The Global Ocean Data Assimilation Experiment

This paper describes the Global Ocean Data Assimilation Experiment, its primary objectives and applications, and the needed observational and modeling components. The Global Ocean Data Assimilation Experiment is envisaged as a one-time, major effort to take oceanography from a predominantly research oriented field to one where operational activities are undertaken routinely, with useful and practical outcomes. It is a practical demonstration of our ability to deliver timely, useful ocean products, derived from a global ocean data set, and assimilated into skilful models so that greatest benefit can be extracted from the gathered information. The Global Ocean Data Assimilation Experiment is founded on the belief that such a demonstration is vital if we are to ever realize a permanent, global ocean observing network and prediction system, with all components functional and operating on a global domain. There is particular emphasis on the integration of the remote and direct data streams, reasoning that each relies on the other to realize its full potential. The Global Ocean Data Assimilation Experiment embodies a range of processes and applications, drawing power from the fact that they all require ocean data and models, and that there are important commonalities among the components that can be exploited for cost-effectiveness. Some of the severe scientific and technical challenges will be discussed, as will the strategy by which we believe the Global Ocean Data Assimilation Experiment will be realized. The schedule includes feasibility studies and scoping during 1998-1999, a pre-operational phase 2000-2002, and an intensive "experimental" phase during 2003-2005. The lead-time is very short for such an undertaking, but it is a challenge that the oceanographic and remote sensing communities must meet. Several current projects will be described. (C) 2000 COSPAR. Published by Elsevier Science Ltd.