The key importance of data assimilation to oceanography and in particular to the development of operational oceanography is now well recognized. A wide range of ocean research and applications depends upon the prompt and dependable availability of ocean information products. The field of physical oceanography has matured to a point where it is now conceivable to combine numerical models and observations via data assimilation to provide ocean analysis and prediction products on various spatial and time scales. Several operational or research data assimilation systems exist in oceanography, mostly based on relatively simple sequential estimation methods. The aim of this project is to go a step further and to develop and implement advanced data assimilation techniques based on four-dimensional variational methods in a state-of-the-art modelling framework of the ocean that includes configurations ranging from low-resolution global scale to high-resolution basin scale. To achieve these goals, we need to improve and extend an existing variational data assimilation system, to develop new, innovative numerical algorithms, and to demonstrate the feasibility of variational assimilation in configurations where it has not yet been proven. The project builds on the NEMO (Nucleus for European Modelling of the Ocean) framework and the NEMOVAR (NEMO VARiational data assimilation system) initiative, both having been already adopted by the operational community. It also provides a valuable research platform for the scientific community. This project leads to improved techniques for ocean forecasting. The societal benefit will be an increased knowledge of the marine environment, ocean climate, predictive skill for societal, industrial, and commercial benefit, tactical and strategic advantage, as well as the provision of a comprehensive and integrated approach to the oceans.