Systematics and refinement of the Neptune ephemeris: Leveraging Gaia FPR and 42-year stellar occultation astrometry with an independent Triton orbit update

¹ Purple Mountain Observatory, Chinese Academy of Sciences, No. 10 Yuanhua Road, Nanjing 210033, China
² School of Astronomy and Space Science, University of Science and Technology of China, No. 96 Jinzhai Road, Hefei, Anhui 230026, China

^★ Corresponding author: yuanye@pmo.ac.cn

Received: 30 March 2025
Accepted: 15 September 2025

Abstract

Context. Neptune, as a critical archetype for ice giant and exoplanetary studies, remains underexplored in the years since the Voyager 2 mission (1989). Obtaining an accurate and precise ephemeris for Neptune system barycenter (NSB) is essential to upcoming 2040–2050 missions (e.g., Neptune Odyssey, Chinese orbiter) and an overall pulsar timing analysis. Current NSB ephemerides lack high-accuracy Gaia Celestial Reference Frame (Gaia-CRF) aligned constraints.

Aims. This study aims to refine NSB orbital accuracy by incorporating multi-source historical and new astrometric data and an updated Triton ephemeris, while addressing systematic errors in historical observations.

Methods. We refined the NSB orbit by incorporating (1) legacy datasets: historical angular measurements with necessary corrections and Voyager 2’s 1989 3D normal point; (2) Gaia Focused Product Release (FPR) Triton astrometry (2014–2019; sub-mas accuracy); (3) stellar occultations (1981–2022) reprocessed with Gaia DR3; and (4) the independently updated FORCES-8-MAIN-2024 (F8M24) Triton ephemeris from new Gaia-aligned ground-based CCD observations. The final orbit-fit solution yields the FORCES-8-BARY-2024 (F8B24) NSB ephemeris. Dynamical model systematics are evaluated through post-fit residuals of synthetic datasets generated from various reference planetary ephemerides. Orbital uncertainties are estimated with a Monte Carlo random parameter method and a Jackknife resampling method. The contributions from each dataset were determined using a cross-validation method against partial datasets.

Results. Key results include a sub-mas consistency between stellar occultation and Gaia FPR astrometry, both reliably aligned with the Gaia-CRF, and a mission-era positional uncertainty of ~500 km in F8B24. The along-scan residuals in Gaia FPR astrometry using F8M24 are near-Gaussian (mean and median <0.5 mas). In terms of essential synergy, Gaia DR3-based stellar occultations have provided dominant long-term constraints, Voyager 2 has preserved historical accuracy, and Gaia FPR has refined both present and future accuracy. We find that the exclusion of any of these datasets introduces rapid deviations from the orbit.

Conclusions. These findings establish modern occultation and Gaia astrometry as indispensable for high-fidelity orbit determinations, while confirming Voyager 2’s continued importance. Future works will incorporate new observations and optimize systematics to further enhance long-term accuracy and scientific applications.