Asteroid Science in the Era of the Large Synoptic Survey Telescope

Nathan Golovich | 20-ERD-025

Project Overview

This project was conceived to address a shortage of detections of near-earth objects (NEOs), which are minor planets that have orbits that bring them into the proximity of earth. In 2005, NASA was directed in the NASA Authorization Act of 2005 to catalog and characterize 90% of the NEOs that are at least 140 meters or larger. Once it became clear that this mandate would not be accomplished until the late 2030s at the earliest, most in government sought new ground- and space-based surveys that would cost billions of dollars. Our vision was to emphasize algorithmic improvements that leveraged LLNL's world-class computing infrastructure and support this effort with modeling and simulation of the NEO population. While we developed our algorithmic approach, we positioned ourselves as world leaders in the study of earth Trojan asteroids (ETAs), which are a niche subpopulation of NEOs with implications for planetary defense and solar-system science.

We set the strongest limits on this population with our own observations taken with the dark-energy camera in Chile. Then, we established LLNL as partners in the Zwicky Transient Facility (ZTF), which is a prototype of the next-generation ground-based survey. With ZTF data, we prototyped our detection method using LLNL high-performance computing (HPC) resources and institutional allocation to use the computing resources through three consecutive Computing Grand Challenge program awards. The method we have developed utilized LLNL investments in GPU-based HPC systems as well as institutional code bases for GPU parallelization, orbit modeling, and object detection. With our computing allocations we completed massive N-body simulations of ETAs and NEOs that have bolstered our understanding of the ETA observations made over the years and also explained why we have not detected a large population. Further, we extended the understanding of NEO orbits down to a few meters in size and made predictions of the detection rates with coming surveys, which are vital for planning follow- up missions, since the asteroid surveys of the 2020s and 2030s will detect so many asteroids that our ability to characterize them will be a substantial challenge.

Mission Impact

LLNL has had a long-running role in the development, building, and planning of the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST). This project continues that longstanding role for the lab but extends the purview to enhancing the asteroid-detection rate and providing projections for the detection rate of certain subpopulations. Additionally, our project has deep ties to LLNL's space program, which has grown rapidly over recent years. As such, we have helped explore solutions to emerging security challenges associated with technology surprise in space, and we have improved the security and resilience of the nation and its critical infrastructure from natural hazards, including hazardous asteroid impact by improving the rate of detection and characterization of these dangerous objects in space. Furthermore, we have provided access to petabytes of relevant data for the mission. We have especially had an impact with respect to recruiting and raising awareness to LLNL's expertise in space sciences and security. We have also curated data for the LLNL Data Science Institute's Open Data Initiative. This data also was used to establish a six-week data challenge for 60 University of California students from Riverside and Merced in the summer of 2021.

Publications, Presentations, and Patents

Rongaus, M., et al. "Asteroid Flyby Detection Performance Assessment for a Twilight Survey with the Vera C. Rubin Observatory." Poster, LLNL USAF ROTC Summer Internship Program. September, 2022

Yeager, T. and N. Golovich. 2022. "MEGASIM: Lifetimes and Resonances of Earth Trojan Asteroids - The Death of Primordial ETAs?" The Astrophysical Journal 938 (1) (2022); doi: 10.3847/1538-4357/ac8e63, LLNL-JRNL-835967.

Yeager, T. and N. Golovich. 2022. "Introducing MEGASIM: Multitudinous Earth Greek/Trojan Asteroid SIMulation." Research Notes of the AAS 6, 4:68 (2022); doi: 10.3847/2515-5172/ac62da. LLNL-JRNL-832783.

Golovich, N. "HPC-Enabled Asteroid Detection System in the Era of LSST." Presentation, LLNL Planetary Defense Group Meeting, Lawrence Livermore National Laboratory, Livermore February, 2022. LLNL-PRES-818716.

Schneider, M. "HPC-Enabled Asteroid Detection System in the Era of LSST." Presentation, SDA Data Science Working Group, Lawrence Livermore National Laboratory, Livermore 2022. LLNL-PRES-818716.

Yeager, T. "Earth Trojans." Presentation, ZTF Spring Meeting, Paris May 2022. LLNL-PRES-835241.

Yeager, T. "Early Results from the MEGASIM: Multitudinous Earth Greek (Not Trojan) Asteroid SIMulation." Presentation, AAS/Division of Dynamical Astronomy Meeting, 54(4) Manhattan April 2022. LLNL-PRES-835241.

Yeager, Travis. "N-Body Simulations of Earth Trojans." PLS Postdoc Seminar Series, Lawrence Livermore National Laboratory, Livermore, CA. March, 2022. LLNL-PRES-835173.

Golovich, N. "HPC-Enabled Asteroid Detection System in the Era of LSST." Presentation, Computing Grand Challenge Seminar Series, Lawrence Livermore National Laboratory, Livermore, CA. February, 2022. LLNL-PRES-818716.

Yeager, T. "The Lifetimes of Earth Trojan Asteroids and Tadpole Orbits." Presentation, AAS/Division of Dynamical Astronomy Meeting, 53(5), Virtual. May 2021.

Nathan G. "Data Science and Machine-Learning Opportunities in Asteroid Science." Presentation, LLNL Data Science Institute Summer, UC Merced-LLNL Data Science Challenge, Virtual. June 2021.

Nathan G. "Data Science and Machine Learning Opportunities in Asteroid Science." Presentation, LLNL Data Science Institute Summer, UC Riverside-LLNL Data Science Challenge. Virtual. June 2021.

Lifset, N., et al. 2021. "A Search for L4 Earth Trojan asteroids Using a Novel Track-Before-Detect Multiepoch Pipeline." The Astronomical Journal 161(6): 282 (2021); doi: 10.3847/1538-3881/abf7af.