Analysis of an experimental integrated model for Nowcasting in complex terrain within a coastal region

Abstract - Real-time monitoring of fast changing weather conditions and providing short term forecasts or nowcasts with sufficient accuracy are big challenges in complex terrain within a coastal region. A system has been developed that can evaluate numerical weather prediction (NWP) model performance in real time, weight several NWP models based on their past performance and continuously generate integrated forecasts by blending bias corrections and weighted forecasts from the NWP models. The Canadian GEM (Global Environmental Multi-scale) regional and LAM (Limited Area Modeling) model predicted parameters were used for generating the integrated model. Seven major forecast parameters consisting of temperature, relative humidity, wind speed, wind direction, visibility, ceiling and precipitation rate were used as predictors for the development and application of the system. This system was used in the Science of Nowcasting Olympic Weather for Vancouver 2010 (SNOW-V10) project. Comparison of the system's forecasts with actual observation data during the Vancouver Olympics shows that the integrated model enhances the accuracy of nowcasts and is the most frequently selected accurate method for the short-term forecasts.