Arxiv:Astro-Ph/0309563V1 21 Sep 2003 02 T20 Ac 91:71.7U.Teﬂec in ﬂuence the UT

arXiv:astro-ph/0309563v1 21 Sep 2003 02 t20 ac 91:71.7U.Teflec in fluence The UT. al. 1 et 11:37:14.67 was (Ricker, 29 range satellite keV March HETE-2 3–400 (Mon- 2003 the SXC at on and 2002) 2003), 2002) al. al. et et (Shirasaki, nelly, was et WXM (Atteia, 2003), FREGATE 2002), instruments, al. al. three the et by (Vanderspek, detected 030329 GRB brightness, 8 9 10 ua(03.Terdhf fteatrlwwsmeasured was afterglow the of Ue- redshift and as (2003) The Torii by (2003). followed mura (2003), Price & Peterson trigger). at after 1995) analysis minutes al. (73 ground et UT (Barthelmy, by 12:50:24 GCN determined through was distributed and direction burst The 11 ihteHTosrain Fuhe,e l 2003). found was al. mag- et Cloud absolute (Fruchter, an Magellanic observations HST with Small the galaxy the Hjorth, with host to 2003; dwarf similar A al. nitude et 2003). Kawabata, a 2003; al. spectroscopi- et faded, al. was and et afterglow (Stanek, which photometrically the cally both emerged As observed 2003dh) clearly 2003). (SN component al. supernova et (Greiner, VLT 1 2 3 rpittpstuigL using typeset Preprint 2021 17, April version Draft 4 5 6 7 IA rjc,241-0,Tuahm-ih,Kohoku-ku, Tsunashima-nishi, 2-4-10-708, Project, MISAO um srnmclOsraoy 808,Nkym,Takay Nakayama, 6860-86, Observatory, Astronomical Gunma omcRdainLbrtr,RKN -,Hrsw,Wk,S Wako, Hirosawa, 2-1, RIKEN, Laboratory, Radiation Cosmic eateto srnm,Fclyo cec,KooUnivers Kyoto Science, of Faculty Astronomy, of Department eateto hsc,Fclyo cec,Kuh Universi Kyushu Science, of Faculty Physics, of Department oaunUiest,26-,Nkn-ah,Hcij,Tok Hachioji, Nakano-machi, 2665-1, University, Kogakuin asgtk aea hw ae o,224Nb,Yokosuka, Nobi, 2-2-4 Co., Kaken Showa camera, Yatsugatake aaoNtoa olg fTcnlg,76Tkm,Ngn 3 Nagano Tokuma, 716 Technology, of College National Nagano IA rjc,718 ohky,Ae,Siaa3206,J 362-0064, Saitama Ageo, Koshikiya, 791-8, Project, MISAO og( long A dnicto fa pia fego a eotdby reported was afterglow optical an of Identification nti etr erpr h aletpaebhvo of behavior phase earliest the report we Letter, this In aiOsraoy aiAtoPr,17-,Tkym,Saji- Takayama, 1071-1, Park, Astro Saji Observatory, Saji ie srnmclOsraoy 737 kr,Bsi Oda Bisei, Okura, 1723-70 Observatory, Astronomical Bisei z 0 = . pc xldstepeec fa diinlflr opnn at headings: component th Subject flare of additional Extrapolation 021211. an GRB of co contrasts. and presence synchrotron density the the ambient excludes above the epoch was distribution to band smooth insensitive optical the is the by that explained is is explanation lightcurve featureless a Such hog h C)adcniudfrmr hn5hus h after The hours. 5 than more for continued 12 at and started GCN) photometries the series time through Follow-up UT. 13:00 and 10:00 3 in tre 7mntsbfr h rge n h us oiinwsc was U position 11:37:14.67 burst than 29, the brighter March and flare 2003 trigger contemporaneous the at or before precursor satellite minutes HETE-2 97 the started by detected was 4mnatrbrt t aigbten12ad63husi elchara well is hours 6.3 and 1.2 between fading Its form burst. after min 74 65fo ihrslto pcrsoywt the with spectroscopy resolution high from 1685 − epeetteeris pia mgn bevtoso R 030329 GRB of observations imaging