Predictability and Ensemble Forecast Skill Enhancement Based on the Probability Density Function Estimation
Abstract
An innovative method for direct assimilation of Lagrangian data in meteorology and oceanography has been formulated based on the extended Kalman filter (EKF). The method has been applied to the point vortex systems. A classification of the Lagrangian trajectories has been made for practical applications of the direct Lagrangian data assimilation method to complex ocean circulation models. In order to examine the predictability of the coherent structures in the stratified ocean, a layer-model for point vortex system has been investigated. Judicious observing system, including parameter-estimation capacity, has been designed for improved estimation of the flow dynamics in the unobserved sub-surface layers. To study predictability concerning transition from oscillatory to singular behavior, an idealized, highly-nonlinear dynamical system has been developed. The model is based on interplay between restoring force and hysteria in inertia, which can be found in many physical systems. The bifurcation diagram has been investigated to enhance forecast skill for the sudden transition. A hybrid transport theory (TIME: transport induced by mean-eddy interaction) has been formulated to estimate transport induced by the anomaly in the large-scale atmosphere and ocean flows. It has been extended to identify signature and analyze role of variability in application to a large-scale atmospheric flow.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 21, 2005
- Accession Number
- ADA429618
Entities
People
- Kayo Ide
Organizations
- University of California, Los Angeles