Ensemble based prediction of water levels and residual currents in Singapore regional waters for operational forecasting

• Highly complex hydrodynamic phenomenon including sea level anomalies investigated.
• A twin experiment was designed to model external storm surge (non-tidal) events.
• Data assimilation schemes based on EnKF and EnSSKF are implemented.
• Simulations were performed in both hindcast and forecast mode for major surge event.
• Comprehensive monitoring network is designed to forecast water levels and currents.

Singapore Strait located between the South China Sea and Andaman Sea is driven by tides coming from both sides and the hydrodynamics in this area is complex. From the viewpoint of long term forecasting, however, models developed for this area suffer from limitations introduced by parametric uncertainty, absence of data for appropriate specification of forcing and lateral boundary conditions. For improving the model forecasts, a data assimilation technique based on ensemble Kalman filter is implemented and applied. Based on the latter, an ensemble based steady state Kalman filter is derived to address the computational limitation for daily operational forecasting. Via a twin experiment on a simulation period that includes a significant storm surge event (sea level anomaly) the skills of both data assimilation schemes are assessed and compared.