly than the peak flux with the redshift. This is confirmed by figure 4
which shows that, at a redshift z=8, the fluence and the softness of GRB 020124 would have
been comparable with the fluences and softness of many GRBs localized by HETE.
We finally note that, at redshifts higher than 10, the fluence of GRB 020124 remains
comparable with the fluence of many GRBs detected and localized with HETE while its
peak flux is well below the detection threshold. A consequence of this fact is that a mission
can greatly improve its ability to detect high-z GRBs if it is able to localize long, faint,
soft transients. This is not the case for BeppoSAX and HETE-2 whose detection strategy
is mainly based on the search for count excesses on relatively short timescales. A strategy
based on the search for long, soft transients appearing in the image of the sky, as is the case
for SWIFT-BAT, appears more promising for detecting high-z GRBs.

Page 9

– 9 –
5.Afterglows and hosts of GRB 020124 and GRB 030323
The discussion in this section is mainly based on the information that has been published
in papers on the afterglows and hosts of the four GRBs with a measured redshift greater
than 3: GRB 971214 (Halpern et al. 1998; Kulkarni et al. 1998; Dal Fiume et al. 2000),
GRB 000131 (Andersen et al. 2000), GRB 020124 (Berger et al. 2002; Hjorth et al. 2003),
and GRB 030323 (Vreeswijk et al. 2004).
5.0.1. Spectroscopy of the afterglows
Thanks to fast localizations by HETE, the afterglows of GRBs 020124 and 030323 were
observed soon after the burst, while they were still bright. The identification of the afterglow
of GRB 020124 is due to Price et al. (2002) in data taken 13.5 hours after the burst, but the
first observation took place only two hours after the trigger and caught the afterglow at a
magnitude R=18.5 (Torii et al. 2002). The identification of the afterglow of GRB 030323 is
due to Gilmore et al. (2003) in data taken 9.5 hours after the burst. The afterglow was first
observed with ROTSE III, when it had a magnitude R=18.4, 6 hours after the burst. This
contrasts with the afterglows of GRB 971214, and GRB 000131, which were identified when
they had magnitudes R=22.1, and R=23.3, respectively. These quick detections allowed
for spectroscopy of the afterglows while they were still moderately bright (R=24 for GRB
020124 and R=21.5 for GRB 030323). The spectroscopic observations of GRB 020124 and
GRB 030323 resulted in high quality spectra and in the detection of strong Absorption Line
Systems (ALS) in the afterglows of both GRBs (Hjorth et al. 2003; Vreeswijk et al. 2004).
5.0.2.Beaming breaks
The light curves of GRB afterglows often display jet breaks, attributed to the confine-
ment of the relativistic outflow into a small cone. Rhoads (1997) has shown that jet breaks
can be used to estimate θj, the jet opening angle, and Eγ, the total energy output in γ-rays.
Eγis given by the following formula
Eγ= Eiso∗ (1 − cos(θj))
where θjis the opening angle of the jet in radians.
The afterglows of GRB 020124 and GRB 030323 were observed on several occasions dur-

Page 10

– 10 –
ing the few days following the bursts, but not frequently enough to unambiguously identify
a possible jet break. Berger et al. (2002) nevertheless argue that GRB 020124 might have a
jet break 10-20 days after the burst. The observations of GRB 030323, on the other hand,
give an afterglow slope s=−1.56±0.03 (Vreeswijk et al. 2004), showing that they took place
before a possible jet break (one expects s ∼ −2 after the break). Fig.1. of Vreeswijk et al.
(2004) shows that we can exclude a jet break in the afterglow, in the first 4 days following
the burst.
Frail et al. (2001) find that Eγis narrowly distributed around 5 × 1050erg in a sample
of GRBs with known redshifts. Bloom, Frail, and Kulkarni (2003) later revised this value to
1.3×1051erg. From the measured value of Eiso, we can calculate the opening angle that the
jet would have to have for Eγto be 1.3×1051erg. For GRB 020124, we find θj= 0.10 radians
(5.8 degrees), and θj= 0.29 radians (16 degrees) for GRB 030323. Following equation (1)
of Frail et al. (2001), and assuming a circumburst density of [0.1 cm−3], we find that the
expected break time is 2.6 days for GRB 020124, and 5.7 days for GRB 030323. With the
freedom allowed by the poor sampling of the light curves, and by the small (but real) scatter
in the size of the energy reservoir, we consider that these numbers do not contradict the
finding of Frail et al. (2001), and Bloom, Frail, and Kulkarni (2003) that there is a standard
radiated energy for GRBs.
5.0.3.Host galaxy identification
GRB 020124 and GRB 030323 have very faint hosts: R ≥ 29.5 for GRB 020124
(Bloom et al. 2002), and V=28.0 for GRB 030323 (Vreeswijk et al. 2004). The early lo-
calization of these two GRBs, and the quick identification of their afterglows, made possible
the identification of the host galaxy of GRB 030323, and placed stringent limits on the mag-
nitude of the host galaxy of GRB 020124. The faintness of the host galaxies of these two
bursts shows the impossibility to measure their redshifts from the spectroscopy of their host
galaxies. GRB 020124 and GRB 030323 are examples of GRBs whose redshifts can only
be measured at early times from the spectrum of the afterglow (unlike what was done for
GRB 971214). The fact that two of the four GRBs known with z ≥ 3 occurred in faint
galaxies, may indicate that a non-negligible fraction of star formation takes place in such
faint galaxies. Gamma-ray bursts appear to be a privileged way to identify this population.

