Magnetars, neutron stars with magnetic fields a quadrillion times stronger than Earth's, occasionally emit short gamma-ray bursts. These bursts are rare and unpredictable, so systematic study requires years of all-sky monitoring. This work follows in the footsteps of the NICER and INTEGRAL/IBIS catalogs, which also accumulate burst statistics but in different energy ranges. Fermi GBM offers a unique combination of sensitivity and time resolution in the soft gamma-ray band, allowing it to catch even weak events missed by the trigger.
We used Fermi GBM trigger-mode data from 2008 to 2025. From the 12 NaI detectors, we selected those for which the angle between the source direction and the detector normal was less than 60 degrees, and the detector was not occulted by Earth or other parts of the satellite. To search for untriggered bursts, we applied Bayesian block analysis to the light curves with 8 ms resolution. Durations were determined from Bayesian blocks with 1 ms resolution in the 8–200 keV band. Spectra were fitted with four models: a power law, a power law with an exponential cutoff, and one or two blackbodies. The preferred model was selected using the Bayesian information criterion with a threshold of ΔBIC ≥ 10.
The catalog contains 1254 bursts from 17 magnetars, of which 688 are triggered and 566 were found in post-trigger data. Two sources dominate: SGR J1935+2154 (567 bursts) and SGR J1550–5418 (459 bursts). The average burst duration is about 200 ms for these two sources, while for the others it is noticeably shorter. The spectra of 68% of bursts are well described by two blackbodies, and 46% by a Comptonized power law. The average temperature of one blackbody is 14.3 keV; for SGR J1550–5418 it is higher (17.3 keV), indicating harder spectra. Burst energies range from 2×$10^{37}$ to 4×$10^{41}$ erg. The fluence distribution follows a power law with an index of about 1.6, consistent with the self-organized criticality model. The dependence of the emitting area on temperature for the two-blackbody model deviates from the Stefan-Boltzmann law: the power-law index is about −4.5…−5.4 instead of −4.
This work differs from neighboring catalogs in that it provides, for the first time, a uniform set of spectral parameters for bursts over such a long period and with such energy coverage. This allows comparison of burst properties across different magnetars and different activity episodes of the same source. It confirms that SGR J1550–5418 is systematically harder than SGR J1935+2154, while the spectral characteristics of SGR J1935+2154 remain almost unchanged from episode to episode. The deviation from the Stefan-Boltzmann law suggests that the emitting region is not a perfect blackbody and likely undergoes rapid evolution.
The catalog will serve as a starting point for studying burst mechanisms and for calibrating future instruments. New missions with polarimetry and higher time resolution will be able to test models of magnetospheric scattering and reconnection. It may also be possible to link burst properties to the magnetic field strength and age of the magnetar.
The results will impact neutron star physics, high-energy astrophysics, and the planning of X-ray and gamma-ray observations.
Further analysis may include searching for correlations between spectral parameters and rotation phase, as well as comparisons with fast radio bursts from SGR J1935+2154.
The catalog directly addresses the unresolved problem of the generation mechanism of short magnetar bursts: whether they result from crustal fractures or magnetospheric reconnection. The deviation from the Stefan-Boltzmann law and the differences in spectral hardness between sources provide constraints on models.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
This work presents a catalog of 1254 short gamma-ray bursts from 17 magnetars, accumulated over 17 years of Fermi GBM observations. It includes both triggered and untriggered events, with duration, spectrum, flux, and energetics determined for each. The catalog is three times larger than the previous five-year one and provides statistics for testing emission models.
Where its strength is
The unique combination of long-term monitoring, a broad energy range, and high time resolution has yielded a uniform set of spectral parameters for a large number of bursts from different sources. This makes it possible to compare burst properties across magnetars and activity episodes.
Where it could go next
Worth trying
Why this may matter
A catalog of this size and uniformity is a foundation for statistical studies of magnetar activity, testing emission models, and preparing future missions. It could become a benchmark for comparison with other instruments and for searching for rare events.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.52, against a usual 0.64. It looks like few have drilled nearby.
On a single day, January 22, 2009, the magnetar SGR J1550–5418 emitted 341 bursts, and accounting for detector dead time, the actual number could have been around 541.