摘要
This paper discusses some of the differences between online and offline approaches for both air quality forecasting and numerical weather prediction, and argues in favor of an eventual migration to integrated modeling systems that allow two-way interactions of physical and chemical processes. Recent studies are used that directly compared online and offline simulations to discuss possible shortcomings for both air quality and weather forecasting. The disadvantages of offline approaches are easy to show for air quality forecasting. On the other hand, a positive impact on short to medium range weather forecasts that is significant enough to justify an implementation at operational weather forecasting centers is more difficult to prove, and may initially only come through an improvement of the meteorological data assimilation. Eventually though, a migration to an integrated modeling system will provide new opportunities for weather prediction modelers as well. The simulation of chemical species will allow identification of shortcomings in currently used forecast models as well as lead to better use of meteorological data assimilation.