Page 11

– 11 –
6. Conclusions
This paper describes the temporal and spectral properties of GRB 020124 and GRB
030323, two GRBs at redshift z > 3 detected and localized with HETE. These two events
are found to be fully consistent with the properties of the rest of the GRB population detected
with HETE.
We have used the chain of events which successfully led to the measurement of the
redshifts of GRB 020124 and GRB 030323 as a baseline to discuss the conditions required
for the identification of high-z GRBs. Our two main conclusions are summarized below.
The fast localization of GRB 020124 and GRB 030323 allowed the quick identification
and the early spectroscopy of their afterglows. We see a posteriori that this was the only way
to measure their redshifts, given the faintness of their host galaxies. In these cases, contrary
to the case of GRB 971214, we could not rely on the spectroscopy of the host galaxy to
measure the redshifts, and this might well be the case for the majority of high-z GRBs.
Study of GRB 020124 shows that even instruments of modest size like FREGATE or
the WXM are able to detect and localize GRBs up to z=7-8, if indeed GRBs occur at these
redshifts. The study of the tracks with redshift of the peak flux and of the fluence of GRB
020124 provides insight into the strategy to be used for the detection of high-z GRBs. A
strategy that relies mainly on the search for count excesses in short time intervals does not
appear to be the most appropriate. A strategy based on the imaging of faint, soft transients
lasting minutes appears more promising.
Acknowledgments
The HETE mission is supported in the U.S. by NASA contract NASW-4690; in Japan,
in part by the Ministry of Education, Culture, Sports, Science, and Technology Grant-in-Aid
13440063; and in France, by CNES contract 793-01-8479. KH is grateful for HETE support
under Contract MIT-SC-R-293291, for Ulysses support under JPL Contract 958056, and for
IPN support under NASA grant FDNAG5-11451. G. Pizzichini acknowledges support by
the Italian Space Agency.
REFERENCES
Amati, L., Frontera, F., Tavani, M., et al. 2002, A&A, 390, 81

Page 12

– 12 –
Andersen, M.I., Hjorth, J., Pedersen, H., et al. 2000, A&A, 364, L54
Band, D., Matteson, J., Ford, L. et al. 1993, ApJ, 413, 281
Barraud, C., Olive, J-F., Lestrade, J.P. et al. 2003, A&A, 400, 1021
Barraud, C., Atteia, J-L., Olive, J-F. et al. 2004, AIP Conf. Proc. 727: Gamma-Ray Bursts:
30 Years of Discovery, 727, 81
Berger, E., Kulkarni, S., Bloom, J. et al. 2002, ApJ, 581, 981
Bloom, J.S., Berger, E., and Kulkarni, S.R. 2002, GCN Circ. 1452
Bloom, J.S., Frail, D.A., and Kulkarni, S.R. 2003, ApJ, 594, 674
Dal Fiume, D., Amati, L., Antonelli, L.A. et al. 2000, A&A, 355, 454
Frail, D.A., Kulkarni, S.R., Sari, R. et al. 2001, ApJ, 562, L55
Frontera, F., Amati, L., Costa, E. et al. 2000, ApJS, 127, 59
Gilmore, A., Kilmartin, P., Henden, A. 2003, GCN Circ. 1949
Graziani, C., Shirasaki, Y., Matsuoka, M. et al. 2003, GCN Circ. 1956
Halpern, J.P., Thorstensen, J.R., Helfand, D.J., and Costa, E. 1998, Nature, 393, 41
Hjorth, J., Møller, P., Gorosabel, J. et al. 2003, ApJ, 597, 699
Hurley, K. et al. 2002, GCN Circ. 1223
Kulkarni, S.R., Djorgovski, S.G., Ramaprakash, A.N., et al. 1998, Nature, 393, 35
Lamb, D. Q., & Reichart, D. E. 2000, ApJ, 535, 1
Lamb, D. Q., Sakamoto, T., Atteia, J-L. et al. 2005, ApJ, in preparation
Preece, R. D., Briggs, M.S., Mallozzi, R.S. et al. 2000, ApJS126, 19
Price, P.A., Fox, D.W., Yost, S.A. et al. 2002, GCN Circ. 1221
Rhoads, J. 1997, ApJ, 487, L1
Ricker, G.R., Atteia, J-L., Kawai, N. et al. 2002, GCN Circ. 1220
Sakamoto, T., Lamb, D.Q., Suzuki, M. et al. 2004a, ApJ, 602, 875

