ITSG‐Grace2018 is a new series of GRACE‐only gravity field solutions based on reprocessed GRACE observation data (L1B RL03) and the latest atmosphere and ocean dealiasing product (AOD1B RL06). It includes unconstrained monthly and constrained daily solutions, as well as a high‐resolution static gravity field. Compared to the previous ITSG release, we implemented a number of improvements within the processing chain and use updated background models. In an effort to better model all known error sources, we propagate synthetic orientation uncertainties of the star camera assembly to the antenna offset correction for intersatellite ranging observations. This enables the disentanglement of the stationary noise of the K‐Band system and the nonstationary noise of the antenna offset correction. We further incorporated uncertainties of the atmosphere and ocean dealiasing product to reduce temporal aliasing effects. To mitigate errors in the applied ocean tide model, we used constrained GRACE estimates of selected tidal constituents as an additional background model. Variability over quiet ocean areas suggests a 27% to 46% lower noise level compared to the current spherical harmonic solutions of the official processing centers (300 km Gaussian filter applied). To ensure that the low noise floor is not accompanied by signal loss, we examined drainage basin averages, which showed consistent amplitudes with the official GRACE time series. These evaluations lead to the conclusion that ITSG‐Grace2018 is a state‐of‐the‐art GRACE time series which exhibits an excellent signal‐to‐noise ratio.

Techniques enabling precise point positioning with ambiguity resolution (PPP-AR) were developed over a decade ago. Several analysis centers of the International GNSS Service (IGS) have implemented such strategies into their software packages and are generating (experimental) PPP-AR products including satellite clock and bias corrections. While the IGS combines individual orbit and clock products as standard to provide a more reliable solution, interoperability of these new PPP-AR products must be confirmed before they can be combined. As a first step, all products are transformed into a common observable-specific representation of biases. It is then confirmed that consistency is only ensured by considering both clock and bias products simultaneously. As a consequence, the satellite clock combination process currently used by the IGS must be revisited to consider not only clocks but also biases. A combination of PPP-AR products from six analysis centers over a one-week period is successfully achieved, showing that alignment of phase clocks can be achieved with millimeter precision thanks to the integer properties of the clocks. In the positioning domain, PPP-AR solutions for all products show improved longitude estimates of daily static positions by nearly 60% over float solutions. The combined products generally provide equivalent or better results than individual analysis center contributions, for both static and kinematic solutions.

This article describes the raw observation approach as implemented at Graz University of Technology to determine GNSS products like satellite orbits, clocks, and station positions. To assess the performance of the approach, 15 years (2003-2017) of observations from a network of 245 globally distributed IGS stations to the GPS constellation were processed on a daily basis using the IGS14 reference frame and antenna calibrations. The resulting products are evaluated against those determined by IGS analysis centers. Orbit fit quality relative to the IGS combination is comparable to the best-fitting solutions used for evaluation. Starting from early 2017, when the IGS switched to IGS14, the determined orbits fit better to the IGS combination than any other considered solution. Midnight discontinuities show good internal orbit consistency and no noticeable satellite block-dependency. Satellite clocks are comparable to the considered IGS analysis center solutions. Station positions differ from the IGS combination on a similar level to the solutions they were evaluated against. The temporal repeatability of station positions is slightly better than that of the IGS combination. The quality of resulting GNSS products confirms that the raw observation approach is well suited for the task of determining satellite orbits, clocks, and station positions. It provides an alternative to well-established approaches used by IGS analysis centers and simplifies the introduction of additional observables from new and modernized GNSS.

No abstract

No abstract