Improving Data Assimilation by Surfing the Waves

The use of models and data, via data assimilation, is one ofthe strategies pursued to improve climate and weather predictions and retrodictions. In these application areas the norm is that thenumber of degrees of freedom in the models is vastly larger than the data available. This is notoriously problematic in the oceans where data gathering is challenging and the dyna- mics have statio-temporal scales that span large ranges.

The Dynamic Likelihood lter is a data assimilation scheme that isdesigned speci cally for hyperbolic and advection-dominated problems. It aims to improve predictions that combine data and model outcomes in a Bayesian framework by extending the application of the like- lihood over longer spatio-temporal ltering frames. When the data has low uncertainty and the data is sparse, the methodology is competitive with other ltering methodswith regard to computational complexity, and superior in its estimates.

The talk will describe the lter and show how it performs on simple wave problems. It will then be compared to other methods on problems where it is clear what the optimal outcome should be, demonstrating its e ciency and e cacy.