Content uploaded by Titus Galama
Author content
All content in this area was uploaded by Titus Galama
Content may be subject to copyright.
Content uploaded by Titus Galama
Author content
All content in this area was uploaded by Titus Galama
Content may be subject to copyright.
arXiv:astro-ph/9904286v1 21 Apr 1999
The X-ray, Optical and Infrared Counterpart to GRB 980703
P.M. Vreeswijk1, T.J. Galama1, A. Owens2, T. Oosterbroek2, T.R. Geballe3, J. van
Paradijs1,4, P.J. Groot1, C. Kouveliotou5,6, T. Koshut5,6, N. Tanvir7, R.A.M.J. Wijers8, E.
Pian9, E. Palazzi9, F. Frontera9,10, N. Masetti9, C. Robinson4,6, M. Briggs4,6, J.J.M. in ’t
Zand11, J. Heise11 , L. Piro12, E. Costa12, M. Feroci12, L.A. Antonelli12, K. Hurley13, J.
Greiner14, D.A. Smith15 , A.M. Levine15, Y. Lipkin16, E. Leibowitz16, C. Lidman17 , A.
Pizzella17, H. B¨ohnhardt17, V. Doublier17, S. Chaty18,19, I. Smail20, A. Blain21, J.H.
Hough22, S. Young23, N. Suntzeff24
– 2 –
1Astronomical Institute ‘Anton Pannekoek’, University of Amsterdam, & Center for High
Energy Astrophysics, Kruislaan 403, 1098 SJ Amsterdam, The Netherlands
2Astrophysics Division, Space Science Department of ESA, European Space Research and
Technology Centre, 2200 AG Noordwijk, The Netherlands
3Joint Astronomy Centre, 660 N. A’ohoku Place, Hilo, Hawaii 96720, USA
4Physics Department, University of Alabama in Huntsville, Huntsville AL 35899, USA
5Universities Space Research Association
6NASA/MSFC, Code ES-84, Huntsville AL 35812, USA
7Institute of Astronomy, Madingley Road, Cambridge CB3 0HA, UK
8Astronomy Program, State University of NY, Stony Brook, NY 11794-3800, USA
9Istituto Tecnologie e Studio Radiazioni Extraterrestri (TESRE), CNR, Via P. Gobetti
101, 40 129 Bologna, Italy
10Dipartimento di Fisica Universita’ di Ferrara, Via Paradiso 12, 44100 Ferrara, Italy
11Space Research Organisation Netherlands (SRON), Sorbonnelaan 2, 3584 CA Utrecht,
The Netherlands
12Istituto di Astrofisica Spaziale, CNR, Via Fosso del Cavaliere, Roma, I-00133, Italy
13University of California at Berkeley, Space Sciences Laboratory, Berkeley, CA, USA
94720-7450
14Astrophysikalisches Institut Potsdam, D-14482 Potsdam, Germany
15Massachusetts Institute of Technology, 77 Mass. Avenue, Cambridge, MA 02139-4307,
USA
16Wise Observatory, Tel Aviv University, Ramat Aviv, Tel Aviv 69978, Israel
17ESO, Casilla 19001, Santiago 19, Chile
18DAPNIA/Service d’Astrophysique, CEA/Saclay, F-91191 Gif-Sur-Yvette, France
19Centre d’Etude Spatiale des Rayonnements, 9, avenue du Colonel Roche BP 4346, F-31
– 3 –
Received ; accepted
028 Toulouse Cedex 4, France
20University of Durham, South Road, Durham, DH1 3LE, UK
21Cavendish Laboratory, Madingley Road, Cambridge CB3 0HE, UK
22Physics & Astronomy, University of Hertfordshire, Hatfield, AL10 9AB, UK
23Cerro Tololo Interamerican Observatory, Casilla 603, La Serena, Chile
– 4 –
ABSTRACT
We report on X-ray, optical and infrared follow-up observations of
GRB 980703. We detect a previously unknown X-ray source in the GRB error
box; assuming a power law decline we find for its decay index α < –0.91 (3σ).
