Interaction of Deterministic Errors in GRACE-Continuity

Abstract

The overall error in gravity field products derived from satellite-to-satellite tracking gravity missions like the Gravity Recovery and Climate Experiment Follow-On (GRACE-FO) is driven by a combination of temporal aliasing error, and measurement system error consisting of both stochastic and deterministic sources. While the impact of each of these error sources has been extensively documented in the literature, there has been relatively little study of how the phasing of deterministic errors, or tones, propagates into the gravity field solution.  Because these errors are primarily driven by on-orbit thermal variations, this study uses flight data from GRACE-FO thermistors alongside high-fidelity thermo-mechanical simulations that map temperature variations to structural deformation to investigate the phase and amplitude of these tones.We utilize these data to derive the most probable phasing of twice-per-revolution tones in both inter-satellite ranging and accelerometer measurements. We are particularly interested in the interplay of these tones and the twice-per-revolution component of absolute frequency error in the inter-satellite ranging signal. Unlike once-per-revolution errors, which are largely filtered out by empirical parameterization during standard orbit determination, twice-per-revolution errors survive the filtering process and map directly into the estimated gravity field. While we exclusively use GRACE-FO data and numerical simulations of the GRACE-Continuity flight system in this work, the principles of this work apply to any gravity mission using an inter-satellite ranging system.  We demonstrate that tone phasing plays a critical role in the magnitude of the resulting gravity field degree variance, particularly in C20. We conclude by discussing strategies to account for deterministic measurement system errors.