Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://mavr.sao.ru/hq/balega/PUBL_BAL/PUB_2007/A&A_462_L57.ps Äàòà èçìåíåíèÿ: Tue May 5 16:21:10 2009 Äàòà èíäåêñèðîâàíèÿ: Tue Aug 18 06:46:56 2009 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: annular solar eclipse

A&A 462, L57--L60 (2007)
DOI: 10.1051/0004­6361:20066660
c
# ESO 2007
Astronomy
&
Astrophysics
L##### ## ### E#####
Extensive multiband study of the X­ray rich GRB 050408 #
A likely off­axis event with an intense energy injection ##
A. de Ugarte Postigo 1 , T. A. Fatkhullin 2 , G. Jhannesson 3 , J. Gorosabel 1 , V. V. Sokolov 2 , A. J. Castro­Tirado 1 ,
Yu. Yu. Balega 2 , O. I. Spiridonova 2 , M. Jelnek 1 , S. Guziy 1,4 , D. Prez­Ramrez 5 , J. Hjorth 6 , P. Laursen 6 , D. Bersier 7 ,
S. B. Pandey 1,8 , M. Bremer 9 , A. Monfardini 7 , K. Y. Huang 10 , Y. Urata 11,12 , W. H. Ip 10 , T. Tamagawa 11 , D. Kinoshita 12 ,
T. Mizuno 13 , Y. Arai 13 , H. Yamagishi 13 , T. Soyano 14 , F. Usui 15 , M. Tashiro 16 , K. Abe 16 , K. Onda 16 , Z. Aslan 17,18 ,
I. Khamitov 17 , T. Ozisik 17 , U. Kiziloglu 19 , I. Bikmaev 20,21 , N. Sakhibullin 20,21 , R. Burenin 22 , M. Pavlinsky 22 ,
R. Sunyaev 22 , D. Bhattacharya 23 , A. P. Kamble 23 , C. H. Ishwara Chandra 24 , and S. A. Trushkin 2
1 Instituto de Astrofsica de Andaluca (IAA­CSIC), Apartado de Correos 3004, 18080 Granada, Spain
e­mail: deugarte@iaa.es
2 Special Astrophysical Observatory, Nizhnij Arkhyz, Zelenchokskaya, Karachaevo­Cherkesia, 369167 Russia
3 Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavk, Iceland
4 Nikolaev State University, Nikolska 24, 54030, Nikolaev, Ukraine
5 Dpto. de Fsica (EPS), Universidad de Jan, 23071 Jan, Spain
6 Dark Cosmology Centre, Niels Bohr Institute, University of Copenhagen, Juliane Maries Vej 30, 2100 Copenhagen, Denmark
7 Astrophysics Research Inst., Liverpool John Moores Univ., Twelve Quays House, Egerton Wharf, Birkenhead, CH41 1LD, UK
8 The UCL Mullard Space Science Laboratory, Holmbury St. Mary, Dorking, Surrey, RH5 6NT, UK
9 Institut de Radio Astronomie Millimtrique (IRAM), 300 rue de la Piscine, 38406 Saint­Martin d'Hres, France
10 Institute of Astronomy, National Central University, Chung­Li 32054, Taiwan, PR China
11 Institute for Physics and Chemical Research (RIKEN), Wako, Saitama 351­0198, Japan
12 Tokyo Institute of Technology, Ookayama, Meguro, Tokyo 152­8550, Japan
13 Department of Astronomy and Earth Sciences, Tokyo Gakugei University, Koganei, Tokyo 184, Japan
14 Kiso Observatory, Institute of Astronomy, University of Tokyo, Mitake­mura, Kiso­gun, Nagano 397­0101, Japan
15 Japan Aerospace Exploration Agency, Institute of Space and Astronautical Science, Sagamihara, Kanagawa 229­8510, Japan
16 Saitama University, Sakura­ku, Saitama 338­8570, Japan
17 TBITAK National Observatory, Akdeniz niversitesi, 07058, Antalya, Turkey
18 Akdeniz University, Physics Department, 07058 Antalya, Turkey
19 Middle East Technical University, Physics Department, Inonu Bulvari, Ankara, 06531, Turkey
20 Departments of Astronomy, Kazan State University, Kremlevskaya Str., 18, Kazan, 420008, Russia
21 Academy of Sciences of Tatarstan, Bauman Str., 20, Kazan, 420111, Russia
22 Space Research Institute (IKI), 84/32 Profsoyuznaya, Moscow, 117997, Russia
23 Raman Research Institute, Bangalore 560 080, India
24 National Centre for Radio Astrophysics, Ganeshkhind, Pune, 411007, India
Received 30 October 2006 / Accepted 14 December 2006
ABSTRACT
Aims. Understand the shape and implications of the multiband ligth curve of GRB 050408, an X­ray rich (XRR) burst.
