Combining Stationary Ocean Models and Mean Dynamic Topography Data

Abstract

In this study, a new estimate for the Mean Dynamic Topography (MDT) and its error description is analysed in terms of its impact on the performance of ocean models. For the first time, a full MDT error covariance matrix is available whose inverse can readily be used as weighting matrix in the optimization. Two different steady-state inverse ocean models are analysed in terms of their response to the new MDT data set. The output of each of these ocean models in turn provides a combined satellite-ocean model MDT. This study investigates whether the inverse ocean models benefit from the new MDT data set and its error covariance. It is examined whether oceanographic features such as the ocean current structure, the overturning circulation and heat transports are improved by the assimilated MDT data set. Special focus is given to the MDT error covariance estimate as it is crucial in the optimization.