J.-P. Dezalay

Energetic Neutrons, Protons, and Gamma Rays during the 1990 May 24 Solar Cosmic-Ray Event

The solar cosmic-ray event on 1990 May 24 can be divided into three phases: a first impulsive production of γ-rays and greater than 200 MeV neutrons; a second slower phase during which there were high-energy protons at the Sun for ~20 minutes producing pions and high-energy neutrons; and a third phase when the protons observed by the IMP 8 and GOES spacecraft and by neutron monitors were injected into interplanetary space. This third phase started after the onset of the event but before the second phase had ceased. We found that high-energy neutron production occurred during the last 60 s of the impulsive phase and at least the first 19 minutes of the second phase. During the second phase the high energy-neutron and γ-ray emissions decayed more slowly than either the 2.2 MeV or 4-7 MeV γ-ray line emissions. A two-component neutron energy spectrum that changes between the first and second phases gives a reasonable fit to the count rate increase recorded by the Climax neutron monitor. From the fit we infer that the integrated neutron emissivity at the Sun was ~3.5 × 1030 sr-1 for E > 100 MeV. The maximum intensity of P > 1.5 GV solar protons near the Earth was 4.5 × 103 (m2 sr s)-1. The differential solar proton energy flux (dJ/dE) as a function of rigidity at the Sun can be described by an evolving power-law spectrum. We estimate that the number of escaping protons with E > 30 MeV in the third phase was 7%-14% of the number of protons required to produce the solar neutron increase at the Earth. Although it is attractive to assume that the interplanetary solar protons simply leaked out from the trapping region at the Sun, the data suggest that the interplanetary solar protons were not from the population of energetic particles that produced the neutron and γ-ray emissions but were freshly accelerated during the third phase of the solar flare.

Hard X-ray and γ-ray observations of solar flares with the Phebus experiment

The Phebus experiment on board the GRANAT satellite provides temporal and spectral observations of solar and cosmic γ-ray bursts in the 0.1 ≃ 100 MeV nominal energy range. The experiment was turned on January 8, 1990 and is still in operation. In this paper we present the main characteristics of the Phebus experiment and we describe and discuss some of the observational properties of the 18 solar hard X-ray/γ-ray events detected during the first semester of the Phebus operation. It is found that: (i) events of a few minutes duration, detected above ∼ 100 keV, systematically show subsecond time variations; (ii) longer duration events (>5 min) do not exhibit fast time variations and generally consist of 1-min peaks superimposed on a less intense, sometimes harder, slowly varying component. In addition to these general trends we discuss in more detail three events for which significant count-rates have been detected above 10 MeV.

The Hardness-Intensity Correlation in Bright Gamma-Ray Bursts

The presence of a hardness-intensity correlation (HIC) is demonstrated in a sample of 77 bright bursts observed by the Phebus and Ulysses gamma-ray burst (GRB) experiments. Using simulations, we find a significant correlation (5 σ) between the peak spectral hardness and the peak intensity in Phebus. Moreover, the HIC is compatible with a single regression line over the full range of intensities. This result, together with earlier ones, shows that the HIC is a fundamental property of GRBs that has important consequences for statistical studies of these events. First, the presence of a HIC implies that the intensity distribution (and V/Vmax) depends on the energy range of the instrument. Second, we show that the HIC, combined with V/Vmax, constrains the spatial distribution of the bursters. Bright events in our sample exhibit a significant HIC and V/Vmax = 0.5. The interpretation of these observations is different for cosmological and galactic models. In Euclidean space, two features are required to explain the observations: bursters must belong to a homogeneous, bounded population and must have an intrinsic hardness-luminosity correlation. In a cosmological scenario, the apparent contradiction between the HIC (which requires high redshifts) and V/Vmax = 0.5 (characteristic of nearby bursters) can be resolved by assuming a density evolution of the sources.

Short Gamma-Ray Bursts observed by PHEBUS

Statistical studies of Gamma-Ray Burst (GRB) properties have recently led to the discovery of a subclass within the population of classical events (Dezalayet al. 1992, Kouveliotouet al. 1993). Bursts belonging to this subclass are characterized by short durations, typically less than 2 seconds, and harder spectra on average. Using the PHEBUS GRB data set, we analyse the distributions of peak intensity, hardness ratio, and duration of the two subclasses. We also compare the sum spectra obtained with the brightest events to determine the ratio of total energies observed for each population.

