Glimpses of the past activity of Sgr A* inferred from X-ray echoes in Sgr C

International audienceContext. For a decade now, evidence has accumulated that giant molecular clouds located within the central molecular zone of our Galaxy reflect X-rays coming from past outbursts of the Galactic supermassive black hole. However, the number of illuminating events as well as their ages and durations are still unresolved questions. Aims. We aim to reconstruct parts of the history of the supermassive black hole Sgr A* by studying this reflection phenomenon in the molecular complex Sgr C and by determining the line-of-sight positions of its main bright substructures. Methods. Using observations made with the X-ray observatories XMM-Newton and Chandra between 2000 and 2014, we investigated the variability of the reflected emission, which consists of a Fe Kα line at 6.4 keV and a Compton continuum. We carried out an imaging and a spectral analysis. We also used a Monte Carlo model of the reflected spectra to constrain the line-of-sight positions of the brightest clumps, and hence to assign an approximate date to the associated illuminating events. Results. We show that the Fe Kα emission from Sgr C exhibits significant variability in both space and time, which confirms its reflection origin. The most likely illuminating source is Sgr A*. On the one hand, we report two distinct variability timescales, as one clump undergoes a sudden rise and fall in about 2005, while two others vary smoothly throughout the whole 2000−2014 period. On the other hand, by fitting the Monte Carlo model to the data, we are able to place tight constraints on the 3D positions of the clumps. These two independent approaches provide a consistent picture of the past activity of Sgr A*, since the two slowly varying clumps are located on the same wavefront, while the third (rapidly varying) clump corresponds to a different wavefront, that is, to a different illuminating event. Conclusions. This work shows that Sgr A* experienced at least two powerful outbursts in the past 300 years, and for the first time, we provide an estimation of their age. Extending this approach to other molecular complexes, such as Sgr A, will allow this two-event scenario to be tested further