The U.S.Army Research Laboratory has developed a mesoscale model; the Battlescale Forecast Model (BFM), which is a hydrostatic model that employs a horizontal resolution of 10 km, with 16 terrain-following vertical levels to a top of 7000 m above the highest elevation on the model domain. Many meteorological variables have been derived in the post-processed software package of the BFM. These parameters include such forecast elements as turbulence, icing, clouds, surface visibility, fog, thunderstorm probability, and precipitation parameters. Since ARL receives 15-km MM5 output from the U.S, Air Force Weather Agency, it was determined to utilize the post-processing routines from the BFM on those MM5 data to investigate the results of the two different modeling systems. The study was conducted during the cold seasons from 1999-2002 and emphasized several variables in a wide range of meteorological conditions. Results of this project show that model biases play a significant role in the output of weather hazards and also give insight into the physics of the respective model. This project and the results will prove that common software routines are not ideal for weather hazards and there should be an effort to develop software based on the individual trends of each model. Additionally, the article will relate the tie between model physics and the weather products being developed by ARL using the BFM and MM5.