The goal of this work is to demonstrate that assimilating the output of a Mesoscale Numerical Weather Prediction (NWP) model into a Computational Fluid Dynamics (CFD) model can improve the fine scale structure that can be modeled in environmental flows. Mesoscale NWP with data assimilation has been used extensively to study mesoscale flow features. Scales finer than about 1 km, however, are not well captured by the turbulence physics of such models. CFD models have proven useful at those smaller scales, capturing the details of flow around features such as buildings and fine-scale topography. This work describes how the output of the MM5 NWP model with four dimensional data assimilation (FDDA) is used to initialize and assimilate into CFD simulations with grid spacing on the order of 50 m. Assimilating the NWP model data is critical to obtaining the spatially varying outer scale patterns. The CFD model is then able to accurately model flow around fine scale topographic features. In addition, weather tower observations are assimilated into the CFD model. The technique is demonstrated in a case study. We expect this technology to prove useful for downscaling flow features.