NEOForCE: Near-Earth Objects' Forecast of Collisional Events

TL;DR

NEOForCE introduces an independent near-Earth object impact monitoring framework that combines Line of Variations sampling with a semi-linear Partial Banana Mapping approach to propagate orbital uncertainties and identify potential Earth impactors up to a century ahead. By focusing on Earth’s proximity to curvilinear uncertainty regions and applying a targeted virtual impactor search on a refined parameter space, the method efficiently yields collision probabilities while maintaining completeness within a quantified threshold. Validation against NASA’s Sentry-II on five representative asteroids shows that NEOForCE reproduces most high-probability events and additionally uncovers extra low-probability collisions (in the range $10^{-7}$–$10^{-6}$) not reported by Sentry-II, highlighting its complementary role in planetary defense. The work also elaborates the classical and modified target-plane analyses, completeness limitations, and practical realizations, underscoring the tool’s potential as a robust cross-check for impact assessments across the community.

Abstract

Robust impact monitoring of near-Earth objects is an essential task of planetary defense. Current systems such as NASA's Sentry-II, the University of Pisa's CLOMON2, and ESA's Aegis have been highly successful, but independent approaches are essential to ensure reliability and to cross-validate predictions of possible impacts. We present NEOForCE (Near-Earth Objects' Forecast of Collisional Events), a new independent monitoring system for asteroid impact prediction. By relying on orbital solutions from DynAstVO at Paris Observatory and using an original methodology for uncertainty propagation, NEOForCE provides an alternative line of verification for impact assessments and strengthens the overall robustness of planetary defense. As other monitoring systems, NEOForCE samples several thousand virtual asteroids from the uncertainty region and integrates their orbits up to 100 years into the future. Instead of searching for close approaches of the virtual asteroids with the Earth, our system looks for times when the Earth comes close to the realistic uncertainty regions around them, which are mostly stretched along their osculating orbits. We also estimate the maximal impact probability, and only if this value is large enough do we continue to the next step. In this second step, we compute how the original asteroid orbit should be modified so that the new trajectory leads to an Earth impact, which allows us to confirm the possible collision and estimate the impact probability. We tested NEOForCE against NASA's Sentry-II system on five representative asteroids: 2000 SG344, 2005 QK76, 2008 JL3, 2023 DO and 2025 JU. NEOForCE successfully recovered nearly all possible collisions reported by Sentry-II with impact probabilities above e-7, demonstrating the robustness of our approach. In addition, NEOForCE identified several potential impacts at the e-7 - e-6 level that Sentry-II did not report.

Figures (16)

• Figure 1: Schematic illustration of the confidence curvilinear ellipsoid. Point A is the nominal position of the asteroid at time $t$, and point B is the VA on the main axis of the confidence ellipsoid at the same time $t$, which is closest to the Earth after projection onto its target plane. The arrow indicates the velocity direction of Earth with respect to the confidence ellipsoid. The bold line shows the nominal orbit of the asteroid. Image credit: Fig. 3 in 2020MNRAS.492.4546V.
• Figure 2: Schematic illustration of the different scenarios for grouping of VAs. The top chart shows the situation when the Earth is between two subsequent VAs (difference in $M$ has opposite signs). In this case we choose j+1-th VA as our primary representative and j-th as a secondary one. The bottom chart shows the situation when all the subsequent VAs are on one side from the Earth (difference in $M$ are all positive). In this situation we choose j+2-th as our primary representative. Note, that j+7-th VA is in another group but it will also be a representative of its group.
• Figure 3: Schematic illustration of the work of NEOForCE system. On the upper chart one can see the nominal asteroid (purple dot), the original uncertainty region at the epoch of observation (black ellipse), the LOV (horizontal line), VAs sampled along LOV (black dots). The green area represents the smaller uncertainty region assigned to this VA. The red denotes the VI. On the bottom chart one can see the nominal position of the asteroid at time of a possible collision, $t$ (point A), the osculating orbit of the nominal asteroid (black part of an ellipse), the Earth, the position of the VA at time $t$ (green dot), the uncertainty region assigned for this VA at time $t$ (green curved ellipse). The red dot (point B) represents the position of the VI. The dashed vector shows the relative motion of the Earth and the green uncertainty region. The Earth collides with the red dot.
• Figure 4: Schematic illustration of the target plane method. The ellipsoid represents the uncertainty region of an asteroid when it enters the sphere of influence, the plane is the target plane, the circle on the plane is the projection of the Earth, and the ellipse on the plane is the projection of the ellipsoid. Image credit: vavilov2018_iaaras.
• Figure 5: Vibrational stability equation of state $S_{\mathrm{vib}}(\lg e, \lg \rho)$. $>0$ means vibrational stability.