As one of the Observing System Simulation Experiments (OSSE) team members, the Forecast Systems Laboratory (FSL) has the responsibility to provide model output to scientists within FSL, Environmental Technology Laboratory (ETL), and other organizations to simulate observational data for comparison with the simulated lidar data of winds, vertical velocity and cloud fields. To fulfill this responsibility at FSL, the nonhydrostatic, numerical weather prediction model MM5 has been implemented to run with initial and boundary conditions obtained from the nature run data from the European Center for Medium-Range Weather Forecasting (ECMWF) Integrated Forecast System(IFS) for 9-15 February 1993. The MM5 domain covers most of North America with a 10-km horizontal grid spacing and 23 vertical levels topping out at 100 mb. This domain has been tailored to include the present domain of the Rapid Update Cycle (RUC-2) hydrostatic numerical weather prediction model. This configuration will allow scientists to perform assimilation experiments with simulated standard data extracted from the mesoscale MM5 nature run, with and without the simulated lidar data. This will allow assessment of the impact of lidar measured winds on regional-scale forecasts. Additional software is under development to interpolate MM5 output to spatial and temporal locations corresponding to a variety of data sources such as lidar, METARs, profilers, RAOBs, and ACARS. This paper will discuss the attributes of the nature run and the design of the observation simulators.