We invoke host galaxy extinction to match the observed spectral slope with the
slope expected from ‘fireball’ models. We find no evidence for a spectral break
in the infrared to X-ray spectral range on 1998 July 4.4, and determine a lower
limit of the cooling break frequency: νc>1.3 ×1017 Hz. For this epoch we
obtain an extinction of AV= 1.50 ±0.11. From the X-ray data we estimate
the optical extinction to be AV= 20.2+12.3
−7.3, inconsistent with the former value.
Our optical spectra confirm the redshift of z=0.966 found by Djorgovski et al.
(1998). We compare the afterglow of GRB 980703 with that of GRB 970508 and
find that the fraction of the energy in the magnetic field, ǫB<6×10−5, is much
lower in the case of GRB 980703, which is a consequence of the high frequency
of the cooling break.
Subject headings: Gamma-rays bursts—gamma-rays:observations—radiation
mechanisms:non-thermal
– 5 –
1. Introduction
Several properties of gamma-ray burst (GRB) afterglows can be well explained by
‘fireball’ models, in which a relativistically expanding shock front, caused by an energetic
explosion in a central compact region, sweeps up the surrounding medium and accelerates
electrons in a strong synchrotron emitting shock (M´eszar´os and Rees 1994, Wijers, Rees
and M´eszar´os 1997; Sari, Piran and Narayan 1998; Galama et al. 1998a). The emission
shows a gradual softening with time, corresponding to a decrease of the Lorentz factor
of the outflow. Most X-ray and optical/infrared (IR) afterglows display a power law
decay (except GRB 980425, which is most likely associated with the peculiar supernova
SN 1998bw; Galama et al. 1998b).
Spectral transitions in the optical/IR have been detected for the afterglows
of GRB 970508 (Galama et al. 1998a; Wijers and Galama 1998) and GRB 971214
(Ramaprakash et al. 1998). These have been explained by the passage through the
optical/IR waveband of the cooling break, at νc(for GRB 970508) and the peak of the
spectrum, at νm(for GRB 971214); see Sari, Piran & Narayan (1998) and Wijers & Galama
(1998) for their definition. For GRB 970508 the observed break is in excellent agreement
with such ‘fireball’ models, while for GRB 971214 an exponential extinction has been
invoked to explain the discrepancy between the expected and observed spectral index.
GRB 980703 was detected on July 3.182 UT with the All Sky Monitor (ASM) on the
Rossi X-ray Timing Explorer (RXTE; Levine et al. 1998), the Burst And Transient Source
Experiment (BATSE, trigger No. 6891; Kippen et al. 1998), Beppo SAX (Amati et al. 1998)
and Ulysses (Hurley et al. 1998). The burst as seen by BATSE consisted of two pulses,
each lasting approximately 100 sec., with a total duration of about 400 sec. (Kippen et al.
1998). The first pulse had significant sub-structure, whereas the second, weaker episode was
relatively smooth. This double-peak morphology has also been seen with the Beppo SAX
– 6 –
Gamma Ray Burst Monitor (Amati et al. 1998). BATSE measured a peak flux of (1.9 ±
0.1) ×10−6erg cm−2s−1(25 - 1000 keV) and fluence of (4.6 ±0.4) ×10−5erg cm−2(>
20 keV), consistent with the Beppo SAX GRBM measurement. A time resolved spectral
analysis of the burst will be presented elsewhere (Koshut et al. 1999).
Observations with the Narrow-Field Instruments (NFIs) of Beppo SAX showed a
previously unknown X-ray source (Galama et al. 1998c) inside both the ASM error box
and the InterPlanetary Network annulus (Hurley et al. 1998). Frail et al. (1998a; see also
Zapatero Osorio et al. 1998) subsequently reported the discovery of a radio (6 cm) and
optical (R-band) counterpart to GRB 980703.
Here we report X-ray (0.1-10 keV), optical (VRI), and infrared (JHK) follow-up
observations of GRB 980703. In §2 we report our NFI X-ray observations of the ASM error
box, and §3 is devoted to the description and results of our spectroscopic and photometric
optical/IR monitoring campaign. We discuss the results of these observations in §4.