optical earliest the present We ≃ Rc %i iue ohustmsae n (∆ and timescales hours to minutes in 5% 0s uaingmarybrto extreme of burst gamma-ray duration s) 50 .Torii K. f bn.N infiatflxvrainwsdtce n pe iisare limits upper and detected was variation flux significant No -band. my (1 = (mJy) A 1. T E tl mltajv 11/12/01 v. emulateapj style X 1 introduction . .Yoshida S. .Kato T. , 2 × . am-as uss—sproa:idvda:(N2003dh) (SN individual: supernovae: — bursts gamma-rays: 99 10 H ALETOTCLOSRAIN FGB030329 GRB OF OBSERVATIONS OPTICAL EARLIEST THE − ± 4 0 rscm ergs 2 . 8 02(statistic) .Yamaoka H. , .Kinugasa K. , − 2 Rce,2003). (Ricker, ± 3 rf eso pi 7 2021 17, April version Draft 0 .Kohmura T. , 9 . .Kohama M. , 14(systematic)) o1201,Japan 192-0015, yo o,Yz-u,Ttoi6911,Japan 689-1312, Tottori Yazu-gun, son, y opnas,Fkoa8086,Japan 810-8560, Fukuoka Ropponmatsu, ty, ooaa aaaa2305,Japan 223-0053, Kanagawa Yokohama, kym 1-41 Japan 714-1411, Okayama , t,Sko-u yt 0-52 Japan 606-8502, Kyoto Sakyou-ku, ity, ABSTRACT apan m,Aasm,Gna3700,Japan 377-0702, Gunma Agatsuma, ama, iaa3109,Japan 351-0198, aitama aaaa2904,Japan 239-0841, Kanagawa 185,Japan 81-8550, f/f V 1 ) RMS 5 = 4 eto atr o h bevdeeg ag n beaming 2003). and al. range et energy observed (Bloom, the for factors rection oa egh35m(/.) h edo iwws85 was view of of field lens The angle wide (f/1.4). 3.5mm a (here- length and camera of focal XC-75 southern field SONY Japan. The the utilized Yatsugatake, of YC-S) camera. out after northern in always the was m GRB of the 1000 view of of position sky altitude The cameras an video at the (North field wide two placed by ultra of (fixed) imaged unguided consists South) continuously system and The was 030329 Camera. Yatsugatake GRB of region aewt h uoae epneTlsoe(R)at (ART) Telescope Response Automated the with made within meteor. to bright YC-N fur- a of We using time data. by read-out the read-out the in Approximate calibrated superimposed was ther frame tape. each and of video SS-10), time timelapse (MSJ memory to 8 frame after written to format stored NTSC integrations, in s output continuously were Images itneo 0 p n h rmtgmaryenergy gamma-ray prompt the and luminosity Mpc a output, use 800 We of 030329. GRB distance of afterglow optical the 1 .Okamoto Y. , .Oribe T. , u bevto o ssmaie ntbe1 h sky The 1. table in summarized is log observation Our olwu bevtoso h us oiinwere position burst the of observations Follow-up . 35 = × ( 1 ( V t/ E day) 1 − 5 = γ < t %i eod omntstimescales. minutes to seconds in 5% 10 7 = ln rqec hr mretflux emergent where frequency oling . niuul bevd efudno found We observed. ontinuously n .Kawabata T. and , trzdb igepwrlwo the of power-law single a by cterized )i 2s(4s iecl between timescale s) (64 s 32 in 5) 5:9U 7 fe oiinnotice position after s (75 UT :51:39 lwwas glow fego ihcret h burst the to lightcurve afterglow e 0mntsa eni R 990123 GRB in seen as minutes 10 − 5 . fcrubrtmdu.Another medium. circumburst of .Ohnishi K. , 0 9 eae oS 03h h burst The 2003dh. SN to related . 890 × .Orwd-edmonitoring wide-field Our T. 10 2. ± 0 51 . 006(statistic) observations eie s(∆ as derived Rc f k 12 = b 6 rs where ergs, .Kadota K. , . 11 35 ± 0 ± . 010(systematic) f/f 0 . 7at 07 k ) 7 RMS , and t f = = b r cor- are ◦ × ± 70 s 4 ◦ . 2 Torii, et al.

