Neural networks as routine for error updating of numerical models

This paper describes a somewhat alternative approach to combining observati ons and numerical model results in order to produce a more accurate forecas t routine. The approach utilizes artificial neural networks to analyze and forecast the errors created by numerical models. The resulting hybrid model provides very good forecast skills that can be extended over a forecasting horizon of considerable length. The method has been developed for the purp ose of operational forecasting of current speeds in the Danish empty setres und Strait. The forecast system was used as a planning tool during the cons truction of the 16 km-long fixed link across the empty setresund Strait, li nking the countries of Denmark and Sweden.