2. X-ray observations
We observed the ASM error box of GRB 980703 with the Beppo SAX Low- and
Medium Energy Concentrator Spectrometers (LECS, 0.1-10 keV, Parmar et al. 1997;
MECS, 2-10 keV, Boella et al. 1997) on July 4.10-5.08 UT (starting 22 hrs after the burst)
and on July 7.78-8.71 UT. The LECS and MECS data show a previously unknown X-ray
source 1SAX J2359.1+0835 at R.A. = 23h59m06.
s8, Decl. = +08◦35′45′′(equinox J2000.0),
with an error radius of 50′′ . The field also contains the sources 1SAX J2359.9+0834 at
R.A. = 23h59m59s, Decl.= +08◦34′03′′ , and 1SAX J0000.1+0817 at R.A. = 00h00m04s,
Decl.=+08◦17′14′′. Both are outside the ASM error box, do not show any variability
and coincide with the known ROSAT sources, 1RXS J235959.1+083355 and 1RXS
– 7 –
J000007.0+081653, respectively.
We extracted 1SAX J2359.1+0835 data at the best fit centroids with radii of 8′(LECS)
and 2′(MECS). To analyze the spectrum we binned the data into channels, such that
each contained at least 20 counts. Using the separate standard background files of the
spectrometers, we simultaneously fitted the LECS and MECS data of July 4-5 UT, with
a power law spectrum and a host galaxy absorption cut-off, using a redshift of z= 0.966
(Djorgovski et al. 1998; see also §3 of this paper). In the fit we fixed the Galactic foreground
absorption at NH= 3.4 ×1020 cm−2(AV= 0.19, as inferred from the dust maps of Schlegel,
Finkbeiner & Davis 19981, and the AV-NHrelation of Predehl & Schmitt 1995). The
average spectrum can be modelled with a photon index Γ = 2.51 ±0.32 and a host galaxy
column density NH(host) = 3.6+2.2
−1.3×1022 cm−2, corresponding to AV(host) = 20.2+12.3
−7.3
(local to the absorber). Modelling the spectrum without NH(host) results in a very poor
fit. We did not account for a possible small position dependent error in the relative flux
normalizations between the LECS and MECS, which is only a few percent near the center
of the image. A change of 25% in the assumed Galactic foreground absorption does not
affect the output values of the fit parameters. For the second epoch of NFI observations we
kept the position fixed at the position determined from the first epoch; we find FX<1.1 ×
10−13 erg cm−2s−1(3σ). Fitting a power law model to the light curve including the upper
limit, we obtain α < –0.91 for the decay index. The X-ray light curve is shown in Fig 1. We
checked for the presence of the 6.4 keV K line at the redshifted energy of 3.26 keV, but do
not detect it. The upper limit on its flux is 8.3 ×10−6photons cm−2s−1(90% confidence
level), corresponding to an equivalent width of 532 eV in the observer’s frame.
1see http://astro.berkeley.edu/davis/dust/index.html
– 8 –
3. Optical and infrared observations
We observed the field of GRB 980703 with the Wise Observatory 1-m telescope (in I);
the 3.5-m New Technology Telescope (NTT; in V, I and H), the 2.2-m (in H and Ks) and
Dutch 90-cm (in gunn i) telescopes of ESO (La Silla); the CTIO 0.9-m telescope (in R) and
UKIRT (in H, J and K).
The optical images were bias-subtracted and flat-fielded in the standard fashion. The
infrared frames were reduced by first removing bad pixels and combining about five frames
around each object image to obtain a sky image. This sky image was then subtracted after
scaling it to the object image level; the resulting image was flat-fielded. Four reference
stars were used to obtain the differential magnitude of the optical transient (OT) in each
frame. These stars were calibrated by observing the standard stars PG2331+055 (in V and
I; Landolt 1992), FS2 and FS32 (in J, H and K; Casali & Hawarden 1992). We used the
R-band calibration of Rhoads et al. (1998). The offsets in right ascension and declination
from the OT, and the apparent standard magnitudes outside the Earth’s atmosphere of the
reference stars are listed in Table 1.
