Zsolt Bagoly

The directly measured redshifts of the long gamma-ray bursts - obtained from the afterglow data - are separated into two samples. The bursts detected first by the Swift satellite gives the "Swift-sample"; th eremaining ones forms the "non-Swift-sample". We perfored five statistical tests to compare the redshifts in these two samples. All these tests show that in the Swift sample the redshifts are on average larger.

Until 6 October 2005 sixteen redshifts have been measured of long gamma-ray bursts discovered by the Swift satellite. Further 45 redshifts have been measured of the long gamma-ray bursts discovered by other satellites. Here we perform five statistical tests comparing the redshift distributions of these two samples assuming - as the null hypothesis - identical distribution for the two samples. Three tests (Student's $t$-test, Mann-Whitney test, Kolmogorov-Smirnov test) reject the null hypothesis on the significance levels between 97.19 and 98.55%. Two different comparisons of the medians show extreme $(99.78-99.99994)$% significance levels of rejection. This means that the redshifts of the Swift sample and the redshifts of the non-Swift sample are distributed differently - in the Swift sample the redshifts are on average larger. This statistical result suggests that the long GRBs should on average be at the higher redshifts of the Swift sample.

Earlier classification analyses found three types of gamma-ray bursts (short, long and intermediate in duration) in the BATSE sample. Recent works have shown that these three groups are also present in the RHESSI and the BeppoSAX databases. The duration distribution analysis of the bursts observed by the Swift satellite also favors the three-component model. In this paper, we extend the analysis of the Swift data with spectral information. We show, using the spectral hardness and the duration simultaneously, that the maximum likelihood method favors the three-component against the two-component model. The likelihood also shows that a fourth component is not needed.

The redshift distribution of gamma-ray bursts collected in the BATSE Catalog is compared with the star formation rate. We aim to clarify the accordance between them. We also study the case of comoving number density of bursts monotonously increasing up to redshift 6-20. A method independent of the models of the gamma-ray bursts is used. The short and the long subgroups are studied separately. The redshift distribution of the long bursts may be proportional to the star formation rate. For the short bursts this can also happen, but the proportionality is less evident. For the long bursts the monotonously increasing scenario is also less probable but still can occur. For the short bursts this alternative seems to be excluded.

We have recalculated the durations of GRBs using a method which calculates the parabolical background fit using the standard background and burst intervals. This analysis indicates that the errors associated with the durations of the short bursts are larger than previously published in gamma-ray burst catalogs. Also the error bars are asymmetric, i.e. non-gaussian. .

We investigate the full randomness of the angular distribution of gamma-ray bursts (GRBs) detected by the BATSE (the Burst and Transient Source Experiment). We divided the BATSE sample into 5 subsamples (short1, short2, intermediate, long1, long2 based on their durations and peak fluxes and studied the angular distributions separately. We used three methods, the Voronoi tesselation, minimal spanning tree and multifractal spectra to search for non-randomness in the subsamples. To investigate the eventual non-randomness in the subsamples we defined 13 test-variables (9 from the Voronoi tesselation, 3 from the minimal spanning tree and one from the multifractal spectrum). We performed Monte Carlo simulations taking into account the BATSE sky-exposure function. We tested the randomness by introducing squared Euclidean distances in the parameter space of the test-variables. We concluded that the short1, short2 groups deviate significantly (99.90%, 99.98%) from the full randomness in the distribution of the squared Euclidean distances, however, it is not the case with the long subsamples. In the intermediate group, the squared Euclidean distances also yield a significant deviation (98.51%).

It is known that the soft tail of the gamma-ray bursts' spectra show excesses from the exact power-law dependence. In this article we show that this departure can be detected in the peak flux ratios of different BATSE DISCSC energy channels. This effect allows to estimate the redshift of the bright long gamma-ray bursts in the BATSE Catalog. A verification of these redshifts is obtained for the 8 GRB which have both BATSE DISCSC data and measured optical spectroscopic redshifts. There is good correlation between the measured and esti redshifts, and the average error is $\Delta z \approx 0.33$. The method is similar to the photometric redshift estimation of galaxies in the optical range, hence it can be called as "gamma photometric redshift estimation". The estimated redshifts for the long bright gamma-ray bursts are up to $z \simeq 4$. For the the faint long bursts - which should be up to $z \simeq 20$ - the redshifts cannot be determined unambiguously with this method.

Gamma-ray bursts are usually classified through their high-energy emission into short-duration and long-duration bursts. A third intermediate group has been identified on statistical grounds but its individual properties have not yet been studied in detail. Using the large sample of follow-up observations of GRBs produced during the Swift era we analyze the individual characteristics of this group. We find that

We observed 50 high-z (0.1 < z < 1) SN Ia host galaxies with the MIPS camera on board the Spitzer Space Telescope at the three photometric bands (24, 70 and 160 μm). The galaxies were selected to be detectable up to z 1. Their 24, 70 and 160 μm Hubblediagrams (redshift versus apparent brightness) provides information about the dust content of the galaxies, and show whether the visible-light observations should be corrected for an unexpected dust content.

