TITLE: GCN GRB OBSERVATION REPORT NUMBER: 3852 SUBJECT: GRB 050820a - corrections to GCN 3846 DATE: 05/08/22 11:36:41 GMT FROM: Dmitry Frederiks at Ioffe Institute V. Pal'shin and D.Frederiks on behalf of the Konus-Wind team: There were several misprints in our GCN 3846. The subject line should have read: ------------------------------------ "GRB 050820a - a long GRB like GRB041219a? (Konus-Wind observation)" ------------------------------------ The correct circular text should be: ------------------------------------ "S. Golenetskii, R.Aptekar, E. Mazets, V. Pal'shin, D. Frederiks and T. Cline on behalf of the Konus-Wind team, and A. Rau, A. von Kienlin, G. Lichti on behalf of the INTEGRAL SPI-ACS GRB team report: A long soft GRB triggered Konus-Wind at T0=23954.512s UT (06:39:14.512) on August 20. It was also detected by INTEGRAL SPI-ACS. Because the burst was soft, the SPI-ACS response was weak, and precise triangulation of this burst is not possible. But the estimated time delay is consistent with the position of GRB 050820a, detected by Swift-BAT ~260 sec before Konus-Wind trigger (Page, et al., GCN 3830; Cummings, et al., GCN 3835). The Konus-Wind GRB consists of two parts: from ~T0-35s to ~T0+29s and from ~T0+135s to ~T0+235s and is much more intense than the Swift burst (precursor?), which was weakly seen by Konus-Wind in background data. The total burst duration is ~270 s. The Konus-Wind light curve recorded in the background mode is very similar to the light curve of the famous GRB 041219a (Vestrand et al., Nature, 435, 178 (2005)): weak precursor, main pulse at ~250 sec after it, second pulse at ~150 s after main and then - weaker pulse(s). The Konus-Wind light curve of this GRB can be seen at http://www.ioffe.rssi.ru/LEA/GRBs/GRB050820_T23954/ Based on these facts, we strongly suggest that this GRB is the main part of the GRB 050820a. In this case the optical transient reported by Fox and Cenko (GCN 3829) and by Wren et al. (GCN 3836) is in fact the prompt optical emission of very long GRB 050820a (the total burst duration including Swift precursor is ~500 sec). We hope, that the further analysis of this event will clarify this issue." --------------------------------- We apologize for these misprints.