On the Development of an Ensemble Data Assimilation and Forecasting System for the Red Sea
With Ibrahim Hoteit, King Abdullah University of Science and Technology (KAUST)
On the Development of an Ensemble Data Assimilation and Forecasting System for the Red Sea
With a growing interest in exploiting the Red Sea resources and protecting its fragile ecosystem, there is more and more pressing demand for building an operational system to predict its circulation. This is a challenging task due to the dominant strong seasonal variability and short-living mesoscales in this basin. This talk will present our approach for building this system within an ensemble Kalman filtering (EnKF) framework, combining a (i) one-step-ahead smoothing formulation to enhance the ensembles sampling with the future observations, (ii) a hybrid formulation of the filter prior covariance for implementation with reasonable-size ensembles, and (iii) a second-order exact sampling of the observations perturbations for efficient implementation of (i) and (ii) with a stochastic EnKF. I will discuss the relevance of each of these schemes and demonstrate their performances with various applications.
- Speaker: Ibrahim Hoteit, King Abdullah University of Science and Technology (KAUST)
- Thursday 24 October 2019, 15:00–16:00
- Venue: MR 14.
- Series: Applied and Computational Analysis; organiser: Edriss S. Titi.
