Abstract Small asteroids and large meteoroids frequently impact the Earth, though their physical and material properties remain poorly understood. When observed as fireballs in Earth's atmosphere, these properties can be inferred from their ablation and fragmentation behavior. The 2022 release of previously classified United States Government (USG) satellite sensor data has provided hundreds of new fireball light curves, allowing for more detailed analysis. Here we present a new Bayesian inference method based on dynamic nested sampling that can robustly estimate these objects' physical parameters from their observed light curves, starting from relatively uninformative, flat priors. We validate our method against seven USG sensor‐observed fireballs with independent ground‐based observations and demonstrate that our results are consistent with previous estimates. We then apply our technique to 13 decameter‐size Earth impactors to conduct the most detailed population‐level study of their structure and material strength to date. We identify three structurally distinct groups within the decameter impactors. The first group are primarily structurally homogeneous, weak objects which catastrophically disrupt below MPa. The second group are heterogeneous objects which progressively fragment starting from MPa typically up to MPa. The third group are strong aggregates which remain mostly intact until MPa. Our results also suggest that decameter‐size asteroids fragment in two distinct phases: an initial phase at MPa and a second at MPa. While decimeter‐ to meter‐size objects typically lose most of their mass in the initial phase, larger decameter‐size objects instead lose most of their mass in the second phase.
Chow et al. (2026) studied this question.