Methods. We present a multiband optical light curve, covering the time from the onset of the #­ray event to several months after,
when we only detect the host galaxy. Together with X­ray, millimetre and radio observations we compile what, to our knowledge, is
the most complete multiband coverage of an XRR burst afterglow to date.
Results. The optical and X­ray light curve is characterised by an early flattening and an intense bump peaking around 6 days after
the burst onset. We explain the former by an o#­axis viewed jet, in agreement with the predictions made for XRR by some models,
and the latter with an energy injection equivalent in intensity to the initial shock. The analysis of the spectral flux distribution reveals
an extinction compatible with a low chemical enrichment surrounding the burst. Together with the detection of an underlying starburst
host galaxy we can strengthen the link between XRR and classical long­duration bursts.
Key words. gamma rays: bursts -- techniques: photometric
# Based on observations collected at SAO, La Silla, Roque de
los Muchachos, Haleakala, Kitt Peak, Cerro Tololo, TBITAK,
Kiso, Observatorio de Sierra Nevada, Plateau du Bure, GMRT and
RATAN­600.
## Appendices A and B are only available in electronic form at
http://www.aanda.org
1. Introduction
X­ray flashes (XRFs) where first identified by Beppo­SAX (Heise
et al. 2001) as those bursts detected by the X­ray camera but
not the #­ray monitor. Later studies based on the larger sam­
ple gathered by HETE­2 (Sakamoto et al. 2005b) gave a more

L58 A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408
general (and instrument­ independent) classification and con­
firmed the intermediate group of events, the X­ray rich (XRR)
class, previously detected by Ginga (Yoshida & Murakami
1994) and Granat/WATCH (Castro­Tirado et al. 1994). It is
now known that long­duration GRBs (LGRBs), XRRs and
XRFs share the same isotropic distribution in the sky, the same
duration range and similar spectrum, with the main di#erence of
having respectively lower observed spectral peak energy E obs
peak in
the #F # spectrum. They seem to form a continuum and thereby,
most of the proposed models have tried to explain them as
a unified phenomena (see a summary of the di#erent models in
Granot et al. 2005).
GRB 050408 was detected by WXM, SXC, and FREGATE
aboard HETE­2 (Sakamoto et al. 2005a) at 2005 April 08
16:22:50.93 UT (t 0 hereafter). With an observed peak energy of
#20 keV it was classified as an XRR event. The 1.0 m Zeiss and
6.0 m BTA telescopes at the Special Astrophysical Observatory
(SAO) in Russia pointed at the position delivered by HETE­2
through the GCN (GRB Circular Network) and detected the opti­
cal afterglow (de Ugarte Postigo et al. 2005a) coincident with the
possition of the X­ray afterglow detected by Swift/XRT, which
began observing 42 min after the burst (Wells et al. 2005). The
precise localisation allowed further optical (Covino et al. 2007)
and spectroscopic (Berger et al. 2005; Prochaska et al. 2005) ob­
servations, this latter ones detemined a redshift of z = 1.236.
In Sect. 2 we present the observations and the reduction
methods that have been used for the analysis of the data.
Section 3 describes the results that have been obtained, includ­
ing observations of the host galaxy and modelling of the light
curve. Section 4 discusses the implications of the analysis of the
light curve.
2. Observations and data reduction
For this work we have compiled over 60 photometric measure­
ments in U, B, V, Rc and Ic bands from 12 telescopes. The im­
ages where reduced using standard techniques based on IRAF 1
and JIBARO (de Ugarte Postigo et al. 2005b).
The burst happened during night time in Japan, where a fast
follow up was carried out. The very wide field camera, WIDGET
was monitoring the field of view of HETE­2 when the event
was reported but detected no optical emission before, during
or after the gamma­ray emission down to an unfiltered limit­
ing magnitude of 9.7 (all limits given throughout the paper are
3­#). The 1.05 m KISO Schmidt telescope pointed to the er­
ror box 20 min after the burst but failed to detect the afterglow.
Finally, the 1 m LOT telescope observed the field 55 min af­
ter the burst, images that later served to confirm the afterglow
(Huang et al. 2005b). The discovery of the afterglow was made
with the data of the 1.0 m Zeiss and 6.0 m BTA telescopes in
Russia (de Ugarte Postigo et al. 2005a), starting 115 min after
the burst, when observations became possible from that site.
