A statistical comparison of the optical/UV and X-ray afterglows of gamma-ray bursts using the Swift Ultraviolet Optical and X-ray Telescopes

S. R. Oates 2 M. J. Page 2 P. Schady 2 M. De Pasquale 2 P. A. Evans 1 K. L. Page 1 M. M. Chester 0 P. A. Curran 2 T. S. Koch 0 N. P. M. Kuin 2 P. W. A. Roming 0 M. H. Siegel 0 S. Zane 2 J. A. Nousek 0 0 Department of Astronomy and Astrophysics, Pennsylvania State University, 104 Davey Laboratory, University Park , PA 16802, USA 1 X-ray and Observational Astronomy Group, Department of Physics and Astronomy, University of Leicester , Leicester LE1 7RH 2 Mullard Space Science Laboratory, University College London , Holmbury St. Mary, Dorking Surrey RH5 6NT A B S T R A C T We present the systematic analysis of the Ultraviolet/Optical Telescope (UVOT) and X-ray Telescope (XRT) light curves for a sample of 26 Swift gamma-ray bursts (GRBs). By comparing the optical/UV and X-ray light curves, we found that they are remarkably different during the first 500 s after the Burst Alert Telescope trigger, while they become more similar during the middle phase of the afterglow, i.e. between 2000 and 20 000 s. If we take literally the average properties of the sample, we find that the mean temporal indices observed in the optical/UV and X-rays after 500 s are consistent with a forwardshock scenario, under the assumptions that electrons are in the slow cooling regime, the external medium is of constant density and the synchrotron cooling frequency is situated between the optical/UV and X-ray observing bands. While this scenario describes well the averaged observed properties, some individual GRB afterglows require different or additional assumptions, such as the presence of late energy injection. We show that a chromatic break (a break in the X-ray light curve that is not seen in the optical) is present in the afterglows of three GRBs and demonstrate evidence for chromatic breaks in a further four GRBs. The average properties of these breaks cannot be explained in terms of the passage of the synchrotron cooling frequency through the observed bands, nor a simple change in the external density. It is difficult to reconcile chromatic breaks in terms of a single component outflow and instead, more complex jet structure or additional emission components are required. - Gamma-ray bursts (GRBs) are intense flashes of gamma-rays that can last from as little as a few milliseconds up to a few thousand seconds after the trigger. The duration and spectral hardness distributions are found to be bimodal, leading to a division of GRBs into two classes: short-hard GRBs (<2 s) and long-soft GRBs (>2 s) (Kouveliotou et al. 1993). The prompt gamma-ray emission is expected to be followed by an afterglow. The afterglow is most commonly seen in the X-rays, but is also observed in the optical/UV and, less commonly, down to radio wavelengths. The duration of the afterglow in the X-ray and optical/UV band varies considerably from GRB to GRB, and it has been observed to last for as little as a few hours up to a few months after the trigger. In the radio, the afterglow emission may be detected up to several years after the prompt gamma-ray emission. The afterglow from a short GRB tends to be fainter and short-lived in comparison with the long GRBs. For this reason, only long GRBs fall into the Oates et al. (2009) selection criteria. This paper is the successor to Oates et al. (2009) and therefore uses the same sample of 26 GRBs. Due to the unpredictability and rapid fading of these cosmic explosions, crucial clues on to their nature, their possible progenitors and their environments could only be obtained through deep and continuous observations of the afterglow. A rapid response satellite, Swift, which was launched in 2004 November, was specifically designed to observe these events. Swift houses three instruments designed to capture the gamma-ray, X-ray and optical/UV emission. The Burst Alert Telescope (BAT; Barthelmy et al. 2005) detects the prompt gamma-ray emission. The X-ray Telescope (XRT; Burrows et al. 2005) and the Ultraviolet/Optical Telescope (UVOT; Roming et al. 2005) observe the afterglow. The energy ranges of the BAT and the XRT instruments are 15350 and 0.210 keV, respectively, and the wavelength range of the UVOT is 16008000 . The coalignment of the XRT and UVOT instruments is ideal for observing GRB afterglows because observations of the X-ray and optical/UV afterglow are performed simultaneously. One of the early results that emerged from the first 27 X-ray afterglows collected by Swift is the existence of a canonical Xray light curve, which typically comprises four segments (Nousek et al. 2006). Here and throughout the paper, we will use the flux convention F t with and being the temporal and spectral indices, respectively. With this notation, the canonical X-ray light curve can be described as an initial steep decay segment (5 < X < 3) transitioning to a shallow decay phase (1.0 < X < 0.0; Liang, Zhang & Zhang 2007), then followed by a slightly steeper decay (1.5 < X < 1.0), which finally breaks again at later times. The last segment is usually identified as a post-jet-break decay (Zhang et al. 2006). However, the application of this model to all GRBs has recently been questioned by Evans et al. (2009) with a larger sample of 327 GRBs, 162 of which are considered by Evans et al. (2009) to be well sampled. This paper found that the canonical behaviour accounts for only 42 per cent of XRT afterglows. A statistical study of the UVOT light curves has recently shown that, although there are some similarities between the optical/UV and X-ray bands, in general the optical/UV afterglow does not behave in the same way as the X-ray one (Oates et al. 2009). In particular, the optical/UV light curves can either decay from the beginning of the observations or exhibit an initial rise and then a decay phase. In both cases, the decay segment is usually well fitted by a power law, although a small number of GRBs require a broken or a doubly broken power law. Moreover, by systematically comparing the optical/UV light curves with the XRT canonical model, Oates et al. (2009) found that among the four segments of the XRT canonical model the shallow decay segment has the most similar range of temporal indices to the optical/UV light curves. The temporal indices of the other segments of the XRT canonical light curve are steeper than the temporal indices of the optical/UV light curves. In this paper, we present a statistical cross comparison of the XRT and UVOT light curves for a sample of 26 GRBs presented in Oates et al. (2009). Table 1 lists these GRBs and their respective redshifts. The paper is organized as follows. In Section 2 we describe the data reduction and analysis. The main results are presented in Section 3. Discussion and conclusions follow in Sections 4 and 5, respectively. All uncertainties throughout this paper are quoted at 1 . 2 D ATA R E D U C T I O N A N D A N A LY S I S The 0.310 keV X-ray light curves were obtained from the GRB light curve repository at the UK Swift Science Data (...truncated)