2025-08-04 –, Online speaker
New asteroids and comets are continuously being discovered in the night sky. Such objects need to be dynamically characterised quickly, to determine the possibility of Earth impact and schedule follow-up observations before the object is lost. Here we present the Meerkat Asteroid Guard, an automated imminent impact warning service developed and operated at the ESA NEO Coordination Centre.
Meerkat continually downloads tracklets for newly discovered objects from the Minor Planet Center. For many of these new objects, the observation arc length is short. While the object's plane of sky position and motion may be well known, the remaining two parameters required to describe the orbit, the topocentric range and range rate, are not. Such short arcs lead to severe errors and degeneracies in traditional orbit determination methods. To overcome this, we employ the method of systematic ranging, whereby a grid of topocentric range and range rates have their orbits fitted with associated weighted root mean square error. From this error we derive a posterior probability distribution. By scanning a suitably dense grid, we can produce a statistical description of the most likely orbital solutions, and derive important information such as estimated size and impact probability. Meerkat operates 24/7, delivering email and phone alerts to subscribed users for imminent impactors and close approaches.
Over its five-year operational lifetime, Meerkat has successfully issued alerts for the past seven imminent impactors, from 2022 EB5 to most recently 2024 XA1. These alerts were vital for coordinating follow-up observations and preparing local authorities for fireball events.
After undergoing major software development and testing, v2.0 is ready to be released. Much of the functionality has been redesigned with ESA's new flight dynamics library, GODOT. This library is comprehensive and versatile, enabling the optimisation of the orbit determination, impact monitoring and ephemeris routines. New features such as the grid memory and propagation error handling have reduced the error rate of the systematic ranging algorithm. Under the hood, the software exploits a fully dockerised architecture for more efficient service and deployment. The software input is upgraded to meet modern standards, able to process ADES 2022 files and retrieve observations from the Minor Planet Center SBN Postgres database. In the output, new metrics flag potential false positive alerts to assist astronomers in time-critical follow-up observations.
The importance of an imminent impactor warning system cannot be overstated. With the advent of new surveys from ESA Flyeye, the Vera Rubin Observatory and NEO Surveyor, the number of new detections is predicted to increase dramatically. To this end, it is vital we ensure our planetary defence systems are as accurate and reliable as possible, ready to inform decision-makers when hazardous objects are found.
Hi! I'm a young graduate trainee at ESA, developing the Meerkat software. I'll be starting a PhD on machine learning for time domain astronomy in September at the University of Manchester.