Further observations were performed from the 1.5 m
Russian­Turkish telescope, in TBITAK National Observatory,
the 4.0 m Blanco telescope in Cerro Tololo, the 4.0 m Mayall
telescope in Kitt Peak, the 2.0 m Faulkes Telescope North (FTN)
in Haleakala and the 3.5 m Telescopio Nazionale Galileo (TNG)
in la Palma. A specially intense multiband campaign was carried
out from the 1.54 m Danish telescope in La Silla, where daily
1 IRAF is distributed by the National Optical Astronomy
Observatories, which is operated by the Association of Universities for
Research in Astronomy, Inc. (AURA) under cooperative agreement
with the National Science Foundation.
observations were obtained during the first 8 days following the
burst onset.
Finally, 8 months after the burst, deep observations were
made from the 3.5 m telescope at Calar Alto. In these images
we detect the host galaxy of the burst in B and Rc bands and
impose a limit in Ic band.
Optical photometric calibration is based on the observa­
tion of several standard fields (Landolt 1992) using the 1.54 m
Danish telescope at La Silla and the 1.5 m telescope at Sierra
Nevada Observatory. From these observations we derive 12 sec­
ondary standards of di#erent brightnesses. A log with the obser­
vations and the calibration stars are given as online material.
Our dataset is completed with several millimetre and radio
limits. 6 epochs of millimetre observations were carried out with
the 6­antenna Plateau de Bure interferometer (PdB, Guilloteau
et al. 1992). No detection was obtained in either of the 1 mm
or 3 mm bands, although a 3­# signal was found on the phase
center in both observing bands on April 18. Careful re­analysis
of the data did not reveal these signals as instrumental artifacts.
Based on the extreme spectral slope and the non­detection on
April 19, we conclude that this result is either due to a statis­
tical fluctuation or an unusual event of interstellar scattering at
high galactic latitude, and not due to a source­intrinsic variation.
Data calibration was done using the GILDAS software package 2
using MWC349 as primary flux calibrator and 3C 273 as ampli­
tude and phase calibrator. Further observations where obtained
13 days after the burst at 1.28 GHz from GMRT and at 8.4 GHz
from RATAN­600.
3. Results
3.1. Light curve
In order to put together all the radio, millimetre, near infrared
(nIR), optical and X­ray data available (including data from
Foley et al. 2006 and Soderberg 2005a,b), we have determined
the corresponding flux density values for all observations. X­ray
afterglow counts, obtained from Foley et al. (2006) have been
converted and corrected for hydrogen column extinction us­
ing WEBPIMMS 3 taking as spectral model a powerlaw with
a slope of #X = 1.16 and a column density of NH = 0.25 â
10 22 cm -1 (Nousek et al. 2006; Chincarini et al. 2005). The
optical data have been corrected for galactic reddening (using
an E(B - V) = 0.026, Schlegel et al. 1998) and intrinsic extinc­
tion (see Sect. 3.2). The measured/estimated host galaxy flux has
been subtracted from the data to obtain the clean afterglow flux
(see Sect. 3.3). The conversion of the optical data to flux den­
sity was done using the transformations given by Fukugita et al.
(1995) for the optical and by Allen et al. (2000) for the nIR.
The resulting light curves are shown in Fig. 1. Note the intense
bump, rising at #3 days and peaking at #6 days, both in op­
tical and X­rays. These kind of fluctuations have already been
detected in the light curves of LGRBs and short­duration bursts
(SGRBs) (de Ugarte Postigo et al. 2005c, 2006).
3.2. Study of the optical­nIR SFD
We have constructed the UBVRcIcJHK­band Spectral Flux dis­
tribution (SFD) of the afterglow 0.6 days after the burst, when
near simultaneous optical and nIR observations were available.
2 GILDAS is the software package distributed by the IRAM Grenoble
GILDAS group.
3 http://heasarc.nasa.gov/Tools/w3pimms.html

A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408 L59
Fig. 1. Multiwavelength light curve of the GRB 050408 afterglow in
the observer frame. The lines show the best fit of a fireball model with
one energy injection (at 2.9 days) seen o#­axis (see text for details). Our
observations are plotted with filled symbols, while the ones obtained
from the literature are represented by empty ones, this convention is
used for all the figures.
Fig. 2. Spectral flux distribution of the afterglow 0.6 days after the burst
onset in the observer frame. The di#erent lines represent results from
fitting the data to various extinction laws: Small Magellanic Cloud
(SMC), Large Magellanic Cloud (LMC), Milky Way (MW) and No
Extinction (NE).