RIKEN. The two computer controlled equatorial mounts sidereal motion. We thus created stacked running mean were used for the observation. These mounts carry a 0.20- of four frames, corresponding to 32 s integration. We also m f/4.0 reflector (hereafter T1) and a 0.25-m f/6.8 reflector created stacked frames of 8 × 8 s integration to see fainter (T2), respectively. T1 was equipped with Apogee AP7p objects. camera which gave 53×53 arcmin field of view. T2 was Figure 1 shows a stacked 8 × 8 s frame at the burst trig- equipped with Apogee AP6E camera with 50×50 arcmin ger time. At the position of the afterglow, no significant field of view. The CCDs were continuously read out with object was found. In all the frames with 32 s (64 s) in- a constant integration times of 30-s and 60-s for T1 and tegration, a nearby V = 5.1 mag (V = 5.5 mag) star is T2, respectively. The burst position was received through clearly seen while fainter stars are not always detected. the GCN (GRB Coordinates Network) (Barthelmy, et al. We therefore derive an upper limit of V = 5.1 (V = 5.5) 1995) and the two telescopes automatically slewed to the for 32 s (64 s) timescales between 10:00 and 13:00 UT for notified position under fine meteorological condition. The any optical emission associated with the GRB. These re- first useful frames were acquired starting at 12:51:39 (75 s sults expand our previous report based on the same data after notification) with T1 and 12:52:09 with T2. On Mar. set (Okamoto, Ohnishi, & Torii 2003). 29, the first images obtained with T2 were matched to The other data, obtained with cooled CCD cameras, USNO-A2.0 catalog (Monet, et al. 1998) by using PIXY12. were reduced in standard ways. Figure 2 shows the first The optical afterglow candidate was identified as an only image obtained with the ART T1. After dark subtrac- one uncataloged object in the field of view. This object tion and flat fielding, aperture photometry was performed was not present in a DSS (POSS E) frame and no cor- by a java-based software developed by T. Kato (ART and responding minor planet13 was cataloged. The informa- GETS data), PIXY-214 (Ageo data), and Astroart (Saji tion of this new transient was reported in a GCN circular data). Photometric analysis for the Ageo data was per- (Torii, 2003). An earlier identification of the afterglow at formed by S. Yoshida. The Saji data were calibrated by the Siding Spring Observatory was reported by Peterson using the standard stars of Landolt (1992), and the two & Price (2003). Between 13:41 UT and 13:43 UT, three frames in each of Rc and V were combined to yield the 30-s exposures were obtained with ART T1 by stopping following measurements: Rc = 13.70 ± 0.05 at 16:58:00 the sidereal tracking. The afterglow was detected as lin- UT and V = 13.99 ± 0.05 at 17:02:45 UT. For the ART, ear trails in these frames and used to study short time GETS, and Ageo data, we used a comparison star at (RA, variability with 0.44 s(/pixel) resolution. Dec)=(10 44 54.485, +21 34 29.80) (J2000; USNO-A2.0 In the city of Ageo, K. Kadota received an e-mail alert 1050.06351075) as measured by Henden (2003), which has message and observed the afterglow. The instrument was a similar color (V − Ic =0.839) to the afterglow. 0.25 m Newtonian reflector equipped with unfiltered SBIG For unfiltered observations, instrumental magnitudes ST-9E CCD camera. The robotic GETS 0.25 m telescope were converted to Rc by using color terms for correspond- at the Gunma Astronomical Observatory was also used ing CCDs (Henden 2000). We assumed that the afterglow to observe the afterglow. A part of the GETS data suf- had a constant color, V − Ic = 0.85, throughout the ob- fered from passages of clouds. At the Saji Observatory, K. servations (Kinugasa, et al. 2003). This value is close to Oribe acquired two Rc and two V frames with the 1.03-m a near simultaneous independent measurement (V − Ic = reflector. 