Dr. Shulman presented information on the existing data assimilation techniques in numerical modeling of oceanic processes. He began his talk by presenting the 3 types of data to be considered:

• High spatial resolution but low temporal resolution (satellite derived data).
• High spatial and temporal resolution over the small area (HF Doppler radar data).
• High temporal resolution of a single point (profilers modeled arrays).

He then presented information on various numerical modeling techniques. The most famous one is the Kalman Filter (KF) which provides the optimal way to optimize you model at time k+1. Another techniques is the Optimum Interpolation (OI) in which the full numerical model is used instead of the linear state transition. He proceeded with information on the Simplification and Order Reduction methods and Variational Methods. After discussing the advantages and problems, he summarized the information presented. Dr. Shulman stated the goal is to research the use of a simplified statistical method to provide a cheap estimate rather than continuing to use a variational method.

He concluded his presentation after presenting examples of data assimilation from the NRL layered ocean model. The assimilation scheme and approach was discussed in detail. He also presented information on the Identical Twin Experiments, which models data from the Sea of Japan. He completed his talk after presenting results from the experiments.