Characterizing Physical and Material Properties of Decameter-Size Earth Impactors

Small asteroids ranging from 1 − 20 meters in size impact the Earth 35 − 40 times per year (Brown et al. 2002; Bland and Artemieva 2006), often appearing as spectacular fireballs in the atmosphere. The largest of these objects can have kinetic energies equivalent to hundreds of kilotons of TNT, posing a hazard if they impact populated areas. As most recovered meteorites originate from 1 − 20 meter-size asteroids (Borovička 2015), this population in particular presents a unique opportunity to link together data from fireball, telescopic and meteorite observations. The properties of these small asteroids have to date been poorly characterized at a population level, as they are often at the detection limit of telescopic near-Earth object (NEO) surveys while also being relatively rare as Earth impactors. However, the amount of data on this population has grown significantly in recent years. In 2022, the US Space Force publicly released decades of previously classified fireball data from US Government (USG) satellite sensors, including light curves of intensity over time¹. This tranche of over one thousand recorded fireballs represents the most comprehensive dataset of meter-size and larger impactors to date.

In Chow and Brown (2025) we undertook the first population-level study characterizing the orbital properties of decameter-size Earth impactors with the new USG sensor data. We analyzed the dynamical origins of decameter-size impactors and NEOs, and evaluated possible explanations for the order-of-magnitude “decameter gap" between the observed impact rate from fireball data and the inferred impact rate from NEO models based on telescopic surveys. Here we present a companion study to our previous paper that characterizes the physical and material properties of these decameter-size impactors using the USG sensor light curves.

Previous studies using light curve data to analyze the physical properties of these small asteroids have generally proceeded by first generating a synthetic light curve by simulating the asteroid’s ablation and fragmentation in the atmosphere and then manually fitting the synthetic light curve to observations by adjusting various model parameters (e.g. Wheeler et al. 2017; Borovička et al. 2020; McFadden et al. 2024). However, this method is slow, labour-intensive, subject to parameter degeneracy and does not quantify uncertainty in the inferred model parameters. Previous attempts to develop automated approaches for ablation modeling using genetic algorithms (Tárano et al. 2019; Henych et al. 2023) have seen only limited success for a small number of fireballs and require an initial manual solution to be found first.

Motivated by the recent release of USG sensor data, we thus develop a novel Bayesian inference method that uses dynamic nested sampling (Skilling 2004, 2006; Higson et al. 2019) in conjunction with the semi-empirical fragmentation model of Borovička et al. (2013) that can probabilistically characterize the physical and material properties of Earth impactors from their light curves. Crucially, our nested sampling-based method allows for robust quantification of parameter uncertainty for the first time by estimating posterior distributions using a Bayesian framework. We first validate our method by applying it to several USG-recorded fireball events for which detailed light curve modeling has previously been conducted using independent ground-based observations and demonstrating that our results are consistent with previous estimates based on manual fitting. We then use our method to model the light curves of 13 decameter-size impactors we previously identified in Chow and Brown (2025), ultimately drawing population-level inferences about their physical properties such as mass and material strength for the first time.

As an example of our procedure, the above figure shows the resulting fit for one of the 13 decameter impactors we analyze, the 1994 February 1 Marshall Islands fireball. On the left, the USG-sensor recorded fireball light curve is plotted in red, while the 1σ, 2σ and 3σ uncertainties of the fit light curve of intensity versus height obtained with nested sampling are shown by the black shaded regions. The maximum log-likelihood solution is plotted as the blue line, while the detection limit of USG sensors is marked by the vertical red line. On the right, the marginal 2D nested sampling posterior distributions of dynamic pressure against mass released at each fragmentation point and at peak dynamic pressure are shown. In this presentation we will summarize the broad results of applying this procedure to all 13 decameter impactors and quantifying their relative strength.

¹jpl.nasa.gov/news/us-space-force-releases-decades-of-bolide-data-to-nasa-for-planetary-defense-studies/

 

References:

• Bland, P.A., & Artemieva, N.A.. 2006, Meteoritics & Planetary Science, 41 (4): 607–31. https://doi.org/10.1111/j.1945-5100.2006.tb00485.x

• Borovička, J. 2015, Proceedings of the International Astronomical Union, 10 (S318): 80–85. https://doi.org/10.1017/S174392131500873X

• Borovička, J., Spurný, P., & Shrbený, L. 2020, The Astronomical Journal, 160 (1): 42. https://doi.org/10.3847/1538-3881/ab9608

• Borovička, J., Tóth, J., Igaz, A., et al. 2013, Meteoritics & Planetary Science, 48 (10): 1757–79. https://doi.org/10.1111/maps.12078

• Brown, P., Spalding, R.E., ReVelle, D.O., Tagliaferri, E., & Worden, S.P. 2002, Nature, 420 (6913): 294–96. https://doi.org/10.1038/nature01238

• Chow, I., and Brown, P.G. 2025, Icarus, 429 (March): 116444. https://doi.org/10.1016/j.icarus.2024.116444

• Henych, T., Borovička, J., & Spurný, P. 2023, Astronomy & Astrophysics, 671 (March): A23. https://doi.org/10.1051/0004-6361/202245023

• Higson, E., Handley, W., Hobson, M., & Lasenby, A. 2019, Statistics and Computing, 29 (5): 891–913. https://doi.org/10.1007/s11222-018-9844-0

• McFadden, L., Brown, P.G., & Vida, D. 2024, Icarus, 422 (November): 116250. https://doi.org/10.1016/j.icarus.2024.116250

• Skilling, J. 2004, in AIP Conference Proceedings, 735: 395–405. Garching (Germany): AIP. https://doi.org/10.1063/1.1835238

• Skilling, J. 2006, Bayesian Analysis, 1 (4): 833–59. https://doi.org/10.1214/06-BA127

• Tárano, A.M., Wheeler, L.F., Close, S., & Mathias, D.L. 2019, Icarus, 329 (September): 270–81. https://doi.org/10.1016/j.icarus.2019.04.002

• Wheeler, L.F., Register, P.J., & Mathias, D.L. 2017, Icarus, 295 (October): 149–69. https://doi.org/10.1016/j.icarus.2017.02.011