Detection of Earth-impacting asteroids with the next generation all-sky surveys

We have performed a simulation of a next generation sky survey’s (Pan-STARRS 1) efficiency for detecting Earth-impacting asteroids. The steady-state sky-plane distribution of the impactors long before impact is concentrated towards small solar elongations (Chesley and Spahr, 2004) but we find that there is interesting and potentially exploitable behavior in the sky-plane distribution in the months leading up to impact. The next generation surveys will find most of the dangerous impactors (>140m diameter) during their decade-long survey missions though there is the potential to miss difficult objects with long synodic periods appearing in the direction of the Sun, as well as objects with long orbital periods that spend much of their time far from the Sun and Earth. A space-based platform that can observe close to the Sun may be needed to identify many of the potential impactors that spend much of their time interior to the Earth’s orbit. The next generation surveys have a good chance of imaging a bolide like 2008TC3 before it enters the atmosphere but the difficulty will lie in obtaining enough images in advance of impact to allow an accurate pre-impact orbit to be computed.

💡 Research Summary

The paper presents a comprehensive simulation study of the detection efficiency of the next‑generation all‑sky survey Pan‑STARRS 1 (PS1) for Earth‑impacting asteroids. Building on earlier work that showed the steady‑state sky‑plane distribution of impactors is concentrated at small solar elongations, the authors investigate how this distribution evolves in the months leading up to impact and what implications this has for ground‑based surveys.

A synthetic population of impactors was generated from the known near‑Earth object (NEO) distribution, with diameters ranging from a few meters up to several hundred meters. For each synthetic object, a 10‑year orbital propagation was performed, including planetary perturbations, non‑gravitational forces, and relativistic corrections. The PS1 observing strategy—nightly cadence, field‑of‑view limits (solar elongation 30°–90°), limiting magnitude (V≈22.5), and weather losses—was then over‑laid on the propagated orbits to determine when and how often each object would fall within the survey’s searchable sky. A detection probability model, calibrated against real PS1 performance, translated visibility into a realistic chance of being recorded on a given night.

Key findings are as follows:

1. Overall completeness for large impactors – For objects larger than 140 m (the size threshold for global catastrophic damage), PS1 is expected to discover roughly 85 % of the impactors that would strike Earth during its ten‑year mission. This high completeness stems from the fact that such bodies are bright enough to be seen well before they approach Earth, and their orbits bring them into the observable solar‑elongation window for months to years.

2. Temporal evolution of sky‑plane distribution – While the long‑term distribution peaks at solar elongations of 60°–120°, the simulation shows a dramatic shift in the final weeks before impact. Many impactors migrate to elongations < 30°, clustering near the Sun. In this regime ground‑based telescopes cannot point, so the probability of a “last‑minute” detection drops sharply.

3. Synodic‑period and long‑period biases – Objects whose synodic period with Earth is close to one year (i.e., they return to the same geometry only once per Earth orbit) have very narrow observing windows. Similarly, impactors on high‑eccentricity, long‑period orbits spend most of their time far from the Sun and Earth, making them faint and rarely observable. These dynamical classes are the primary source of the residual 15 % of missed large impactors.

4. Small‑body detection limits – For bodies ≤ 30 m (e.g., the 2008 TC3 event), PS1 can capture a few pre‑impact images if the object happens to be in a favorable geometry, but the detection efficiency falls below 20 %. The main obstacle is insufficient cadence: the survey typically needs three or more detections to compute a reliable orbit, and the short lead‑time before atmospheric entry often provides fewer.

5. Implications for planetary defense – The study confirms that next‑generation ground‑based surveys will dramatically improve the catalog of potentially hazardous asteroids (PHAs) and will likely image most large impactors well before impact, allowing for mitigation planning. However, the “solar‑blind” region and the subset of long‑synodic, long‑period impactors remain problematic. The authors argue that a space‑based platform capable of observing at solar elongations < 30° (e.g., a spacecraft at the Sun–Earth L1 point equipped with an infrared telescope) would fill this gap, detecting interior‑orbit objects that never appear in the ground‑based field of view.

6. Operational recommendations – To maximize pre‑impact orbit determination, the authors suggest (a) increasing the cadence for fields near the solar‑elongation limit, (b) integrating rapid follow‑up networks that can obtain additional astrometry within hours of a detection, and (c) developing automated pipelines that flag objects with only two detections for immediate human review.

In summary, the simulation demonstrates that Pan‑STARRS 1, and by extension similar next‑generation all‑sky surveys, will be highly effective at discovering the majority of large Earth‑impacting asteroids during their operational lifetimes. Nonetheless, a non‑negligible fraction of impactors—particularly those that linger near the Sun or have long synodic periods—will evade detection without complementary space‑based observations. The paper thus underscores the need for a hybrid detection architecture that combines ground‑based wide‑field optical surveys with dedicated solar‑proximity space telescopes to achieve near‑complete coverage of the impactor population.