The origin of dark bursts - i.e. that have no observed afterglows in X-ray, optical/NIR and radio ranges - is unclear yet. Different possibilities - instrumental biases, very high redshifts, extinction in the host galaxies - are discussed and shown to be important. On the other hand, the dark bursts should not form a new subgroup of long gamma-ray bursts themselves.

Several statistical studies - done also by the authors of this contribution - show that there are three subclasses of gamma-ray bursts. They can be called as short, intermediate and long ones, because they can be separated with respect to their durations. The short and intermediate bursts are distributed anisotropically on the sky. This behavior is highly remarkable, and can have a cardinal impact on the cosmology. The subject of this contribution is a survey of this topic.

In this paper we make an attempt to combine the two kinds of data from the Swift-XRT instrument (windowed timing and photon counting modes) and the from BAT. A thorough desription of the applied procedure will be given. We apply various binning techniques to the different data: Bayes blocks, exponential binning and signal-to-noise type of binning. We present a handful of lightcurves and some possible applications.

We have carried out a principal component analysis for 625 gamma-ray bursts in the BATSE 3B catalog for which non-zero values exist for the nine measured variables. This shows that only two out of the three basic quantities of duration, peak flux and fluence are independent, even if this relation is strongly affected by instrumental effects, and these two account for 91.6% of the total information content. The next most important variable is the fluence in the fourth energy channel (at energies above 320 keV). This has a larger variance and is less correlated with the fluences in the remaining three channels than the latter correlate among themselves. Thus a separate consideration of the fourth channel, and increased attention on the related hardness ratio $H43$ appears useful for future studies. The analysis gives the weights for the individual measurements needed to define a single duration, peak flux and fluence. It also shows that, in logarithmic variables, the hardness ratio $H32$ is significantly correlated with peak flux, while $H43$ is significantly anticorrelated with peak flux. The principal component analysis provides a potentially useful tool for estimating the improvement in information content to be achieved by considering alternative variables or performing various corrections on available measurements

Two applications of the multifractal (MFR) point pattern analysis are presented. First, we study the angular distribution of subclasses of gamma-ray bursts (GRBs), then we analyse the structure of extinction maps of dark molecular clouds obtained by near-infrared (NIR) star counts.

... Jakub ˇRípa Astronomical Institute, Faculty of Mathematics and Physics, Charles University, Czech Republic E-mail: ripa@sirrah.troja.mff.cuni.cz ... Acknowledgements. Thanks are due to István Horváth, Péter Mészáros and Roland Vavrek for valuable remarks. ...

Gamma-ray bursts are usually classified through their high-energy emission into short-duration and long-duration bursts. A third intermediate group has been identified on statistical grounds but its individual properties have not yet been studied in detail. Using the large sample of follow-up observations of GRBs produced during the Swift era we analyze the individual characteristics of this group. We find that

Gamma-ray bursts are usually classified through their high-energy emission into short-duration and long-duration bursts, which presumably reflect two different types of progenitors. However, it has been shown on statistical grounds that a third, intermediate population is needed in this classification scheme, although an extensive study of the properties of this class has so far not been done. The large amount of follow-up studies generated during the Swift era allows us to have a suficient sample to attempt a study of this third population through the properties of their prompt emission and their afterglows. Our study is focused on a sample of GRBs observed by Swift during its first four years of operation. The sample contains those bursts with measured redshift since this allows us to derive intrinsic properties. Intermediate bursts are less energetic and have dimmer afterglows than long GRBs, especially when considering the X-ray light curves, which are on average one order of magnitude fainter than long bursts. There is a less significant trend in the redshift distribution that places intermediate bursts closer than long bursts. Except for this, intermediate bursts show similar properties to long bursts. In particular, they follow the Epeak vs. Eiso correlation and have, on average, positive spectral lags with a distribution similar to that of long bursts. Like long GRBs, they normally have an associated supernova, although some intermediate bursts have shown no supernova component. This study shows that intermediate bursts are different from short bursts and, in spite of sharing many properties with long bursts, there are some differences between them as well. We suggest that the physical difference between intermediate and long bursts could be that for the first the ejecta are thin shells while for the latter they are thick shells.

Principal component analysis is a statistical method, which lowers the number of important variables in a data set. The use of this method for the bursts' spectra and afterglows is discussed in this paper. The analysis indicates that three principal components are enough among the eight ones to describe the variablity of the data. The correlation between spectral index alpha and the redshift suggests that the thermal emission component becomes more dominant at larger redshifts.