An Alternative Concept for Validation of GOCE Gradiometry Results Based on Regional Gravity R. Haagmans Earth Sciences Division, ESA/ESTEC, Keplerlaan 1, 2200 AG Noordwijk, The Netherlands K. Prijatna DEOS, Delft University of Technology, Thijsseweg 11, 2629 JA Delft, The Netherlands O. Omang Department of Mapping Sciences, Agricultural University of Norway, PO Box 5034,1432 Ås, Norway Abstract. The GOCE mission (ESA, 1999) aims at providing an improved global geoid to 1-2 cm at a spatial resolution of about 100km. For this purpose measurements from a 3-axis satellite gradiometer and high-low satellite-to-satellite tracking (SST) need to be combined. The SST and gradiometry data are combined in complementary sense to derive information on the gravity field. The gradiometer capacity is limited due to inferior performance of the accelerometers outside a specific measurement bandwidth. The mission profile contains two short calibration phases preceding two measurements phases of 6 months. Once the calibration is performed, which includes the complex process for accelerometers and the gradiometer (Koop et al. 2001), there is a need to check the validity of the calibration throughout the measurements phases without direct interference with the satellite control. Here concept is proposed for checking several of the gradiometer components against ground gravity data in a band-limited sense at the Earth’s surface or at satellite altitude. The band-limits need to be tuned to keep most of the signal inside the accelerometer measurement bandwidth allowing a direct comparison with the calibrated gradiometer results only. The same concept can also be applied between various (combinations of) components of the gradiometer. The band-limited filters allow numerical integration approaches with truncated kernels, which means that several areas in the world with good gravity coverage can be repeatedly checked against the gradiometer results in a limited area yielding the same truncation (omission) error. Analysis of the repeated solutions in time and from different components will improve our inside in possible changes in calibration characteristics over time. Keywords. Satellite gravity gradiometry, gravity, GOCE validation, band limited kernels 1 Introduction The idea is to compare regional gravity and satellite gradiometer data from GOCE in a spectrally band- limited sense. This application follows the recently proposed idea by Prijatna and Haagmans (2001) for a kind of multi-resolution concept for regional geoid determination with modified kernels based on heterogeneous data. It implies the combination of a low pass filtered global geopotential model, a regional band pass filtered satellite gradiometer contribution, and a regional high pass filtered gravity contribution. In case we isolate the band pass filtered satellite gradiometer contribution and apply the same band pass filter to modify the kernel of gravity, we are able to compare and possibly validate the results directly in a limited area in terms of geoid heights or disturbing potential, or even gravity or gravity gradients (cf. Arabelos and Tscherning, 1998, Koop et al. 2001). The initial focus will be on the potential at the Earth’s surface or geoid heights, because the main interest is to analyze the effect of a possible build up of systematic errors or change in calibration characteristics in the accelerometers in the final end products. A natural extension of the approach to satellite level is possible and allows also to check the propagation of possible disturbing effects in the accelerometers to the potential from combinations of gradiometer disturbance components like T zz (or T rr ) alone, -T xx – T yy , or {T xx – T yy ,2 T xy } (see van Gelderen and Rummel, 2001). First the general global solution is introduced, followed by a focus on the medium wavelength component for gravity and gravity gradients. The next step is the change to a regional solution with an example of the consequences for the choice of filters and the relation to truncation errors for a solution at the Earth’s surface. Finally, a discussion and conclusions are presented.