The UBVRcIc­band magnitudes from this work were comple­
mented with the JHK­band values reported by Foley et al.
(2006). Synchronisation to a unique timing is done by assum­
ing a powerlaw with an index of # = 0.7 (F # t -# ), as derived
from a linear fit of the nearby multiband data of the afterglow.
The fluxes are used for fitting an extincted powerlaw (F # #
10 -0.4A # # -# ) with 3 di#erent extinction laws: Milky Way (MW),
Large Magellanic Cloud (LMC) and Small Magellanic Cloud
(SMC) as described by Pei (1992). This allows us to obtain
AV and # simultaneously. The results of these 3 fits are com­
plemented with an unextincted powerlaw case (NE), see Fig. 2
and Table 1. The best fit to the SFD of the afterglow is obtained
when considering a SMC extinction law (# 2 /d.o.f. = 5.0/5).
This is consistent with what has been previously found for other
LGRB afterglows (Kann et al. 2006).
3.3. The host galaxy
Several months after the gamma­ray event we revisited the
GRB field with the 3.5 m telescope at Calar Alto Observatory
in order to search for the host galaxy. Images were obtained
in B, Rc and Ic bands, yielding a faint detection in B and Rc
Table 1. Results of the SFD fitting at 0.6 days for di#erent extinction
laws.
Extinction law # A V # 2 /d.o.f.
MW 1.85 ± 0.30 --0.18 ± 0.22 24.5/5
LMC --0.12 ± 0.48 1.19 ± 0.32 11.5/5
SMC 0.28 ± 0.33 0.73 ± 0.18 5.0/5
NE 1.62 ± 0.07 0 21.0/5
and imposing a limit on Ic. We derive galaxy colour indices of
(B -Rc) = 0.7 ± 0.5 and (Rc - Ic) # 0.73. These values are cor­
rected for Galactic extinction. We have compared these values to
the ones derived from the templates computed by Kinney et al.
(1996) for a wide variety of galaxy types. We may conclude that
only starburst galaxy templates are consistent with them. The
best correlation is obtained with the starbust 2 template, with
an intrinsic extinction of E(B - V) = 0.16.
3.4. Modelling of the multiband data
Using the model and methods described by Jhannesson et al.
(2006) we fitted the multiband observations of the afterglow
(galaxy subtracted) to a fireball model with energy injections,
viewed both on­axis and o#­axis (with varying viewing angles).
At least one injection is needed in order to account for the bump
seen at 6 days which would carry as much energy as the ini­
tial shock. Another characteristic of the light curve is a flat­
tening of the early light curve, seen in Rc and X­rays during
the first hours of the burst, which has already been reported by
Foley et al. (2006). This can be explained either by an early en­
ergy injection (single or continous), or an outflow with a low
initial Lorentz factor, or as the result of an o#­axis viewed burst.
Similar early behaviour has already been found in other bursts
(Nousek et al. 2006; Zhang et al. 2006).
Our preferred scenario (giving the best fit) describes the
burst as a collimated (# 0 = 2.7 # ) fireball seen o# axis (# v =
1.45# 0 ) expanding into a uniform low density environment (n 0 =
0.01 cm -3 ) with an electron index p = 2.03 and having an ad­
ditional energy injection after 2.9 days with 1.2 times the ini­
tial energy. No further injections are needed to explain the light
curve with the available amount of data. From the fit we obtain
a # 2 /d.o.f. = 158.1/93. The jet break, as defined by Sari et al.
(1999), would be expected initially at 1.6 days, or at 3.8 days
due to the energy injection. However, due to the e#ect of the
equal arrival time surface, it is further delayed to approximatelly
30 days. Figure 3 shows the radio to X­ray SFD predicted by
our model for 3 epochs, together with observational data, host
subtraced and corrected for galactic and intrinsic extinction.
4. Discussion
The optical­nIR SFD shows a clear curvature, implying a need
for extinction along the afterglow line of sight in the host galaxy.
The only reliable fit (#/d.o.f. = 5.0/5) is based on a SMC ex­
tinction law. This result points towards a low stage of chemical
enrichment in the region of the progenitor, as is usually found
for LGRBs (Kann et al. 2006). The detection of a starburst host
galaxy is also a common feature with most LGRBs (Fruchter
et al. 2006), facts that once again favour the hypothesis of shared
nature between XRR bursts and LGRBs.
