How do uncertainties in hurricane model forecasts affect storm surge predictions in a semi-enclosed bay?

With the increasing demand for accurate storm surge predictions in coastal regions, there is an urgent need to understand and quantify how the predictive skill of the hydrodynamic model is affected by various uncertainties in the atmospheric model forecasts for hurricanes. In this study, a series of numerical sensitivity experiments is conducted for the storm surge in a semi-enclosed Chesapeake Bay generated during the passage of Hurricane Isabel (2003). It is found that the predicted storm surge is sensitive to errors in predicting the hurricane's track, intensity and propagation speed. The surge height is more sensitive to the wind forcing in the upper Bay than in the lower Bay due to different response mechanisms in the two regions. Errors in the translation speed change the prediction of not only the surge height, but also its arriving time and duration of high water, whereas errors in the hurricane track and intensity mainly affect the model prediction on the surge height. In addition, the storm surge prediction is more sensitive to the spatial than temporal resolution of the predicted horizontal wind field. (C) 2010 Elsevier Ltd. All rights reserved.