Gravity Field Recovery and Observation Noise Separation from Simultaneous Laser Ranging Interferometer and K-Band Ranging System Measurements

The unique instrumentation of GRACE Follow-On (GRACE-FO) offers two independent inter-satellite ranging systems with concurrent observations. Next to a K-Band Ranging System (KBR), which has already been proved during the highly-successfully GRACE mission, the GRACE-FO satellites are equipped with an experimental Laser Ranging Interferometer (LRI), which features a drastically increased measurement precision compared to the KBR. Having two simultaneous ranging observations available allows for cross-calibration between the instruments and, to some degree, the separation of ranging noise from other sources such as noise in the on-board accelerometer (ACC) measurements.  

In this contribution we present a stochastic description of the two ranging observation types provided by GRACE-FO, which also takes cross-correlations between the two observables into account. We determine the unknown noise covariance functions through variance component estimation and show that this method is, to some extent, capable of separating between KBR, LRI, and ACC noise. A side effect of this stochastic modelling is that the formal errors of the spherical harmonic coefficients fit very well to empirical estimates, which is key for combination with other data types and uncertainty propagation. We further compare the gravity field solutions obtained from a combined least-squares adjustment to KBR-only and LRI-only results and discuss the differences between the time series with a focus on gravity field and calibration parameters. Even though, at the moment, global statistics only show a minor improvement when using LRI ranging measurements instead of KBR observations, some parts of the spectrum and geographic regions benefit significantly from the increased measurement accuracy of the LRI. Specifically, we see a higher signal-to-noise ratio in low spherical harmonic orders and the polar regions.