The clear characteristic time-scale picked out by the low-frequency quasi-periodic oscillations (QPOs) seen in many black hole and neutron star binaries has the potential to provide a very powerful diagnostic of the inner regions of the accretion flow. However, this potential cannot be realized without a quantitative model for the QPO. We have recently shown that the same truncated disc/hot inner flow geometry which is used to interpret the spectral transitions can also directly produce the QPO from Lense–Thirring (vertical) precession of the hot inner flow. This correctly predicts both the frequency and spectrum of the QPO, and the tight correlation of these properties with the total spectrum of the source via a changing truncation radius between the disc and hot flow. This model predicts a unique iron line signature as a vertically tilted flow illuminates different azimuths of the disc as it precesses. The iron line arising from this rotating illumination is blueshifted when the flow irradiates the approaching region of the spinning disc and redshifted when the flow irradiates the receding region of the disc. This gives rise to a characteristic rocking of the iron line on the QPO frequency which is a necessary (and probably sufficient) test of a Lense–Thirring origin. This is also an independent test of disc truncation models for the low/hard state, as vertical precession cannot occur if there is a disc in the midplane.
We show that it may be possible to observe this effect using archival data from the Rossi X-ray Timing Explorer or XMM–Newton. However, a clean test requires a combination of moderate resolution and good statistics, such as would be available from a long XMM–Newton observation or with data from the proposed European Space Agency mission Large Observatory for X-ray Timing.