Page 13

– 13 –
Sakamoto, T., Lamb, D.Q., Graziani, C. et al. 2004b, ApJ, submitted (astro-ph/0409128)
Torii, K., Kato, T., Yamaoka, H. and Yoshida, A. 2002, GCN Circ. 1378
Vreeswijk, P., Ellison, S.L., Ledoux, C. et al. 2004, A&A, 419, 927
This preprint was prepared with the AAS LATEX macros v5.2.

Page 14

– 14 –
Table 1.Temporal Properties of GRB 020124 and GRB 030323.
GRB / InstrumentEnergy
(keV)
t90
(s)
t50
(s)
GRB 020124
HETE WXM2–25
2–5
5–10
10–25
6–400
6–15
15–30
30–85
85–400
50.2 ± 2.3
41.8 ± 0.4
50.4 ± 8.0
32.5 ± 1.2
49.4 ± 1.3
51.4 ± 1.4
52.9 ± 2.0
45.6 ± 0.7
43.0 ± 6.1
18.6 ± 1.1
23.5 ± 1.7
11.7 ± 3.0
16.7 ± 3.5
22.6 ± 1.0
26.0 ± 2.7
23.2 ± 2.4
22.2 ± 1.3
19.2 ± 1.8
HETE FREGATE
GRB 030323
HETE WXM2–25
2–5
5–10
10–25
6–400
6–30
30–400
32.6 ± 2.7
31.5 ± 0.9
36.1 ± 0.6
19.5 ± 2.0
15.0 ± 2.6
12.8 ± 2.5
12.2 ± 3.6
13.9 ± 1.6
16.2 ± 0.8
19.4 ± 1.3
12.5 ± 1.3
7.1 ± 1.2
6.6 ± 1.5
5.2 ± 1.6
HETE FREGATE
Note. — Errors are 1-σ. The significantly longer dura-
tion measured by the WXM for GRB 030323 is explained
by the better sensitivity of this instrument to low energy
photons which are hardly detected with FREGATE.

Page 15

– 15 –
Table 2.Spectral models for GRB020124 and GRB 030323. See section 3.2 for a
description of the spectral models.
GRB / ModelAlphaE0
Beta
GRB 020124 PLE
−0.79+0.15
−0.14
87+34
−21
N/A
GRB 020124 GRBM
−0.87−0.16
+0.19
82+31
−31
−2.6−0.65
GRB 030323 PLE
−0.80+0.8
−0.83
44+90
−26
N/A
GRB 030323 GRBMa
−0.96+1.31
−0.85
60−45
−2.3 (frozen)
Note. — Errors are for 90% confidence, there is no upper
limit for Beta.
constrained. β is frozen at −2.3, and there is no upper limit
on E0.
aFor this model, the parameters are poorly
Table 3.Emission Properties of GRB 020124 and GRB 030323.
Energy
(keV)
Peak photon Flux
(ph cm−2s−1)
Peak energy Flux
(erg cm−2s−1)
Energy Fluence
(erg cm−2)
GRB 020124
2–306.9 ± 1.6
2.5 ± 0.40
1.4 ± 0.28
1.8 ± 0.18 × 10−7
4.5 ± 0.46 × 10−7
3.2 ± 0.33 × 10−7
2.0+0.14
6.1+0.88
4.7+0.82
−0.14× 10−6
−0.76× 10−6
−0.82× 10−6
30–400
50–300
GRB 030323
2–303.4 ± 2.1
0.49 ± 0.22
0.29 ± 0.15
.57 ± .16 × 10−7
1.5 ± .43 × 10−7
1.0 ± .29 × 10−7
3.4+1.3
8.9+3.8
6.5+2.8
−1.2× 10−7
−3.5× 10−7
−2.8× 10−7
30–400
50–300
Note. — Errors are given at the 90% confidence level.