OBSERVATION OF AN EXTRA EMISSION AROUND 2 MeV IN THE SPECTRUM OF AN INTENSE GAMMA-RAY BURST

We present a time-averaged spectral analysis (0.1¨100 MeV) of two intense c-ray bursts detected by the PHEBUS experiment. The spectra of these two events are observed to peak between 1 and 2 lf l MeV. For the least intense of the two bursts, a single maximum in is seen. For the other, there is a lf l suggestion of two peaks, with similar strengths, at D1 and D2 MeV. The most commonly used photon models, with four or —ve parameters, provide inadequate —ts to the latter spectrum. We have tested two seven-parameter models that satisfactorily —t the observations by the two brightest detectors of the stronger event. Both of them are built from and compared with the standard four-parameter c-ray burst (GRB) function. The —rst model assumes an additional emission described by a Gaussian expression. The second one supposes a further spectral break at higher energy. For the two models, we —nd an extra emission around 2 MeV relative to the standard GRB model. In addition, the second seven-parameter model supports a sudden change in the photon spectral index from about ( 2t o(3.6 over a restricted energy range near 2.4 MeV. The probability of rejecting, as due to chance, these two models relative to the standard four-parameter expression does not exceed 6.8 ) 10~8. Although its origin is unclear, the 2 MeV emission excess may provide important constraints on the physical conditions and radiation mechanisms at the emission site of c-ray bursts. Subject heading: gamma rays: bursts

Search for extended hard?-ray emission in GRB's

The PHEBUS instrument aboard the GRANAT observatory is able to record spectra of gamma-ray bursts in a broad energy range from 0.1 to 100 MeV. Over 180 bursts were observed by PHEBUS during five years of operation. They represent a very rich sample of events widely differing in durations, light curves and energy spectra. The problem of burst behaviour in the decay phase of the GRB events is discussed in this paper. We present here the results of the search for emission from GRB sources in 100 keV - 100 MeV range after the main burst has finished. Presence of extended emission was found in about 10% of the gamma-ray bursts detected by PHEBUS.

Gross spectral differences between bright and very bright gamma-ray bursts

We consider a sample of 77 GRBs recorded by Phebus on GRANAT and the GRB detector on Ulysses in four years of operation. The comparison of the peak counts measured by these two instruments clearly shows the existence of a correlation between the intensities and the spectral hardnesses of these bursts. As sources homogeneously distributed in an Euclidean space should not exhibit any Hardness-Intensity correlation, we conclude that our sample is made of bursters which are not distributed homogeneously in space. This conclusion applies to bright (nearby?) bursters with peak flux above 8 ph cm−2 s−1. We also note that, while non-homogeneous, our GRB sample exhibits the canonical value 〈V/Vmax〉=0.5, confirming (if needed) that the test V/Vmax is efficient to reject homogeneity but not to prove it.

The hardness-intensity correlation and the spatial distribution of GRBs

We demonstrate the presence of a hardness-intensity correlation (HIC) in a sample of 77 bright GRBs detected by Phebus/GRANAT and the Ulysses GRB detector. Using simulations, we find that the peak spectral hardness and the peak intensity in Phebus are correlated at a 5 σ confidence level. This significant correlation is associated with a 〈V/Vmax〉=0.5 in Ulysses. These two results together imply several constraints on the spatial distribution of bursters. In a cosmological scenario, the apparent contradiction between the presence of a HIC, requiring high redshifts, and 〈V/Vmax〉=0.5, characteristic of nearby bursters, can be resolved by assuming a density evolution of sources. In this case our calculations show that the typical bright sources observed by Phebus and Ulysses should be at zmax≈2−3 to explain the strength of the hardness-intensity correlation.

Spectral properties of GRB's observed by PHEBUS

More than 180 gamma-ray bursts have been observed by the PHEBUS instrument aboard GRANAT observatory since its launch in December, 1989. A large database of the burst spectra in 0.1-100 MeV energy range has been accumulated that shows a wide variety of their evolution. Among the great majority of events displaying standard hard to soft evolution of spectra some bursts exhibit unusual behaviour with hardness ratio rising with time. We present here the brightest examples for both types of events and discuss also the question of the presence of lines in burst spectra.