The Battlescale Forecast Model (BFM) was developed by the U.S. Army Research Laboratory for operational short-range (12.h) forecast The BFM is composed of two major programs. The first program called 3dobj, creates initial and boundary values for a forecast model by analyzing the U.S. Air Force Global Spectral Model (GSM) forecast field output data and upper air sounding data if it is available. The second program is a forecast model adapted from the Higher order Turbulence Model for Atmospheric Circulation developed by Yamada. From a case study of wind field validation, it became clear that wind fields interpolated from OSM data deviated significantly from those observed because of coarse grid spacing and lack of boundary layer physics of the GSM. Thus, it is concluded that incorporation of observed surface wind data into the initial field is important. The performance of the BFM is evaluated using meteorological data observed at White Sands Missile Range, NM. Three different initialization methods are examined to identify optimal model initialization methods. Statistical parameters such as mean residual fields of surface wind and temperature are derived from comparisons of corresponding observations and 25 12-h forecast calculations. Results indicate that incorporation of surface wind observation data into the initial field is essential to produce good, short-range BFM forecasts. (MM).