Toward the Next Generation of Altimeter Data Assimilation for Physical Ocean and Marine Ecosystem Monitoring and Prediction
Approach and work statement. The methods that will be explored and further developed to assimilate altimeter data will be multiple, taking advantage of the most recent advances in the field of stochastic modelling, statistical estimation and optimal control. On the one hand, well established methodologies such as Kalman-type filters and smoothers and variational methods will be expanded and complexified to take into account non-Gaussian error statistics or non-linear model dynamics. On the other hand, more generic methods such as particle filters will be adapted to cope with the huge dimension of realistic ocean models.
Altimetry from the JASON suite, and from forthcoming missions (HY-2A, Sentinel-3, SARAL/Altika), will be the primary source of data for this project, in conjunction with ENVISAT, Cryosat-2 and other historical altimetry data sets, GRACE and GOCE for gravimetry, and SMOS for salinity. A particular focus will be set on the use of multiple data sources including in situ observations and to their optimal complementarity to altimetry.
Expected impact. The application of the assimilation methods will enable improved ocean real-time analyses as well as multi-year reanalyses, providing a unique source of ocean information to
Proposing team. The group of PIs involved in this proposal has played a significant role in France and Europe with regard to the development of data assimilation in oceanography through their fundamental research work, and their contribution to the emergence of increasingly efficient operational systems. The expertise of the proposing team is broad enough to cover the variety of most advanced assimilation approaches, in such a way as to stimulate open dialog and cross-fertilization of ideas between the proposal participants.