The GOCE satellite was launched in March 2009 and collected a unique gravity gradient dataset at a mean altitude of 260 km, later descending to 228 km, until re-entry in November 2013. The gravity gradients are not measured directly, but calculated from the measurements of six accelerometers that form the gravity gradiometer and three star trackers. Earlier investigations on the calibration and processing of those measurements showed that a significant improvement of the entire gravity gradient dataset was possible. In particular the perturbation that affected the cross-track gravity gradient Vyy in the regions around the geomagnetic poles was reduced to a large extent. These achievements led to the decision to perform a full reprocessing of the GOCE mission dataset in 2018, starting from accelerometer and star tracker measurements and processing all the way to global gravity field models. In our presentation, we will explain the new processing and calibration of the star tracker and accelerometer measurements and highlight the essential advances with respect to the previous processing and calibration. In addition, we will demonstrate the higher accuracy of the reprocessed GOCE gravity gradients.

ESA's dedicated gravity field mission GOCE (Gravity field and steady-state Ocean Circulation Explorer) successfully completed its science operations in 2013 as the first Earth Explorer mission in orbit. In 2014, the fifth releases of GOCE based global gravity field models of the Earth were published. Already at that time, the entire mission data set was used. Until now, the release 5 models are the highest resolution models of the static Earth's gravity field derived by satellites. On the one hand, high-low satellite-to-satellite tracking observations of GOCE by GPS satellites is used to derive the long wavelength part of the gravity field. On the other hand, observations of the core instrument - a gravitational gradiometer - are sensitive to the medium and short wavelengths of the Earth gravity field. Combining both observation groups provides a global model of the gravity field with a spatial resolution of about 70 km.
Recent studies have shown that the quality of the derived level 1B gravity gradients can be significantly improved, if a quadratic term is additionally considered in the calibration procedure. Consequently, ESA funded a reprocessing campaign by the High-level Processing Facility (HPF). Within that framework, global gravity field models are estimated and provided as higher level GOCE products.
One approach which can be used for the processing of the GOCE observations is the so called time-wise approach. Within this approach, only GOCE observations are used to estimate the gravity field model in a least-squares sense. The highly correlated gravity gradients in the gradiometer reference frame are used to derive normal equations in terms of global spherical harmonics. A lot of effort is spend on the stochastic modeling of the noise characteristics of the gravity gradients, to deliver in addition to the high quality gravity field a meaningful covariance matrix. The gradiometry normal equations are combined with normal equations determined from high-low satellite to satellite tracking to better capture the long wavelength of the gravity field.
Within this contribution, the release six of the official ESA time-wise model is presented. An improved time-wise processing is used, to estimate the new EGM_RIM_RL06 model form the reprocessed GOCE observations. An advanced detection of suspicious data is implemented and used in the estimation of decorrelation filters as well as for gravity field recovery. These decorrelation filters serve as a stochastic model for the gravity gradients in the least squares estimation of the spherical harmonic coefficients which represent the gravity field.
Within this contribution, the processing used to derive the new release is summarized. The improvements with respect to the former 5th release are highlighted. The characteristics and the quality level of the new solution is discussed. Finally the model is validated with comparisons to independent data sets. As an outlook, the combination with GRACE models towards a combined GOCO06S is shown.

The recently completely recalibrated and reprocessed Gravity Field and Steady-State Ocean Circulation Explorer (GOCE) gravity gradient data were combined with data from Laser Geodynamics Satellite (LAGEOS) 1/2 and Gravity Recovery and Climate Experiment (GRACE) to generate a new satellite-only gravity field model to degree 300. This work was done jointly by GFZ/Potsdam w+and CNES/Toulouse wiithin ESA's High Level Processing Facility (GOCE-HPF). The generation of this new global satellite-only gravity field model was done by using the direct approach and it's named DIR-R6. When compared to the previous gravity field models based on GOCE data, for instance to the earlier releases of ESA’s GOCE models, DIR-R6 is clearly more accurate, especially in its medium to short psherical marmonic wavelengths. The improvement of DIR-R6 is demonstrated among others by GPS/Leveling comparisons, orbit determination tests and an oceanographic evaluation. The improvement is due mainly to the presently excellent quality of the YY gradients, but also thanks to a different data filtering approach (low pass filter only).