The optical spectral index obtained from this fit (# o = 0.28 ±
0.33) and the X­ray one (# X = 1.14 ± 0.19, Nousek et al. 2006)
are consistent with a synchrotron spectra in which the cooling

L60 A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408
Fig. 3. Spectral Flux Distribution of the afterglow from radio to X­rays
0.6, 6.0 and 13.0 days after the burst in the observer frame. Several 3­#
upper limits from radio and millimetre observations are plotted.
break frequency (# c ) is located between optical and X­rays.
A linear fit of the optical (Rc) and X­ray data between 0.1 and
1.0 days (where there is more data available and the light curve
seems more stable) returns temporal slope values of # o = 0.69 ±
0.04 and #X = 0.99 ± 0.21. These numbers, together with the
optical and X­ray spectral slopes, are consistent with a standard
fireball model (Sari et al. 1999) in which a relativistic outflow is
expanding in a uniform density environment in the slow cooling
regime with an electron power law distribution index of p # 2.0.
A more complex multirange model, confirms these results
and is used to account for the bump that has been detected to
peak at #6 days by allowing for refreshed shocks. This fluctu­
ation is simultaneously observed in optical and X­rays and can
be explained by an energy injection of the order of the initial
shock. This achromaticity and the simultaneity at both sides of # c
rules out other explanations such as a density fluctuation, a dust
echo or a supernova bump (which could also be ruled out by
amplitude and onset time). Other explanations involving an re­
freshed energy release such as a double jet (Berger et al. 2003)
or a patchy shell (Meszaros et al. 1998) can not be discarded.
This injection delays the break, that would be expected for about
1.6 days, to about 30 days and, mainly due to the e#ect of the
equal arrival time surface, transforms it to a very smooth break
that expands over a decade in time.
To explain the flattening seen in the earliest points of the
light curve we have studied the case of a collimated fireball seen
o#­axis, as predicted by some unified models (Yamazaki et al.
2002) that simultaneously intend to explain LGRBs, XRR bursts
and XRFs by only varying the viewing angle. Our fit accounts
reasonably well for the multiband and long scale behaviour of
the light curve. However, the fits obtained with an on­axis model
with an additional early injection or a low initial gamma fac­
tor (dirty fireball) can also interpret the data (although returning
worse fits) and can not be ruled out.
Regarding the energetics of the afterglow we find that, with
an observed peak energy of #20 keV and a fluence of #3.3 â
10 -6 erg cm -2 (2--400 keV) it has an isotropic equivalent energy
release in #­rays E #,iso >
# 1.3 â 10 52 erg, at least 6 times greater
than the predicted by E peak - E iso relation (Amati 2006).
We encourage polarimetric observations of XRR bursts and
XRF events (i.e. Gorosabel et al. 2006) as they will be extremely
useful to better understand the physics and geometry of the emis­
sion and to discriminate between energy models when explain­
ing the fluctuations seen in the light curves.
Acknowledgements. We acknowledge the generous allocation of observing
time by di#erent Time Allocation Committees. This work was partially
supported by the Spanish MCyT under programmes AYA2004­01515 and
ESP2005­07714­C03­03 (including FEDER funds), RFBR grants 04­02­16300
and 05­02­17744 and grant NSh­784.2006.2. The Dark Cosmology Centre is
supported by the Danish National Research Foundation. A.d.U.P. acknowledges
support from FPU grant AP2002­0446 from the Spanish MCyT. G.J. acknowl­
edges support from the Icelandic Research Council.