0.86 at Mar 29.744 UT; Burenin, et al. 2003a). We find that our corrected Rc magnitudes obtained from different 3. analyses and results unfiltered instruments (ART, GETS, Ageo) agree within 0.07 mag. These measurements agree within 0.07 mag to Analog data of the Yatsugatake Camera were analyzed the Rc-filtered Saji data and we estimate that our ab- after digitization as described below. A single frame of solute flux calibration is accurate at ≃ 7% level. Our 8-s integration is stored as a single frame of NTSC format best-fit power-law function as described below predicts (1/30 s) in a VHS video tape. This means that nom- 0.07 − 0.10 mag brighter values compared to the simulta- inal playback of the tape shows a movie of one minute neous Rc data obtained at the Siding Spring Observatory which corresponds to 4 hours in original time. The tape (Price, et al. 2003). The function also predicts ≃ 0.07 mag was played back and read to a personal computer by us- brighter values compared to the first part of Rc lightcurve ing a capture card (Canopus, Power Capture Pro). The obtained with RTT150 (Burenin, et al. 2003b). data were stored as uncompressed AVI format movies of Figure 3 shows the afterglow lightcurves. Among these, 640×480 resolution. From these movie files, each frame of the data from ART T2 have the longest time coverage 8 s integration was extracted as an 8-bit FITS file. We cre- with highest signal-to-noise ratio; we discuss here on this ated background frames as an average of many adjacent- data set. We estimated a total error of each measure- time frames and subtracted the background from the raw ment by RMS variations of comparison stars in 11–14 frames. This procedure largely reduced the effect of noisy mag range. Total errors (combination of statistical and hot pixels. We found that relatively faint (∼ 5 mag) stars other errors except for a uniform shift of zero point) for were not steadily apparent in a single 8 s integration frame. T2 were estimated as ≃ 0.03 mag at early part (when However, these stars could be steadily seen if we combine the afterglow was Rc ≃ 12.5) and ≃ 0.05 mag at the last four consecutive frames. This is probably explained by the part (when the afterglow was Rc ≃ 14.0) in RMS. Pho- fact that the CCD chip of YC-S is interline type and that tometric data from ART T2 are summarized in Table 2. it takes 32 s for a stellar image to pass a pixel structure by 12 http://www.aerith.net/misao/pixy1/index.html 13 http://scully.harvard.edu/∼cgi/CheckMP 14 http://www.aerith.net/misao/pixy/index.html TheEarliestOpticalObservationsofGRB030329 3

We fit the ART T2 lightcurve with a power-law function, the latter larger break to α3 ≃ −1.9 which lasted over Rc = a + b · log (t/1 day). This function is statistically ac- 1000 hours in X-rays (Tiengo, et al. 2003). The steep- 2 cepted with χ /d.o.f. = 1.0. The best-fit parameters are ening to α3 was achromatic (Burenin, et al. 2003b) and a = 15.21 ± 0.01(statistic) and b = 2.22 ± 0.02(statistic). could probably be interpreted as jet break in constant den- Flux density at Rc band (658 nm) is then derived as sity medium (e.g., Sari, Piran, & Halpern 1999; Kumar & f(Jy) = (1.99±0.02(statistic)±0.14(systematic))×10−3 × Panaitescu 2000). For a jet break, the decay index evolves −0.891±0.006(statistic)±0.010(systematic) (t/1 day) . Statistic er- from α1 = −(3p − 2)/4 to α3 = −p. While this interpreta- rors