Optimal Spectral Decomposition (OSD) for Remotely Sensed Ocean Data Assimilation

Abstract

Assimilation of remotely sensed ocean data (velocity, temperature, and salinity) into numerical model is of great importance in oceanic and climatic research. However, the data should be reconstructed (onto grids) before assimilation since the original datasets are usually noisy and sparse. This paper describes a recently developed optimal spectral decomposition (OSD) method for mapping and noise filtration with examples of reconstructing the data from the Argo profiling and trajectories, Ocean Surface Current Analyses - Real time (OSCAR), shore-based high-frequency (HF) Doppler radar (CODAR) and Global Temperature-Salinity Profile Program (GTSPP).