Combination of statistical Kalman filters and data assimilation for improving ocean waves analysis and forecasting
The issue of ocean wave analysis and forecasting is today of increasing importance for a variety of scientific and social-economic purposes. In this framework, a number of state-of-the-art research and operational tools have been developed, mainly based on numerical modeling and advanced statistical...
| Main Authors: | , , |
|---|---|
| Format: | Article in Journal/Newspaper |
| Language: | unknown English |
| Published: |
2012
|
| Subjects: | |
| Online Access: | https://pergamos.lib.uoa.gr/uoa/dl/object/uoadl:3028100 |
| Summary: | The issue of ocean wave analysis and forecasting is today of increasing importance for a variety of scientific and social-economic purposes. In this framework, a number of state-of-the-art research and operational tools have been developed, mainly based on numerical modeling and advanced statistical techniques. The performance of the latter is essentially dependant on the utilization of external information (remote sensing and in situ measurements). In this work, a combination of ocean wave numerical models, statistical Kalman filters and data assimilation techniques is used for improvement of simulation-accuracy. More precisely, the systematic deviations of the wave model results are minimized by the use of Kalman filtering algorithms in areas with continuous flow of observations. Then, the improved outputs are assimilated by an optimum interpolation scheme, into the forecasting period of the wave model, in order to extend the assimilation impact in time and space. The case studied concerns four one-monthly intervals in the North Atlantic Ocean. © 2012 Elsevier Ltd. |
|---|