References
Allen, C. W. 2000, Astrophysical Quantities, 4th Ed. (London: Athlone)
Amati, L. 2006, MNRAS, 372, 233
Berger, E., Kulkarni, S. R., Pooley, G., et al. 2003, Nature, 426, 154
Berger, E., Gladders, M., & Oemler, G. 2005, GRB Coordinates Network, 3201
Castro­Tirado, A. J., Brandt, S., Lund, N., et al. 1994, Gamma­Ray Bursts, AIP
Conf. Proc., 307, 17
Capalbi, M., Malesani, D., Perri, M., et al. 2006, A&A, in press
[arXiv:astro­ph/0610845]
Covino, et al. 2007, submitted
Chincarini, G., Moretti, A., Romano, P., et al. 2005
[arXiv:astro­ph/0506453]
Foley, R. J., Perley, D. A., Pooley, D., et al. 2006, ApJ, 645, 450
Fruchter, A. S., Levan, A. J., Strolger, L., et al. 2006, Nature, 441, 463
Fukugita, M., Shimasaku, K., & Ichikawa, T. 1995, PASP, 107, 945
Gorosabel, J., Larionov, V., Castro­Tirado, A. J., et al. 2006, A&A, 459, L33
Guilloteau, S., Delannoy, J., Downes, D., et al. 1992, A&A, 262, 624
Granot, J., Ramirez­Ruiz, E., & Perna, R. 2005, ApJ, 630, 1003
Huang, K. Y., Ip, W. H., Kinoshita, D., et al. 2005b, GRB Coordinates Network,
3196
Heise, J., in't Zand, J., Kippen, R. M., &Woods, P. M. 2001, Gamma­ray Bursts
in the Afterglow Era, 16
Jhannesson, G., Bjrnsson, G., & Gudmundsson, E. H. 2006, ApJ, 647, 1238
Kann, D. A., Klose, S., & Zeh, A. 2006, ApJ, 641, 993
Kinney, A. L., Calzetti, D., Bohlin, R. C., et al. 1996, ApJ, 467, 38
Landolt, A. U. 1992, AJ, 104, 340
Meszaros, P., Rees, M. J., & Wijers, R. A. M. J. 1998, ApJ, 499, 301
Nousek, J. A., Kouveliotou, C., Grupe, D., et al. 2006, ApJ, 642, 389
Pei, Y. C. 1992, ApJ, 395, 130
Prochaska, J. X., Bloom, J. S., Chen, H.­W., Foley, R. J., & Roth, K. 2005, GRB
Coordinates Network, 3204
Reichart, D. E. 2001, ApJ, 554, 643
Sakamoto, T., Ricker, G., Atteia, J.­L., et al. 2005a, GRB Coordinates Network,
3189
Sakamoto, T., Lamb, D. Q., Kawai, N., et al. 2005b, ApJ, 629, 311
Sari, R., Piran, T., & Halpern, J. P. 1999, ApJ, 519, L17
Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525
Soderberg, A. M. 2005a, GRB Coordinates Network, 3210
Soderberg, A. M. 2005b, GRB Coordinates Network, 3234
Stanek, K. Z., Garnavich, P. M., Nutzman, P. A., et al. 2005, ApJ, 626, L5
de Ugarte Postigo, A., Komorova, V., Fathkullin, T., et al. 2005a, GRB
Coordinates Network, 3192
de Ugarte Postigo, A., et al. 2005b, JIBARO: Un conjunto de utilidades para la
reduccin y anlisis automatizado de imgenes, in Astrofsica Robtica en
Espaa, ed. A. J. Castro­Tirado, B. A. de la Morena, & J. Torres, Madrid,
35
de Ugarte Postigo, A., Castro­Tirado, A. J., Gorosabel, J., et al. 2005c, A&A,
443, 841
de Ugarte Postigo, A., Castro­Tirado, A. J., Guziy, S., et al. 2006, ApJ, 648, L83
Wells, A. A., Abbey, A. F., Osborne, J. P., et al. 2005, GRB Coordinates Network,
3191
Yamazaki, R., Ioka, K., & Nakamura, T. 2002, ApJ, 571, L31
Yoshida, A., & Murakami, T. 1994, Gamma­Ray Bursts, AIP Conf. Proc., 307,
333
Zhang, B., Fan, Y. Z., Dyks, J., et al. 2006, ApJ, 642, 354

A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408, Online Material p 1
Online Material

A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408, Online Material p 2
Appendix A: Calibration stars
Table A.1. Calibration stars in the field of GRB 050408, as marked in Fig. A.1.