The light curves of the OT are shown in Fig. 2 (see Table 2 for a list of the
magnitudes). In view of the flattening of the light curves after t∼5 days we fitted a model
Fν=F0·tα+Fgal to our own I and H band light curves (in these bands we have sufficient
data for a free parameter fit). Here tis the time since the burst in days, and Fgal is the flux
of the underlying host galaxy. The values for m0=−2.5·log (F0) + C, the decay index α
and mgal =−2.5·log (Fgal) + C, are listed in Table 3. The photometric calibration, which
determines C, has been taken from Bessell (1979) for V, R and I and Bessell & Brett (1988)
for J, H and K. The weighted mean value of αfor the I and H-bands equals −1.61 ±0.12,
while an I and H-band joint fit, with a single power law decay index, gives α=−1.63 ±0.12
(χ= 15.4/16). For the V, R, J and K bands we fixed αat –1.61, included also data from the
– 9 –
literature, and fitted m0and mgal. The fits are shown as solid lines in Fig. 2. We note that
when we fit all three parameters to the R band data (mainly data from the literature) we
obtain a temporal slope of −1.94 ±0.22, consistent with the adopted value of −1.61 ±0.12.
However, we do not include this value in the average, since the R band magnitudes are
taken from the literature and thus not consistently measured.
For four epochs we have reconstructed the spectral flux distribution of the OT (times
t1, t2, t3and t4in Fig. 2, corresponding to 1998 July 4.4, 6.4, 7.6 and 8.4 respectively). The
host galaxy flux, obtained from the fits to the light curves, was subtracted. We corrected
the OT fluxes for Galactic foreground absorption (AV= 0.19). If more than one value per
filter was available around the central time of the epoch, we took their weighted average.
All values were brought to the same epoch by applying a correction using the slope of the
fitted light curve. For the first epoch (t1) we fitted the resulting spectral flux distribution
with a power law, Fν∝νβ, and find β=−2.71 ±0.12.
We took three 1800 sec. spectra of the OT with the NTT, around July 8.38 UT. The
#3 grism that was used has a blaze wavelength of 4600 ˚
A and a dispersion of 2.3 ˚
A/pixel.
The slit width was set at 1′′. The three spectra were bias-subtracted and flat-fielded in the
usual way. The co-added spectrum was then extracted the same way as the standard star
Feige 110. We wavelength calibrated the spectrum with a Helium/Argon lamp spectrum,
with a residual error of 0.03 ˚
A. The spectrum was flux calibrated with the standard star
Feige 110 (Massey et al. 1988). We estimate the accuracy of the flux calibration to be 10%.
The wavelength and flux calibrated spectrum shows one clear emission line at 7330.43 ±
0.14 ˚
A with a flux of 3.6 ±0.4 ×10−16 erg cm−2s−1. At the redshift z= 0.966 determined
by Djorgovski et al. (1998) this is the λ3727 line of [O II]; our wavelength measurement
corresponds to z= 0.9665 ±0.0005.
– 10 –
4. Discussion
From the optical/IR light curves presented in §3 we have obtained an average power
law decay constant α=−1.61 ±0.12. This value is consistent with the ones derived by
Bloom et al. (1998) (αR=−1.22 ±0.35, and αI=−1.12 ±0.35) and Castro-Tirado et al.
(1999) (αR=−1.39 ±0.3, and αH=−1.43 ±0.11).
If we make the assumption that the OT emission is due to synchrotron radiation
from electrons with a power law energy distribution (with index p), one expects a relation
between p, the spectral slope β, and the decay constant α(Sari, Piran and Narayan
1998). We assume that our observations are situated in the slow cooling, low frequency
regime (e.g. for GRB 970508 this was already the case after 500 sec.; Galama et al.
1998a). One must distinguish two cases: (i) both the peak frequency νmand the cooling
frequency νcare below the optical/IR waveband. Then p= (−4α+ 2)/3 = 2.81 ±0.16
and β=−p/2 = −1.41 ±0.08, (ii) νmhas passed the optical/IR waveband, but νchas not
yet. In that case p= (−4α+ 3)/3 = 3.15 ±0.16 and β=−(p−1)/2 = −1.07 ±0.08. In
both cases the expected value of βis inconsistent with the observed β=−2.71 ±0.12.
Following Ramaprakash et al. (1998) we assume that the discrepancy is caused by
host galaxy extinction (note that we have already corrected the OT fluxes for Galactic
foreground absorption). To determine the host galaxy absorption we first blueshifted the
OT flux distribution to the host galaxy rest frame (using z= 0.966), and then applied an
extinction correction using the Galactic extinction curve of Cardelli, Clayton & Mathis
(1989), to obtain the expected spectral slope β. For epoch t1(July 4.4 UT), we obtain
AV= 1.15 ±0.13 and AV= 1.45 ±0.13 for the cases (i) and (ii), respectively (see Fig. 3).
