Determining the capability of atmospheric models to predict the temporal and spatial evolution of upper level conditions is of great interest in the design of high-altitude, long-endurance (HALE) aircraft and high altitude airships. For instance, strategies for avoiding upcoming high-speed wind storms can be made to affect the operation of unmanned aircraft systems (UAS), such as Global Hawk and Helios. In this paper, we will examine experimental runs of the Weather Research and Forecasting (WRF) model in order to evaluate the performance of these forecasts in the area of predicting high-altitude (~20 km) winds and temperatures. The configuration of the WRF model will be set to allow for fairly high resolution in both time and space, for example, sub-hourly output available at 4 km horizontal grid spacing with a model top set at 10 mb. The model will be executed for several different types of weather events, such as Hurricane Katrina, the Nor'easter of 2007, a typical warm-season convective squall line case, and a wintertime case containing a strong jet stream across the continental U.S. Output from the WRF model forecasts will be compared against available analyses and observations to in an attempt to determine if there are large biases in the forecasts and if realistic spatial structures are included. Results from this ongoing work will be presented at the conference.