With the delivery of the final GOCE rel. 5 gravity models in 2015, the nominal processing of GOCE data has been completed very successfully. All delivered level 1 and level 2 products are of high quality and meet the expected requirements. Since then, several experts from ESA and the GOCE High Level Processing Facility (HPF) team investigated in more detail remaining features visible in the GOCE products and identified further improvements in the processing strategies, which led to the decision in late 2017 to start a reprocessing campaign for the GOCE gradiometer data and to prepare a new release of GOCE level 2 products, i.e. precise science orbits and GOCE gravity field models and gradient grids. In addition, a few new GOCE level 2 products have been identified which further enhance the science return of the mission. In particular these are combined gravity field models incorporating newly observed airborne gravity data over Antarctica (filling the polar gap of GOCE) and making use of GOCE data for studies of ionospheric electron content and thermosphere density and winds. The reprocessing campaign recently has been completed with delivery of the reprocessed gradients, the improved orbits, the new level2 products and, most important, the final rel. 6 GOCE gravity field models. The paper provides an overview of the new release of GOCE products and specifically investigates the performance of the rel. 6 GOCE gravity field models.

We analysed recent gravity field models (given in the form of spherical harmonic expansion to degree and order 2190) EIGEN 6C4 and SatGravRET14 plus the bedrock topography from the Bedmap 2 model, describing the topography of the ground under the ice of Antarctica. From the gravity models we computed the “gravity aspects”: the gravity anomalies, the Marussi tensor of the second derivatives of the disturbing potential, the gravity invariants and their specific ratio, the strike angle and the virtual deformations. We applied these results for selected parts of Antarctica; not too far from Lake Vostok (LV) we discovered at least two objects that might be subglacial volcanoes. We also studied the “lake district” between the Gamburtsev Subglacial Mountains (GSM) and LV. We found that the lake itself is divided not only to its N and S parts by a ridge, which is well known, but also to E and W parts. A tension across LV has a prevailing direction E-W. We identified at least three candidates for subglacial lakes south of the lakes “90 degree E” and “Sovetskaya” (they are not listed in any generic inventory of the lakes in Antarctica). In the gravity aspects, this area looks like a basin, where the mentioned lakes might be just a remnant of a large bowl; the orientation of this basin is the same as that of LV; the size comparable to LV; we call it “90 Degree Basin”. A subglacial connection between all the mentioned lakes is possible and documented via the gravity aspects, namely by the virtual deformations (by this independent way we confirm what is already known from other data). Finally we contributed to a longer discussion with contradictory opinions about a huge impact crater, rather a basin with a mascon in its centre, in Wilkes Land. It would be the largest impact crater found on our planet. Based on the gravity aspects and bedrock topography we support von Frese hypothesis about the impact character of this structure and its possible continuation to southern Australia.

Only 18% of the seafloor has been mapped by ships at a resolution of 1 km or better. Depths in the remaining 82% are predicted using sea surface gravity measurements derived from satellite radar altimeters. Spatially dense altimeter tracks are available from Geosat, ERS-1, Jason-1/2, CryoSAT-2 and AltiKa. We have assessed the relative importance of each altimeter data set to the short-wavelength accuracy of the gravity field. The contributions depend on range precision, the duration of the measurements, and the inclination of the orbital plane. Ranked from most to least important are CryoSAT-2, AltiKa, Jason-1/2, ERS-1, and Geosat. We assess the accuracy and resolution of the marine gravity through a comparison of a spatially dense gravity survey of the Alaminos Canyon in the Gulf of Mexico. A previous gravity model that did not include CryoSat-2, Jason-1/2 and Altika data had an accuracy and half wavelength resolution of 2.03 mGal and 9 km, respectively. This new gravity field, which includes all the latest data, has an accuracy and resolution of 1.33 mGal and 5.5 km, respectively.

This dramatic reduction in noise level of the global marine gravity revealing previously unknown seafloor structures such as small seamounts, abyssal hills, now extinct oceanic microplates, and fault networks buried beneath thick sediments along continental margins. One of the more interesting discoveries is the pervasive nature of small-offset, non-rigid discontinuities, known as tectonic discordant zones. Discordant zones are produced by transform-like offsets that migrate along the ridges, resulting in wavy symmetrical discontinuities on the ridge flanks (e.g., Matthews et al., 2011). They are analogous to propagating rifts but are concentrated along slow- to intermediate-spreading ridges. Globally, the total length of discordant zones is about one quarter that of fracture zones. However, less than 1% of discordant zones have been mapped by ships in high resolution. During the presentation we will “fly” over the remote ocean basins to highlight the new discoveries.