Gravity field modeling on the basis of GRACE rangerate combinations: Current results and challenge

Abstract

A new approach to gravity field modeling from GRACE KBR data is presented in detail. The functional model explicitly connects gravity field parameters with a linear combination of range-rate measurements at three successive epochs. Unlike the functional model based on inter-satellite accelerations, the new one does not require the radial component of intersatellite velocity vectors to be determined from GPS data. Performance of the approach is tested with real GRACE data. A 176-day set and a one-month set are seperately processed. The mean gravity field model computed from the 176-day data set is as accurate as the GGM01S model. The computed monthly model is 2-3 times less accurate than a model obtained by GFZ. The probable reason for a limited accuraccy of the computed monthly model is noise in inter-satellite baselines, which are still determined from GPS data.