In case (i) we find that an extrapolation of the optical flux distribution to higher
frequencies predicts an X-ray flux that is significantly below the observed value, whereas in
case (ii) the extrapolated and observed values are in excellent agreement. The mismatch in
– 11 –
case (i) is in a direction that cannot be interpreted in terms of the presence of a cooling
break between the optical and X-ray wavebands. When we include the X-ray data point
in the fit to obtain a more accurate determination of AV, we find AV= 1.50 ±0.11, and
β=−1.013 ±0.016. We conclude that the optical/IR range is not yet in the cooling
regime, and so p= 3.15 ±0.16. Where would the cooling frequency, νc, be located? The
X-ray photon index, Γ = 2.51 ±0.32, corresponding to a spectral slope of β=−1.51 ±0.32
suggests that perhaps the X-ray waveband is just in the cooling regime, in which case the
expected local X-ray slope would be β=−p/2 = −1.57 ±0.08, while if not in the cooling
regime, it would be β=−(p−1)/2 = −1.07 ±0.08. However, the large error on the
measured X-ray spectral slope would also allow the cooling break to be above 2-10 keV. We
estimate the (2σ) lower limit to the cooling frequency to be νc>1.3 ×1017 Hz (hνc>0.5
keV).
We performed the same analysis for the other epoch (t4) with X-ray data (see Fig.
3). At this epoch, the X-ray upper limit does not allow us to discriminate between the
two cases. However, we can still estimate a lower limit to the cooling break from its time
dependence: νc∝t−1/2, which would allow the break to drop to νc>6.3 ×1016 Hz only,
between epoch t1and t4.
On the basis of our analysis we conclude that there is no strong evidence for a cooling
break between the optical/IR and the 2-10 keV passband before 1998 July 8.4 UT. This
conclusion is at variance with the inference of Bloom et al. (1998), who infer from their
fits that there is a cooling break at about 1017 Hz. Upon closer inspection, there is no real
disagreement: Bloom et al. found a slightly shallower temporal decay, and therefore a bluer
spectrum of the afterglow, which causes their extrapolated optical spectrum to fall above
the X-ray point. However, their error of 0.35 on the temporal decay leads to an error of
0.24 on their predicted spectral slope, and this means that a 1σsteeper slope in their Fig.
– 12 –
2 would be consistent with no detected cooling break.
Assuming that the spectral slope (β=−1.013 ±0.016) did not change during the time
spanned by the four epochs t1- t4(as suggested by the lack of evidence for a break in the
light curve during this timespan) we have derived the V-band extinction AVas a function of
time: AV= 1.50 ±0.11, 1.38 ±0.35, 0.84 ±0.29 and 0.90 ±0.25 for the epochs 1 through
4, respectively. Fitting a straight line through these, we obtain a slope of –0.16 ±0.06,
i.e. not consistent with zero at the 98.8% confidence level. Such a decrease of the optical
extinction, AV, might be caused by ionization of the surrounding medium (Perna and Loeb
1997).
The V-band extinction AV= 20.2+12.3
−7.3, derived from the host galaxy NHfit to the
MECS and LECS data (July 4-5 UT) is not in agreement with AV= 1.50 ±0.11 as derived
from the fit from the optical spectral flux distribution. This may be due to a different dust
to gas ratio for the host galaxy of GRB 980703, or a higher abundance than normal of the
elements that cause the X-ray absorption.
With the above derived constraint on νcwe can partially reconstruct the broad-band
flux distribution of the afterglow of GRB 980703: from the radio observations of Frail et al.
(1998b) at 1.4, 4.86 and 8.46 GHz, we determine the self-absorption frequency νaand its
flux Fνafrom the fit Fν=Fνa(ν/νa)2(1 −exp[−(ν/νa)−5/3]) to the low-energy part of the
spectrum (e.g. Granot, Piran and Sari 1998). We have used averages of the 1.4 and 4.86
GHz observations to obtain best estimates of the radio flux densities as particularly those
frequencies suffer from large fluctuations due to interstellar scintillation (Frail et al. 1998b).