# RA (J2000) Dec (J2000) U B V R I
1 12:02:09.16 +10:49:24.1 19.114 ± 0.053 19.174 ± 0.024 18.601 ± 0.009 18.213 ± 0.008 17.835 ± 0.024
2 12:02:10.38 +10:51:37.3 18.982 ± 0.053 19.241 ± 0.024 18.776 ± 0.009 18.438 ± 0.008 18.062 ± 0.025
3 12:02:08.70 +10:51:52.4 19.743 ± 0.054 18.759 ± 0.023 17.167 ± 0.008 16.125 ± 0.007 14.776 ± 0.022
4 12:02:14.68 +10:54.02.8 18.200 ± 0.051 18.508 ± 0.023 18.028 ± 0.008 17.671 ± 0.007 17.292 ± 0.023
5 12:02:11.76 +10:54:59.6 18.204 ± 0.051 17.855 ± 0.023 17.089 ± 0.008 16.627 ± 0.007 16.218 ± 0.022
6 12:02:22.95 +10:53:57.8 18.896 ± 0.051 19.098 ± 0.023 18.638 ± 0.009 18.300 ± 0.008 17.980 ± 0.024
7 12:02:25.20 +10:51:43.6 20.034 ± 0.058 19.143 ± 0.023 17.681 ± 0.008 16.742 ± 0.007 15.695 ± 0.022
8 12:02:24.70 +10:49:54.7 17.237 ± 0.051 17.406 ± 0.023 17.020 ± 0.008 16.652 ± 0.007 16.324 ± 0.022
9 12:02:17.02 +10:51:45.5 20.510 ± 0.064 20.805 ± 0.030 20.216 ± 0.018 19.673 ± 0.020 18.951 ± 0.041
10 12:02:14.76 +10:52:54.6 -- 21.885 ± 0.058 20.528 ± 0.019 19.617 ± 0.014 18.631 ± 0.028
11 12:02:17.10 +10:51:02.5 -- 23.674 ± 0.051 22.376 ± 0.098 21.823 ± 0.092 20.947 ± 0.137
12 12:02:11.32 +10:51:11.3 21.129 ± 0.089 21.249 ± 0.034 20.810 ± 0.023 20.507 ± 0.029 20.025 ± 0.066

A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408, Online Material p 3
Fig. A.1. Identification chart of GRB 050408. The calibration stars and
the afterglow location have been marked. The field of view is 6.0 # â
6.5 # , North is to the top and East to the left.
Optical photometric calibration is based on the observa­
tion of several standard fields (Landolt 1992) using the 1.54 m
Danish at La Silla (Chile) and the 1.5 m telescope at Sierra
Nevada Observatory (Spain). From these observations we derive
12 secondary standards of di#erent brightnesses in the field of
the GRB. Identification of these stars is shown in Fig. A.1, while
their photometric values are displayed in Table A.1.
Appendix B: Observations
In the following tables we use the HETE­2 onset time t 0 =
2005 April 08 16:22:50.93 UT.
Table B.1. Observations in millimetre wavelengths from the Plateau du
Bure interferometer (compact 6­antenna configuration on all dates). The
errors are based on point­source fits in the UV plane to the phase center.
(t - t 0 ) (days) Band (GHz) Flux (mJ) 3­# limit (mJy)
3.23 86.789 0.0 ± 0.3 0.9
3.23 229.068 1.0 ± 1.6 4.8
5.19 115.477 0.5 ± 0.8 2.5
5.19 232.295 3.3 ± 1.8 5.4
10.34 86.251 0.9 ± 0.3 0.9 a
10.34 232.171 8.4 ± 2.3 6.9 a
11.19 108.995 --1.7 ± 1.7 5.2
11.19 228.534 --9.9 ± 6.5 19.5
12.40 108.995 --0.9 ± 0.7 2.2
12.40 228.534 3.6 ± 2.7 8.2
14.29 111.619 --0.8 ± 0.4 1.3
14.29 224.680 1.4 ± 2.1 6.3
a The faint detections found on this epoch are considered to be due to
a statistical fluctuation or to an unusual event of interstellar scattering
at high galactic latitud, taking into account the non detection the next
night and the extreme spectral slope.
Table B.2. Observations in radio wavelengths from the Giant
Metrewave Radio Telescope (GMRT) and Radio Astronomical
Telescope Academy Nauk (RATAN­600).
(t - t 0 ) (days) Telescope Band (GHz) 3­# limit (mJy)
13.0 GMRT 1.28 0.45
13.0 RATAN­600 8.4 5.0

A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408, Online Material p 4
Table B.3. Optical observations carried out for GRB 050408. The magnitudes are in the Vega system and are not corrected for Galactic reddening.
