WaterTalk: Storm surge forecasting using data assimilation

Thursday, October 5, 2017 2:30 pm - 3:30 pm EDT (GMT -04:00)

As part of the Water Institute's WaterTalks lecture series, Arnold Heemink, professor of Applied Mathematics at Delft University of Technology, is presenting "Storm surge forecasting using data assimilation."

Register today.

More information

Data assimilation methods can be used to combine the results of a large scale numerical tidal model with the measurement information available in order to obtain accurate forecasts. Many data assimilation schemes are based on the well-known Kalman filtering algorithm. The last 20 years a number of ensemble based algorithms have been proposed.

Variational data assimilation or "the adjoint method" can also be used for data assimilation. This approach is especially attractive for model calibration problems. Variational data assimilation, however, requires the implementation of the so-called adjoint model. Therefore also ensemble approaches to variational data assimilation have been proposed that does not require the implementation of the adjoint model.

In the presentation, the various ensemble approaches for solving data assimilation problems will be summarized. The characteristics and performance of the methods will be illustrated with real life data assimilation applications: 

  • Operation storm surge forecasting in the Netherlands using ensemble Kalman filtering
  • Calibration of the tidal model used for storm surge forecasting

Speaker bio

Arnold Heemink
After graduating from the University of Twente in 1980 Arnold Heemink worked for 13 years at Rijkswaterstaat, the governmental hydraulic institute of the Netherlands. Since 1993 he has been a full professor of Applied Mathematics at Delft University of Technology. For more then 35 years he worked on tidal modelling with a focus on the development of data assimilation methods for storm surge forecasting. More recently he also became interested in stochastic modelling and uncertainty quantification.