We find νa= 3.68 ±0.33 GHz and Fνa= 789 ±42 µJy. (The fit is shown in Fig. 4.) The
intersection of the extrapolation from the low-frequency to the high-frequency fit gives a
rough estimate of the peak frequency, νm∼4×1012 Hz, and of the peak flux, Fνm∼8 mJy
(see Fig. 4). By assuming such a simple broken power law spectrum the peak flux density
– 13 –
will likely be overestimated (realistic spectra are rounder at the peak); it is clear from Fig.
4 that 1 < Fνm<8 mJy.
Following the analysis of Wijers and Galama (1998) we have determined the following
intrinsic fireball properties: (i) the energy of the blast wave per unit solid angle: E>5×
1052 erg/(4πsterad), (ii) the ambient density: n > 1.1 nucleons cm−3, (iii) the percentage
of the nucleon energy density in electrons: ǫe>0.13, and (iv) in the magnetic field:
ǫB<6×10−5. The very low energy in the magnetic field, ǫB, is a natural reflection of the
high frequency of the cooling break νc.
We have compared this afterglow spectrum with that of GRB 970508. Scaling the
latter in time according to νa∝t0,νm∝t−3/2and νc∝t−1/2, the results of GRB 970508
(Galama et al. 1998a; see also Granot, Piran and Sari 1998) would correspond to νa∼2.3
GHz, νm= 2.8×1012 Hz, νc= 4.8×1014 Hz, and Fνm= 1.3 mJy. In this calculation we
have corrected for the effect of redshift (see Wijers and Galama 1998) such that the values
represent GRB 970508, were it at the redshift of GRB 980703 and observed 1.2 days after
the event. The greatest difference between the two bursts is in the location of the cooling
frequency, νc.
The observations on the United Kingdom Infrared Telescope, which is operated by the
Joint Astronomy Centre on behalf of the U.K. Particle Physics and Astronomy Research
Council, were carried out in Service mode by UKIRT staff. The BeppoSAX satellite is a
joint Italian and Dutch programme. PMV is supported by the NWO Spinoza grant. TJG
is supported through a grant from NFRA under contract 781.76.011. CK acknowledges
support from NASA grant NAG 5-2560. TO acknowledges an ESA Fellowship. KH is
grateful for support under JPL Contract 958056 for Ulysses, and under NASA grant NAG
5-1560 for IPN operations.
– 14 –
REFERENCES
Amati, L., et al. 1998, GCN circular, 146
Bessell, M.S. 1979, PASP, 91, 589
Bessell, M.S. & Brett, J.M. 1988, PASP, 100, 1134
Bloom, J.S., et al. 1998, ApJ, 508, L21
Boella, G., et al. 1997, A&AS, 122, 299
Cardelli, J.A., Clayton, G.C. & Mathis, J.S. 1989, ApJ, 345, 245
Castro-Tirado, A.J. et al. 1999, ApJ, 511, L85
Casali, M.M. & Hawarden, T.G. 1992, JCMT-UKIRT Newsletter, No. 3, 33
Djorgovski, S.G., et al. 1998, ApJ, 508, L17
Frail, D.A., et al. 1998a, GCN circular, 128
Frail, D.A., et al. 1998b, GCN circular, 141
Galama, T.J., et al., 1998a, ApJ, 500, L97
Galama, T.J., et al. 1998b, Nature, 395, 670
Galama, T.J., et al. 1998c, GCN circular, 127
Granot, J., Sari, T. and Sari, R. 1998, submitted, astro-ph/9808007
Henden, A.A., et al. 1998, GCN circular, 131
Hurley, K., et al. 1998, GCN circular, 125
Kippen, M., et al. 1998, GCN circular, 143
Koshut, T., et al. 1999, in preparation
Landolt, A.U. 1992, AJ 104, 340
Levine, A., et al. 1998, IAU Circ., 6966
Massey, P., et al. 1988, ApJ, 328, 315
M´eszar´os, P. & Rees, M.J. 1994, MNRAS, 269, L41
Parmar, A.N., et al. 1997, A&AS, 122, 309
Pedersen, H., et al. 1998, GCN circular, 142
– 15 –
Perna, R. & Loeb, A. 1998, ApJ, 501, 467
Predehl, P. and Schmitt, J.H.M.M. 1995, A&A, 293, 889
Ramaprakash, A.N., et al. 1998, Nature, 393, 43
Rhoads, J., et al. 1998, GCN circular, 144
Sari, R., Piran, T., Narayan, R. 1998, ApJ, 497, L17
Schlegel, D.J., Finkbeiner, D.P. and Davis, M. 1998, ApJ, 500, 525
Sokolov, V., et al. 1998, GCN circular, 147
Wijers, R.A.M.J. and Galama, T.J. 1998, in press, ApJ
Wijers, R.A.M.J., Rees, M.J. and M´eszar´os, P. 1997, MNRAS, 288, L51
Zapatero Osorio, M.R., et al. 1998, IAU Circ., 6967
– 16 –
Fig. 1.— The 2-10 keV light curve of GRB 980703. Time and flux are on a logarithmic scale.