(t - t 0 ) (days) Tel. + Inst. Filter Texp (s) Mag ErMag
0.4780 Dk1.54 m+DFOSC U 9000 23.09 0.18
1.5939 4.0 mKPNO U 1600 23.75 0.26
0.1297 6.0 mBTA+SCORPIO B 600 22.37 0.05
0.1385 6.0 mBTA+SCORPIO B 500 22.44 0.05
0.1467 6.0 mBTA+SCORPIO B 600 22.46 0.05
0.1554 6.0 mBTA+SCORPIO B 600 22.55 0.05
0.2301 6.0 mBTA+SCORPIO B 500 22.79 0.07
0.3712 Dk1.54 m+DFOSC B 1200 23.05 0.10
0.5681 Dk1.54 m+DFOSC B 1200 23.54 0.17
1.5001 Dk1.54 m+DFOSC B 9000 24.26 0.10
242.5213 3.5 mCAHA B 1800 25.32 4 0.22
0.2245 6.0 mBTA+SCORPIO V 300 22.11 0.06
0.3599 Dk1.54 m+DFOSC V 600 22.24 0.09
1.6206 4.0 mKPNO V 1200 23.98 0.22
2.4394 4.0 mCTIO V 1500 24.10 0.17
0.0155 Kiso 1.05 Schmidt Rc 300 >19.50 --
0.0388 1.0 mLOT Rc 5 180 20.34 0.10
0.0859 1.0 mZeiss Rc 900 20.99 0.09
0.0919 1.0 mLOT Rc 5 180 20.80 0.18
0.0956 1.0 mLOT Rc 5 300 20.98 0.14
0.1029 6.0 mBTA+SCORPIO Rc 180 20.92 0.03
0.1068 6.0 mBTA+SCORPIO Rc 180 20.94 0.03
0.1104 6.0 mBTA+SCORPIO Rc 180 20.92 0.03
0.1139 6.0 mBTA+SCORPIO Rc 180 20.92 0.02
0.1172 6.0 mBTA+SCORPIO Rc 180 20.95 0.02
0.1207 6.0 mBTA+SCORPIO Rc 180 21.03 0.03
0.1240 6.0 mBTA+SCORPIO Rc 180 21.05 0.03
0.1471 RTT150+TFOSC Rc 450 21.13 0.08
0.1629 RTT150+TFOSC Rc 450 21.05 0.06
0.1771 RTT150+TFOSC Rc 540 21.10 0.06
0.1863 RTT150+TFOSC Rc 540 21.19 0.06
0.1984 RTT150+TFOSC Rc 480 21.42 0.08

A. de Ugarte Postigo et al.: Extensive multiband study of the X­ray rich GRB 050408, Online Material p 5
Table B.3. continued.
(t - t 0 ) (days) Tel. + Inst. Filter Texp (s) Mag ErMag
0.2080 RTT150+TFOSC Rc 480 21.40 0.08
0.2175 RTT150+TFOSC Rc 480 21.39 0.09
0.2204 6.0 mBTA+SCORPIO Rc 180 21.32 0.04
0.2296 RTT150+TFOSC Rc 480 21.19 0.06
0.2380 RTT150+TFOSC Rc 480 21.52 0.09
0.2463 RTT150+TFOSC Rc 480 21.56 0.08
0.2742 RTT150+TFOSC Rc 720 21.52 0.06
0.2867 RTT150+TFOSC Rc 720 21.58 0.06
0.3392 RTT150+TFOSC Rc 600 21.80 0.12
0.3496 RTT150+TFOSC Rc 600 21.50 0.09
0.3540 Dk1.54 m+DFOSC Rc 300 21.81 0.13
0.3600 RTT150+TFOSC Rc 600 21.84 0.14
0.3705 RTT150+TFOSC Rc 600 21.95 0.15
0.4362 Dk1.54 m+DFOSC Rc 600 21.91 0.08
0.5366 Dk1.54 m+DFOSC Rc 900 22.00 0.07
0.5963 Dk1.54 m+DFOSC Rc 1200 22.02 0.08
0.7166 2.0 mFTN Rc 2400 22.50 0.11
1.4024 Dk1.54 m+DFOSC Rc 5700 23.08 0.09
1.5864 Dk1.54 m+DFOSC Rc 3600 23.31 0.14
2.4911 Dk1.54 m+DFOSC Rc 14 700 23.86 0.17
3.4506 Dk1.54 m+DFOSC Rc 14 000 23.85 0.13
5.1575 RTT150+ANDOR CCD Rc 9000 23.70 0.20
5.5124 Dk1.54 m+DFOSC Rc 7200 23.72 0.18
7.4663 Dk1.54 m+DFOSC Rc 15 600 23.97 0.11
30.2567 3.5 mTNG Rc 3600 24.64 0.17
242.4885 3.5 mCAHA Rc 2500 24.60 4 0.15
0.3871 Dk1.54 m+DFOSC Ic 1200 21.25 0.07
0.6137 Dk1.54 m+DFOSC Ic 1500 22.52 0.11
6.4766 Dk1.54 m+DFOSC Ic 13 800 23.50 0.18
242.5446 3.5 mCAHA Ic 1500 >24.0 --
--0.00340 WIDGET Unfiltered 95 >9.7 --
--0.00224 WIDGET Unfiltered 95 >9.7 --
--0.00108 WIDGET Unfiltered 95 >9.7 --
0.00007 WIDGET Unfiltered 95 >9.8 --
0.00123 WIDGET Unfiltered 95 >9.7 --
4 Host galaxy magnitude.
5 VR broad band filter was transformed to Rc.