HR 8799 is a young F0-type star with four directly imaged giant planets and two debris belts, one located exterior and another one interior to the region occupied by the planetary orbits. Having an architecture similar to that of our Solar System, but also revealing dissimilarities such as high masses of planets, a huge radial extent and a high mass of the outer debris belt, HR 8799 is considered to be a benchmark to test formation and evolution models of planetary systems. Here we focus on the outer debris ring and its relation to the planets. We demonstrate that the models of the outer disc, proposed previously to reproduce Herschel observations, are inconsistent with the ALMA data, and vice versa. In an attempt to find a physically motivated model that would agree with both observational sets, we perform collisional simulations. We show that a narrow planetesimal belt and a radiation pressure induced dust halo cannot account for the observed radial brightness profiles. A single, wide planetesimal disc does not reproduce the data either. Instead, we propose a two-population model, comprising a Kuiper-Belt-like structure of a low-eccentricity planetesimal population ("the classical Kuiper Belt") and a high-eccentricity population of comets ("scattered disc"). We argue that such a structure of the exo-Kuiper belt of HR 8799 could be explained with planet migration scenarios analogous to those proposed for the Kuiper Belt of the Solar System.

Many debris discs reveal a two-component structure, with a cold outer and a warm inner component. While the former are likely massive analogues of the Kuiper belt, the origin of the latter is still a matter of debate. In this work we investigate whether the warm dust may be a signature of asteroid belt analogues. In the scenario tested here the current two-belt architecture stems from an originally extended protoplanetary disc, in which planets have opened a gap separating it into the outer and inner discs which, after the gas dispersal, experience a steadystate collisional decay. This idea is explored with an analytic collisional evolution model for a sample of 225 debris discs from a Spitzer/IRS catalogue that are likely to possess a twocomponent structure. We find that the vast majority of systems (220 out of 225, or 98%) are compatible with this scenario. For their progenitors, original protoplanetary discs, we find an average surface density slope of −0.93 ± 0.06 and an average initial mass of 3.3 +0.4 −0.3 × 10 −3 solar masses, both of which are in agreement with the values inferred from submillimetre surveys. However, dust production by short-period comets and -more rarely -inward transport from the outer belts may be viable, and not mutually excluding, alternatives to the asteroid belt scenario. The remaining five discs (2% of the sample: HIP 11486, HIP 23497, HIP 57971, HIP 85790, HIP 89770) harbour inner components that appear inconsistent with dust production in an "asteroid belt." Warm dust in these systems must either be replenished from cometary sources or represent an aftermath of a recent rare event, such as a major collision or planetary system instability.

No abstract