Tanuka Chattopadhyay et al. 2007 ApJ 667 1017 doi:10.1086/520317
Tanuka Chattopadhyay1, Ranjeev Misra2, Asis Kumar Chattopadhyay3 and Malay Naskar4
Show affiliationsTwo different multivariate clustering techniques, the K-means partitioning method and the Dirichlet process of mixture modeling, have been applied to the BATSE gamma-ray burst (GRB) catalog, to obtain the optimum number of coherent groups. In the standard paradigm, GRBs are classified into only two groups, the long and short bursts. However, for both of the clustering techniques, the optimal number of classes was found to be three, a result that is consistent with previous statistical analysis. In this classification, the long bursts are further divided into two groups that are primarily differentiated by their total fluence and duration and hence are called low- and high-fluence GRBs. Analysis of GRBs with known redshifts and spectral parameters suggests that low-fluence GRBs have nearly constant isotropic energy output of 1052 ergs, while for the high-fluence ones the energy output ranges from 1052 to 1054 ergs. It is speculated that the three kinds of GRBs reflect three different origins: mergers of neutron star systems, mergers between white dwarfs and neutron stars, and collapse of massive stars.
gamma rays: bursts; methods: data analysis; methods: statistical
Issue 2 (2007 October 1)
Received 2007 March 26, accepted for publication 2007 May 25
Tanuka Chattopadhyay et al. 2007 ApJ 667 1017
Kyohei Terao et al 2006 J. Phys.: Condens. Matter 18 S653
Leslie W. Looney et al 1997 ApJ 484 L157
Nami Sakai et al. 2007 ApJ 663 1174
G C McConnell et al 2009 J. Neural Eng. 6 055005
Pedro Marronetti et al 2007 Class. Quantum Grav. 24 S43
C Roder et al 2003 J. Phys. D: Appl. Phys. 36 A188
K. L. Luhman et al. 2007 ApJ 659 1629
David N. Esch et al. 2004 ApJ 610 1213
W J Moses et al 2009 Environ. Res. Lett. 4 045005