ImprovedEp-TL-Lp Diagram and a Robust Regression Method

Abstract

The accuracy and reliability of gamma-ray bursts (GRBs) as distance indicators are strongly restricted by their systematic errors which are larger than statistical errors. These systematic errors might come from either intrinsic variations of GRBs, or systematic errors in observations. In this paper, we consider the possible origins of systematic errors in the following observables, (i) the spectral peak energies (E p ) estimated by Cut-off power law (CPL) function, (ii) the peak luminosities (L p ) estimated by 1 second in observer time. Removing or correcting them, we reveal the true intrinsic variation of the E p -T

Possible existence of the Ep-Lp and Ep-Eiso correlations for short gamma-ray bursts with a factor 5–100 dimmer than those for long gamma-ray bursts

“…This is because they adopted the values of Lp reported by multiple observation teams. The value of 4-msec peak luminosity is typically a few times larger than that of 1024-msec peak luminosity (Tsutsui et al 2011(Tsutsui et al , 2012 so that using different time resolution to define Lp would make artificial dispersion in the Ep-Lp correlation (Yonetoku et al 2010). Secondly, we integrate energy spectra between 1-100,000 keV in GRB rest frame to calculate Lp, while an energy range of 1-10,000 keV was considered in Zhang et al (2012).…”

“…To make Table 1, we collected all GRBs by the end of 2011 with the value of T rest 90 < 2s, the measured redshift z, the spectral peak energy Ep, the peak flux F p,obs and the fluence S obs within the energy range between Emin and Emax of each instrument. In order to obtain tighter correlations, the time bin of Fp, and then Lp, should be defined in the time in GRB rest frame as discussed in Tsutsui et al (2011Tsutsui et al ( , 2012 for LGRBs. However, the number of SGRBs is so small to determine the best time bin of Lp that we simply adopt here 64 msec in the observer frame for all SGRBs candidates.…”

Identification of subclasses of long gamma-ray bursts with cumulative light curve morphology

“…The needs for E p determined by the Band model and 64 msec light curve to correct the time scale of L p reduce the number of LGRBs to one thirds. However the small systematic error of our gold sample help us studying not only GRB physics but also dark energy density at high redshift universe [2]. …”

“…There were, however, some crucial issues with respect to the accuracy of the observation of prompt emissions of GRBs, which prevent us from recognizing the underlying diversity. To reduce systematic errors of observation, we compiled a gold data set of the prompt emission of 36 long gamma-ray bursts [1]. In table 1, we summarize the difference between the sample of previous works and our gold sample.…”

“…The division reveals that there are two peak energy (E p ) -peak luminosity (L p ) correlations, one for small-ADCL events and the other for large-ADCL events instead of one correlation [1].…”