Observational evidence for the formation of trapping horizons in astrophysical black holes

Black holes in general relativity are characterized by their trapping horizon, a one-way membrane that can be crossed only inwards. The existence of trapping horizons in astrophysical black holes can be tested observationally using a reductio ad absurdum argument, replacing black holes by horizonless configurations with a physical surface and looking for inconsistencies with electromagnetic and gravitational wave observations. In this approach, the radius of the horizonless object is always larger than but arbitrarily close to the position where the horizon of a black hole of the same mass would be located. Upper bounds on the radius of these alternatives have been provided using electromagnetic observations (in optical/IR band) of astronomical sources at the center of galaxies, but lower bounds were lacking, leaving unconstrained huge regions of parameter space. We show here that lower bounds on the radius of horizonless objects that do not develop trapping horizons can be placed using observations of accreting systems. This result is model-independent and relies only on the local notion of causality dictated by the spacetime geometry around the horizonless object. These observational bounds reduce considerably the previously allowed parameter space, boosting the prospects of establishing the existence of trapping horizons using electromagnetic observations.