are 90% confidence values and the systematic error for tion suggests a flat particle distribution, p ≃ 1.9 (with high the decay index was estimated as the difference to the best- energy cut-off), the evolution of decay index is naturally fit value for the ART T1 lightcurve. We therefore conclude reproduced. that deviation from a single power-law decay was small Since this gamma-ray burst was related to the supernova < < ((∆f/f)RMS ∼ 3 % at first part and (∆f/f)RMS ∼ 5 % (Stanek, et al. 2003; Kawabata, et al. 2003; Hjorth, et al. at last part) in minutes to hours time scales during the 2003), significant amount of circumstellar medium must observation. Uemura, et al. (2003b) reported the pres- have been present around the progenitor (e.g., Chevalier ence of wavy structures in the early lightcurve. Since the & Li 1999). Massive stars (such as a Wolf-Rayet star) reported amplitude seems ≃ 0.1 mag in the time region of blow blob-like winds and hydrodynamic instability is also our observations, the presence of such structures does not expected (e.g., Garcia-Segura, Langer, & Mac Low 1996). seem to be highly inconsistent with the results presented The shock interactions in such media modify the lightcurve herein. while the effect depends on the observed frequency range The three frames of ART T1 without sidereal track- (e.g., Lazzati, et al. 2002; Nakar, Piran, & Granot 2003). ing were investigated by constructing afterglow lightcurves At above the cooling frequency, νc, radiated flux is almost with 0.44, 0.88, 1.76, and 3.52 s time bins. Compared to insensitive to the density fluctuation for p ≃ 2. While the a similar magnitude star, no significant short-term varia- radiated flux sensitively traces the density variation be- tion was found. The variation of these data (upper limit tween νc and νm. In several afterglows, bumpy structures of the afterglow fluctuation) is derived as, (∆f/f)RMS = (e.g., GRB 021004: Fox, et al. 2003) or irregular fluctua- 35 (∆t/1 s)−1/2 [%]. Combining the results from ART T2 tions (GRB 030418: Smith, Rykoff, & McKay 2003) were lightcurve, this relation is valid between seconds to min- observed at early phase. These features may be attributed utes time scales. to density fluctuations in the circumburst medium and the cooling frequency above the optical band (Lazzati, et al. 4. discussion 2002). In the early afterglow of GRB 030329, the absence of Afterglow lightcurves of gamma-ray bursts are gener- significant fluctuation may be attributed to the smooth ally described within a framework of synchrotron radi- ambient medium. For small density contrasts, the overall ation from external forward shock (e.g., Wijers, Rees, hydrodynamical structure is not modified and the radiated & Mészáros 1997) as applied for the first afterglow of flux scales as ∆f/f ∝ (∆n/n)1/2 (Lazzati, et al. 2002). GRB 970228 (van Paradijs, et al. 1997; Costa, et From our measurements of (∆f/f)RMS = 3 − 5%, upper al. 1997). For an adiabatic evolution into uniform limits for density contrasts are estimated as (∆n/n)RMS = medium, the shock radius is calculated as r = 2.1 × 6 − 10%. Another explanation is that a sign of inhomo- 17 1/4 1/4 −1/4 10 (t/1day) (E/Eγ ) (n/10) cm (Sari, Piran, & geneities did not emerge in the lightcurve because the op- Narayan 1998). For our lightcurve between 1.2 and 6.3 tical band was above νc during the observation. Burenin, hours after burst, the radius corresponds to r ≃ 0.03 − et al. (2003b) observed the afterglow between 0.25 and 0.6 −1/4 0.05(n/10) pc. days and derived upper limits of 10–1% for flux fluctuation A single power-law decay in the optical data (α1 = on 0.1–1000 s time