Over the past decade, the accuracy of forecasting models has improved quite drastically. This increase in accuracy is widely due to the greater access of data via satellites, upper air observations, and better representation of various processes in the predictive models. This is particularly true in studying tropical cyclogenesis and hurricane evolution. However, in the process of forecasting, it is important to know the strengths and weaknesses of the tool that is used for atmospheric predictions. During the summer of 2006, a field intensive observation program, the African Monsoon Multidisciplinary Analysis (AMMA) was conducted in West Africa. In conjunction with AMMA, the National Center of Environmental Prediction (NCEP), and the NOAA Educational Partnership Program (NOAA/EPP), global and mesoscale model real-time forecasts were undertaken. The focus of this presentation is on the case studies of the 2006 Atlantic Hurricane Season using the WRF-NMM (Weather Research Forecast – Nonhydrostatic Mesoscale Model) version 2 run twice daily at 00Z and 12Z for 48 hour forecast from June 1 - October 1, 2006. There are two domains in the WRF-NMM model covering the continental West Coast of Africa and the Eastern Tropical Atlantic Ocean; the outer parent domain is run at a 12km resolution and a one-way nested inner domain is run at 4km resolution. This case study will evaluate the following parameters: 10m wind field, 850mb geopotential height, and precipitation. There are five storms from the 2006 season that this study will give particular attention to: Tropical Storm Debby, Hurricanes Ernesto, Florence, Gordon, and Helene.