Kayo Ide, University of Maryland
Multi-scale 4D Local Ensemble Transform Kalman Filter for Coastal Ocean
Abstract: Estimation and prediction of coastal ocean dynamics remains challenging because it involves complex physical processes spanning a wide range of spatial and temporal scales, while observing network systems are usually sparse and inhomogeneous. Recent development in data assimilation algorithms has significantly advanced our ability by addressing these challenges. Our approach based on the local ensemble transform Kalman filter (LETKF) uses the algorithms for dynamic constraints and multi-scale data assimilation schemes originally implemented in the 3DVar framework for the California Coastal Ocean (Li et al, 2008, 2014,2015) with multi-resolution observing network systems. Using observing system simulation experiments, we examine the impact of these algorithms in the LETKF, particularly on the mixed layer depths and vertical propagation of information by surface observations.