2015MNRAS.454.2798S

We present unpublished data from a tidal disruption candidate in NGC 3599 which show that the galaxy was already X-ray bright 18 months before the measurement which led to its classification. This removes the possibility that the flare was caused by a classical, fast-rising, short-peaked, tidal disruption event. Recent relativistic simulations indicate that the majority of disruptions will actually take months or years to rise to a peak, which will then be maintained for longer than previously thought. NGC 3599 could be one of the first identified examples of such an event. The optical spectra of NGC 3599 indicate that it is a low-luminosity Seyfert/low-ionization nuclear emission-line region (LINER) with Lbol ∼ 10⁴⁰ erg/s. The flare may alternatively be explained by a thermal instability in the accretion disc, which propagates through the inner region at the sound speed, causing an increase of the disc scaleheight and local accretion rate. This can explain the ≤ 9 yr rise time of the flare. If this mechanism is correct then the flare may repeat on a time-scale of several decades as the inner disc is emptied and refilled.