scales. Therefore, the current discus- −0.890) and a similar behavior in X-rays (−0.9 ± 0.3; sions apply for a broader time range between 1.2 hours Tiengo, et al. 2003) suggest that the optical and X-ray and 0.6 days after burst. bands were in the same segment of the multi-wavelength If we simply extrapolate the afterglow lightcurve toward spectrum divided by the typical synchrotron frequency, the burst epoch, expected magnitudes are Rc = 6.5 and νm, and the synchrotron cooling frequency, νc (Sari, et al. Rc = 8.7 at t = 10 s and t = 100 s after trigger, respec- 1998). Just after the end of our observations, the decay tively. Our contemporaneous upper limits of V =5.1−5.5, − ± index suddenly steepened to α2 = 1.22 0.03 (Lipunov, as well as similar limits from contemporaneous photog- − ± et al. 2003) or α2 = 1.19 0.01 (Burenin, et al. 2003b). raphy (Sasaki, et al. 2003), are above these estimates. At t = 0.57 days, the decay index further steepened to In some gamma-ray bursts (GRB 990123: Akerlof, et al. ≃ − α3 1.9 (Burenin, et al. 2003b; Garnavich, Stanek, & 1999; GRB 021211: Wozniak, et al. 2002; Park, Williams, Berlind 2003). & Barthelmy 2002; Li, et al. 2003a), bright and rapidly ≃ ≃ − The value of ∆α12 0.3 at t 6 hours is close to 1/4 as fading (f ∝ t 2) optical flare is detected. This component expected from the passage of a cooling frequency through dominates the ordinary afterglow within 10 minutes after the observing band. In this case, the flux evolves from burst and considered to originate from reverse shock (e.g., −3(p−1)/4 −(3p−2)/4 f ∝ t to f ∝ t (Sari, et al. 1998) and Kobayashi 2000). Our contemporaneous upper limit ex- p =2.19 is derived for electron number distribution. This cludes the presence of such a flare component, as well as interpretation is interesting, since the index p is close to bright precursor emission. the universal value, p ≃ 2.2 − 2.3, expected from parti- In summary, the early optical afterglow of GRB 030329 cle acceleration in ultrarelativistic shock (e.g., Abraham, is well characterized by a single power-law decay with in- et al. 2001). However, there is difficulty in explaining 4 Torii, et al. dex α1 = −0.890. The absence of significant flux fluc- This work is partly supported by a grant-in-aid tuation in our lightcurve suggests that the circumburst 15740129 (Torii), 13640239 and 15037205 (Kato), and medium was smoothly distributed or the optical band was 14740131 (Yamaoka) from the Japanese Ministry of Ed- above the synchrotron cooling frequency during the obser- ucation, Culture, Sports, Science and Technology. vation.

REFERENCES

Abraham, A., Gallant, Y. A., Kirk, J. G., & Guthmann, A. W. 2001, Monet, D., et al. 1998, USNO-A 2.0, A Catalog of Astrometric MNRAS, 328, 393 Standards (Flagstaff: U.S. Naval Observatory Flagstaff Station) Akerlof, C., et al. 1999, Nature, 398, 400 Monnelly, G., et al. 2002, in Gamma-Ray Burst and Afterglow Atteia, J.-L., et al. 2002, in Gamma-Ray Burst and Afterglow Astronomy 2001: A Workshop Celebrating the First Year of the Astronomy 2001: A Workshop Celebrating the First Year of the HETE Mission, ed. G. R. Ricker & R. K. Vanderspek (New York: HETE Mission, ed. G. R. Ricker & R. K. Vanderspek (New York: AIP), p.25 AIP), p.17 Nakar, E., Piran, T., Granot, J. 2003, New Astronomy, 8, 495 Barthelmy, S. D., et al. 1995, Ap&SS, 231, 235 Okamoto, Y., Ohnishi, K., & Torii, K. 2003, GCN Circ. 2225 Bloom, J. S., et al. 2003, ApJ, in press (astro-ph/0302210) Park, H. S., Williams, G., & Barthelmy, S. 2002, GCN Circ. 1736 Burenin, R., et al. 2003a, GCN Circ. 2024 Peterson, B. A., & Price, P. A. 2003, GCN Circ. 1985 Burenin, R., et al. 2003b, Astronomy Letters, 2003, 