Ejecta from the DART-produced active asteroid Dimorphos

The study investigates how impact-generated ejecta from an asteroid evolve into observable tails, a process previously inferred but not directly observed. Active asteroids are often detected after tails are fully developed, obscuring early-time dynamics. The DART mission provided a controlled, well-characterized impact on Dimorphos, enabling direct observation of ejecta evolution from minutes to weeks post-impact. This work aims to characterize the morphology, kinematics, and dynamical processes (gravity of the Didymos–Dimorphos system and solar radiation pressure) shaping the ejecta, and to relate these to the activation mechanisms inferred for naturally impacted active asteroids.

Prior observations linked some active asteroids (e.g., P/2010 A2, 311P/PANSTARRS) to impact-triggered activity, with multi-tailed and episodic behaviors documented. The Deep Impact mission on comet 9P/Tempel 1 provided a planetary-scale comparison, revealing diffuse, high-speed ejecta (average ~100 m s−1, up to ~300 m s−1) from a volatile-rich, porous comet nucleus. Impact ejecta theory (e.g., Housen & Holsapple) and experiments highlight how target properties (granularity, internal friction, inclusions, curvature, and projectile geometry) control ejecta cone angles and distributions. Models of ejecta dynamics under solar radiation pressure explain size sorting along tails and the formation of narrow, fan-like morphologies. This context motivates interpreting DART ejecta structures and tail development and reassessing past active asteroid observations in light of controlled-impact data.

Observational campaign: The Hubble Space Telescope (HST) imaged the Didymos–Dimorphos system from T + 15 minutes to T + 18.5 days after impact, with a cadence of approximately every 1.6 hours during the first 8 hours. Spatial resolution was ~2.1 km per pixel. Images covered various epochs to capture early cone formation through tail development. Image geometry (Sun, DART vector, and sky-plane orientation) was characterized. Image analysis: The team identified and tracked morphological features (linear features, clumps/blobs, arcs, curved streams s1 and s2, circular and arc-like structures, and the tail). Radial motions were measured to estimate projected speeds. A simple geometric model using feature position angles constrained the 3D ejecta cone opening angle and centerline orientation. Feature evolution was analyzed to separate regimes dominated by system gravity versus solar radiation pressure. Dynamics and photometry: The effect of Didymos–Dimorphos gravity on slow ejecta (<~1 m s−1) was inferred from wrapping/rotation around the system. Solar radiation pressure effects were diagnosed by size-dependent spreading along the sunward–antisunward axis and the development of wing-like structures. Tail brightness profiles were used to derive differential particle size distribution power-law exponents, distinguishing micrometer–millimeter and millimeter–centimeter regimes. Tail length and width were measured to infer initial speeds of the slowest ejecta. Secondary tail appearance and geometry were cataloged and compared against candidate mechanisms in extended analyses.

- Early-time ejecta morphology: Within hours, a hollow cone-shaped ejecta curtain developed with distinct linear features and clumps extending to ~500 km; features expanded radially at constant projected speeds of a few to ~30 m s−1, indicating little immediate influence from gravity or radiation pressure. - Ejecta cone geometry: Modeling of position angles indicates a 3D opening angle of 125° ± 10° with a centerline at position angle 67° ± 8°, nearly parallel to DART’s incoming trajectory; the cone is wider than typical vertical impacts into granular media, consistent with target curvature, internal friction, and projectile geometry effects. - Comparison to Deep Impact: Dimorphos ejecta were more structured and slower than the diffuse, faster ejecta from Tempel 1 (average ~100 m s−1, max ~300 m s−1), plausibly reflecting different target structures (rubble-pile, bouldery surface vs. porous, volatile-rich cometary subsurface). - Gravity-dominated phase (T ~ 0.7–2.1 days): Slower ejecta (<~1 m s−1) formed curved streams (s1 north, s2 south) that wrapped around Didymos under the system’s gravity, with small curvilinear features (116–119) behaving similarly. - Transition to radiation-pressure dominance: Solar radiation pressure sorted particles by size along the sunward–antisunward axis. The northern stream (s1) widened into a wing-like feature with a sharp sunward edge (indicating a maximum particle size cutoff), while southern features separated into individual strands (120–124) and were stretched antisunward. - Tail formation and growth: A narrow antisunward dust tail emerged within hours and quickly exceeded 1,500 km in projected length. Around several days post-impact, the tail showed a sharp southern edge and a more diffuse northern edge, resembling impact-produced active-asteroid tails (e.g., P/2010 A2). Tail width (~1 arcsec) is consistent with initial speeds comparable to Dimorphos’ orbital speed, implying dominance of the slowest ejecta. - Particle size distribution: Tail brightness profiles imply a differential size distribution with power-law exponent −2.7 ± 0.2 for radii ~1 µm to a few mm, steepening to −3.7 ± 0.2 for larger particles up to a few cm; particle sizes along the tail evolved over time, with early tails dominated by micrometer-sized dust and later tails by cm-scale particles within the HST field. - Continued mass loss: Ejecta continued to leave the system through at least T + 15 days in the observations. - Secondary tail: A second, fainter tail appeared between ~T + 5.7 and T + 8.8 days, offset by ~4° to the north of the primary tail, producing a transient fan-shaped morphology; its cause remains unclear but is morphologically consistent with multi-tailed active asteroids.

The observations directly document the transformation of impact ejecta into an active-asteroid-like tail under controlled impact conditions. The early kinematics and morphology confirm that initial ejecta structures are set by impact dynamics (cone geometry, velocity field) before evolving under the combined influences of the binary’s gravity and solar radiation pressure. The gravity-dominated wrapping of slow ejecta and subsequent size-dependent spreading under radiation pressure explain the emergence of wing-like features and the narrow primary tail. The measured particle-size distributions and their temporal evolution suggest that observed particle sizes in active asteroid tails depend strongly on the age of the tail; size sorting and preferential retention of larger particles at later times reconcile previous observations, such as the relative lack of sub-millimeter dust reported months after the P/2010 A2 event. Differences from the Deep Impact case emphasize the role of target structure and composition in ejecta morphology, supporting interpretations that rubble-pile asteroids produce heterogeneous, structured ejecta. Overall, the DART observations provide a robust template to interpret naturally impacted active asteroids and to link tail morphology to impact parameters, target properties, and time since activation.

This study shows that an asteroid can be activated by an impact, producing complex ejecta that evolve into a sustained tail through gravitational interactions and solar radiation pressure. The controlled DART impact on Dimorphos provides the first direct, time-resolved observations from minutes to weeks of this process, constraining ejecta cone geometry, speeds, tail formation, and particle-size distributions. These results offer a framework for reassessing and interpreting observations of previously discovered active asteroids likely triggered by impacts, including the dependence of observed particle sizes on tail age. Future work should determine the origin of secondary tails, refine models linking target properties to ejecta morphology, and extend observations over longer timescales and geometries to track the full evolution and dispersion of ejecta.

- Affiliation mapping and some extended data/methods details are not included in the provided excerpt; certain mechanistic interpretations (e.g., origin of the secondary tail) remain unresolved. - Observational constraints include HST pointing drift (smearing several to ~14 pixels in some exposures), limited field of view relative to >1,500 km tail extents, and reliance on projected (sky-plane) measurements that can obscure true 3D velocities and orientations. - The observing window spans up to T + 18.5 days; longer-term evolution and final dispersal are not captured here. Some feature identifications became uncertain at later times due to faintness and rapid morphological changes.