Ensemble prediction and intercomparison analysis of GRACE time- variable gravity fieldmodels

Precise measurements of the Earth's time-varying gravitational field from the NASA/Deutsches Zentrum fur Luft- und Raumfahrt Gravity Recovery and Climate Experiment (GRACE) mission allow unprecedented tracking of the transport of mass across and underneath the surface of the Earth and give insight into secular, seasonal, and subseasonal variations in the global water supply. Several groups produce these estimates, and while the various gravity fields are similar, differences in processing strategies and tuning parameters result in solutions with regionally specific variations and error patterns. This study examined the spatial, temporal, and spectral variations between the different gravity field products and developed an ensemble gravity field solution from the products of four such analysis centers. The solutions were found to lie within a certain analysis scatter regardless of the local relative water height variation, and the ensemble model is clearly seen to reduce the noise in the gravity field solutions within the available scatter of the solutions. Key Points Ensemble GRACE gravity model reduces error by 5-10mm RMS in total water storage Variations between models on the order of noise in the data Ensemble improves w/in analysis scatter, constrained by rel. model variability