9, 1 Price, P. A., et al. 2003, Nature, 2003, 423, 844 Chevalier, R. A. & Li, Z-Y. 1999, ApJ, 520, L29 Ricker, G. R., et al. 2002, in Gamma-Ray Burst and Afterglow Costa, E., et al. 1997, Nature, 387, 783 Astronomy 2001: A Workshop Celebrating the First Year of the Fox, D. W., et al. 2003, Nature, 422, 284 HETE Mission, ed. G. R. Ricker & R. K. Vanderspek (New York: Fruchter, A., et al. 2003, GCN Circ. 2243 AIP), p.3 Garnavich, P., Stanek, K. Z., &Berlind, P. 2003, GCN Circ. 2018 Ricker, G. R. 2003, IAUC, 8101 Garc´ıa-Segura, G., Langer, N., & Mac Low, M-M. 1996, A&A, 316, Sari, R., Piran, T., & Narayan, R. 1998, ApJ, 497, L17 133 Sari, R., Piran, T., & Halpern, J. P. 1999, ApJ 519, L17 Greiner, J., et al. 2003, GCN Circ. 2020 Sasaki, K., et al. 2003, GCN Circ. 2217 Henden, A. 2000, JAAVSO, 29, 35 Shirasaki, Y., et al. 2003, PASJ, in press Henden, A. 2003, GCN Circ. 2082 Smith, D. A., Rykoff, E. S., & McKay, T. A. 2003, GCN Circ. 2153 Hjorth, J., et al. 2003, Nature, 423, 847 Stanek, K. Z., et al. 2003, ApJ, 591, L17 Kawabata, K. S., et al. 2003, ApJ, 593, L19 Tiengo, A., Mereghetti, S., Ghisellini, G., Rossi, E., Ghirlanda, G., Kinugasa, K., et al. 2003, in preparation & Schartel, N. 2003, A&A, in press (astro-ph/0305564) Kobayashi, S. 2000, ApJ, 545, 807 Torii, K. 2003, GCN Circ. 1986 Kumar, P., & Panaitescu, A. 2000, ApJ, 541, L9 Uemura, M. 2003, GCN Circ. 1989 Landolt, A. U. 1992, AJ, 104, 340 Uemura, M., et al. 2003, Nature, 423, 843 Lazzati, D., Rossi, E., Covino, S., Ghisellini, G., Malesani, D. 2002, Vanderspek, R., et al. 2003, GCN Circ. 1997 A&A, 396, L5 van Paradijs, J., et al. 1997, Nature, 386, 686 Li, W., Filippenko, A. V., Chornock, R., & Jha, S. 2003, ApJ, 586, Wijers, R. M. A. J., Rees, M. J., & Mészáros, P. 1997, MNRAS, 288, L9 L51 Lipnov, V., et al. 2003b, GCN Circ. 2091 Wozniak, P., et al. 2003, GCN Circ. 1757 TheEarliestOpticalObservationsofGRB030329 5

Fig. 1.— A wide ﬁeld YC-S image around the constellation of Leo at the burst trigger time. This image is a stack of 8×8 s frames starting at 11:36:42 UT (33 s before trigger time). A circle shows the position of the afterglow where no signiﬁcant object is apparent.

Fig. 2.— The ﬁrst 30 s exposure (mean epoch 12:52:54 UT) by ART T1. 30′×30′ ﬁeld centered at the afterglow is shown. North is up and east to the left. 6 Torii, et al.

Fig. 3.— From top to bottom, ART T2 (circles), ART T1 (x marks), Ageo (plus marks), Saji (triangle), and GETS (squares) lightcurves are shown. The latter four data sets are shifted by +0.5, +1.0, +1.0, and +1.5 magnitudes. Total errors for T2 data are 0.03 mag at the ﬁrst part and 0.05 mag at the last part. For clear presentation of ART T1 and Ageo lightcurves, a bright comparison star with high signal to noise ratio, USNO A2.0 1050.06353017, is used here. TheEarliestOpticalObservationsofGRB030329 7

Table 1 Observation Log.

Site Instruments Filter Start Time (UT) End Time (UT) Integration [s] Number of frames Observer / PI Yatsugatake YC-S IR-cut 10:00:00 13:00:00 8 1350 Okamoto Wako ART 0.20 m + AP7p No 12:51:39 17:55:00 30 437 Torii Wako ART 0.25 m + AP6E No 12:52:09 17:54:52 60 277 Torii Ageo 0.25m + ST-9E No 13:38:12 16:03:50 4, 60 59 Kadota Gunma GETS 0.25m + AP7p No 14:21:55 17:55:06 30 181 Kinugasa Saji 1.03 m Rc, V 16:54:30 17:05:00 120 4 Oribe

Table 2 Photometric data obtained by ART T2.

Mean Epoch a Rc mag Mag error (1 σ) 0.052365 12.383 0.025 0.053106 12.419 0.026 0.053858 12.430 0.026 0.054610 12.441 0.026 0.055363 12.453 0.026 a days after burst trigger at 2003 March 29, 11:37:14.67 UT. Only the first 5 rows are tabulated here. The other data are available in electric format (machine readable table).