filter star 1 star 2 star 3 star 4
∆ R.A.(′′) –18.0 –11.7 –8.0 –13.9
∆ Decl.(′′) 3.9 –9.7 –21.7 –31.8
V 21.33 ±0.06 17.02 ±0.05 22.06 ±0.08 22.68 ±0.12
R 20.39 ±0.04 16.64 ±0.02 20.72 ±0.05
I 19.55 ±0.07 16.30 ±0.05 19.21 ±0.06 20.01 ±0.08
J 18.45 ±0.13 15.75 ±0.11 17.52 ±0.12 18.25 ±0.13
H 17.85 ±0.12 15.44 ±0.10 16.96 ±0.11 17.69 ±0.12
K 17.71 ±0.13 15.41 ±0.12 16.72 ±0.12 17.52 ±0.13
Table 1: The magnitudes and offset from the OT in arc seconds of the four comparison
stars used. The error is the quadratic average of the measurement error (Poisson noise) and
a constant offset, which we estimate to be 0.05 for the optical passbands and 0.1 for the
infrared filters.
– 17 –
Fig. 2.— V, R, I, J, H and K light curves of GRB 980703. The filled symbols denote our
data, while the open symbols represent data taken from the literature (Zapatero Osorio et
al. 1998; Rhoads et al. 1998; Henden et al. 1998; Bloom et al. 1998; Pedersen et al. 1998;
Djorgovski et al. 1998; Sokolov et al. 1998 & private communication). For each filter a
power law model plus a constant: Fν=F0·tα+Fgal is fitted (solid lines). The fit parameters
are listed in Table 3. The times t1- t4, at which we have reconstructed the spectral flux
distribution of the OT, are indicated by the dashed lines.
– 18 –
Fig. 3.— Left figure: Broad-band spectrum of GRB 980703 at July 4.4 UT (i.e., at t1in
Fig. 2). The open symbols are the R, I and H OT fluxes (interpolated to July 4.4, corrected
for Galactic foreground absorption and the host galaxy flux) and the MECS (2-10 keV) de-
absorbed flux (the absorption correction is 7%). The filled symbols are obtained by invoking
an interstellar extinction, AV, to force the slope of the data points to take on the two possible
theoretical spectral slopes. The two slopes βand their 1σerrors are indicated by the solid
and dotted lines. Right figure: Broad-band spectrum of GRB 980703 at July 8.4 UT (i.e.,
at t4in Fig. 2). The open symbols are the V, R, I, J, H and K OT fluxes and the MECS
(2-10 keV) de-absorbed 3σupper limit.
– 19 –
Fig. 4.— Radio to X-ray spectrum of GRB 980703 at July 4.4 UT (i.e., at t1in Fig. 2).
Shown are data from Fig. (3) as well as 1.4, 4.86 and 8.46 GHz observations from Frail et
al. (1998b). The fit Fν=Fνa(ν/νa)2(1 −exp[−(ν/νa)−5/3]) to the low-energy part of the
spectrum with νa= 3.68 ±0.33 GHz and Fνa= 789 ±42 µJy is shown by the dotted line.
The best fit to the optical/IR and X-ray data is also shown.
