Recent GRACE Follow-On Laser Ranging Interferometer Gravity Field Processing at JPL

The GRACE Follow-On (GRACE-FO) missions Laser Ranging Interferometer (LRI) technology demonstration has produced ranging data functionally equivalent to the missions dual-band microwave (KBR) ranging data for most months between June 2018 and June 2023. Despite the LRI having much better measurement precision than KBR, in the presence of temporal aliasing errors the quality of unconstrained spherical harmonic gravity solutions using LRI data has been at best mostly comparable to solutions using KBR data. We present recent efforts at JPL to make improved LRI gravity solutions in view of the upcoming GRACE Continuity (GRACE-C) missions sole reliance on LRI.Incorporating previous improvements in our de-glitching algorithm for removal of phase jumps in LRI piston phase (both "normal" magnitude and "mega-jumps"), and incorporating master cavity frequency corrections, we consistently reprocessed the whole missions LRI data. We reprocessed multiple flavors through level-1 with various choices: 1) estimating and applying both LRI scale factor and LRI time tag bias (ttb); 2) estimating both but applying only ttb; 3) estimating both but constraining ttb to zero, thus relying on LRI datation reporting (for select months); and 4) estimating neither scale factor nor ttb. For the first two of these, our scale factor and ttb estimates agreed well with the Albert Einstein Institute (AEI) LRI1B v50 and v54 values. Motivated by initial experiments, we used settings for our compression and differentiation algorithm acting on LRI data that were identical to such settings for the same algorithm acting on KBR data. This minimized data processing differences between the two ranging products and produced lower-rate (5 s rather than the default 2 s) LRI1B containing instantaneous range, range rate, and range acceleration.For level-2 processing, we adapted prior 2 s LRI data edits made based on LRI range acceleration postfit residuals for the new 5 sec LRI data, but also needed to do fresh data editing for the early mission timespan (due to GPS data). We optimized relative weighting between LRI and GPS and tried both traditional (white noise model) and VCE (determining full colored noise model) solution strategies. We omitted additional estimation of scale factor or ttb within level-2 solution parameterizations. Comparison of gravity fields between repro flavors supports necessity of the planned scale factor unit (SFU) on GRACE-C. Comparison of the "best-flavor" and "best weighting" monthly gravity fields (52 in all) against their KBR equivalents (accounting for KBR-vs-LRI data coverage differences) also supports the sufficiency of our LRI processing and solution methods for GRACE-C.The research presented in this abstract has been carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration.The authors' copyright for this abstract/supplementary material is transferred to 2024 California Institute of Technology. Government sponsorship acknowledged.