SIMBAD references

We studied an exceptional period of activity of the anomalous X-ray pulsar 1E 1547.0-5408 in January 2009, during which about 200 hard X-ray/soft γ-ray bursts were detected by different instruments on board of ESA's γ-ray observatory INTEGRAL.The major activity episode (22 January 2009), which was detected by NASA's γ-ray telescope Swift, happened when the source was outside the field of view of all the INTEGRAL instruments. But we were still able to study the statistical properties as well as spectral and timing characteristics of 84 short (100ms-10s) bursts from the source during this activity period detected simultaneously by the anti-coincidence shield of the spectrometer SPI on board of INTEGRAL and by the detector of the imager IBIS/ISGRI. We find that the luminosity of the 22 January 2009 bursts of 1E 1547.0-5408 was ≥10⁴²erg/s above ∼80keV. This luminosity is comparable to that of the bursts of soft gamma repeaters (SGR) and is at least two orders of magnitude larger than the luminosity of the previously reported bursts from AXPs. Similarly to the SGR bursts, the brightest bursts of 1E 1547.0-5408 consist of a short spike of ∼100ms duration with a hard spectrum, followed by a softer extended tail of 1-10s duration, which occasionally exhibits pulsations with the source spin period of ∼2s. The ``short spike + tail'' template is not valid for all the observed bursts. We also observe a certain amount of ∼1s duration bursts either without a spike or with a spike delayed with respect to the onset of the burst. A third type of bursts is the ``flat-top'' burst of sub-second duration, which is similar to the ``precursors'' observed in the SGR flares. We find that the bursts of 1E 1547.0-5408 harden with increasing luminosity. Such a behavior is opposite of those observed in SGR bursts, but is similar to the hardness-luminosity relation observed in AXP 1E 2259+586. We confirm the existence of a correlation between the burst fluences and durations and the absence of a correlation between the burst fluence and the waiting time between subsequent bursts, previously established for the AXP 1E 2259+586. The observation of AXP bursts with luminosities comparable to the one of SGR bursts strengthens the conjecture that AXPs and SGRs are different representatives of one and the same source type, and that the AXP and SGR bursts are triggered by a similar type of instabilities.