– 20 –
UT date magnitude filter exp. time seeing telescope/reference
(1998 July) (seconds) (′′)
4.059 20.07 ±0.19 I 2100 2.39 Wise 1-m
4.347 20.43 ±0.04 I 900 1.16 ESO NTT (EMMI)
4.359 20.49 ±0.03 I 900 1.10 ESO NTT (EMMI)
4.372 20.54 ±0.03 I 900 1.14 ESO NTT (EMMI)
4.383 20.55 ±0.04 I 900 1.02 ESO NTT (EMMI)
4.439 17.61 ±0.04 H 810 ESO NTT (SOFI)
5.059 20.73 ±0.29 I 1800 3.09 Wise 1-m
5.339 21.84 ±0.08 R 3600 1.86 CTIO 0.9-m
6.395 18.86 ±0.14 H 540 ESO NTT (SOFI)
7.609 19.25 ±0.12 H 1200 UKIRT
7.622 18.36 ±0.13 K 600 UKIRT
8.361 19.27 ±0.22 H 2700 ESO 2.2m
8.375 21.60 ±0.06 I 900 0.92 ESO NTT (EMMI)
8.396 18.14 ±0.35 Ks 2700 ESO 2.2m
8.438 22.64 ±0.08 V 900 1.93 ESO NTT (EMMI)
8.578 19.54 ±0.08 H 1620 UKIRT
8.608 20.28 ±0.10 J 2160 UKIRT
8.633 18.77 ±0.24 K 600 UKIRT
9.509 20.40 ±0.12 J 2160 UKIRT
9.554 19.76 ±0.12 H 2160 UKIRT
9.614 18.94 ±0.09 K 2160 UKIRT
10.353 21.62 ±0.16 gunn i 4800 1.22 ESO Dutch
10.380 20.09 ±0.20 H 3750 ESO 2.2m
10.435 19.24 ±0.16 Ks 3900 ESO 2.2m
10.440 22.87 ±0.34 V 900 1.06 ESO NTT (EMMI)
11.496 19.98 ±0.21 H 2160 UKIRT
11.527 19.47 ±0.27 K 2160 UKIRT
13.414 18.76 ±0.30 Ks 4950 ESO 2.2m
13.438 21.47 ±0.41 gunn i 2400 1.98 ESO Dutch
13.558 20.42 ±0.13 J 2160 UKIRT
14.536 20.00 ±0.15 H 2160 UKIRT
14.545 19.36 ±0.14 K 2160 UKIRT
15.582 20.56 ±0.12 J 3240 UKIRT
17.359 22.68 ±0.12 V 900 1.05 ESO NTT (SUSI2)
17.371 21.91 ±0.12 I 900 0.75 ESO NTT (SUSI2)
23.501 20.04 ±0.12 H 4860 UKIRT
23.578 19.28 ±0.11 K 4860 UKIRT
Table 2: The log of the observations with the columns: UT Date, magnit ude and error, filter, exposure time, seeing and the telescope.
Instruments and CCDs used: N TT EMMI: red arm with TEK 2k ×2k CCD (#36), 0.27′′ /pixel; NTT SUSI2: EEV 4k ×2k CCD (#45 & #46),
0.08′′/pixel; N TT SOFI: Hawaii 1k ×1k HgCdTe array, 0.29′′/pixel; ESO Dutch: CCD Camera with TEK 512 ×512 CCD (#33), 0.47′′ /pixel;
Wise 1-m: TEK 1k ×1k CCD, 0.70′′ /pixel; CTIO 0.9-m: TEK 2k ×2k CCD, 0.38′′/pi xel; UKIRT: IRCAM3 with FPA42 256 ×256 detector,
0.29′′/pixel; 2.2-m (IRAC2b): NICMOS-3 256 ×256 array, 0.507′′/pixel. We note that we do not list an estimate of the seeing in case of infrared
observations, since the real seeing is overestimated due to the process of co-adding the individu al frames.
– 21 –
Table 3: Fit parameters for the model m= -2.5 log(10−0.4m0tα+ 10−0.4mgal )
filter m0αmgal χ2
red
V 21.22+0.48
−0.33 –1.61 23.04+0.08
−0.08 5.5/5
R 21.18+0.09
−0.08 –1.61 22.58+0.06
−0.05 14.7/10
I 20.60+0.04
−0.04 −1.36+0.27
−0.36 21.95+0.25
−0.16 4.5/8
J 18.32+0.33
−0.25 –1.61 20.87+0.07
−0.11 5.6/4
H 17.29+0.06
−0.06 −1.67+0.13
−0.15 20.27+0.19
−0.15 6.5/7
K 16.48+0.18
−0.15 –1.61 19.62+0.12
−0.11 11.3/9
