Battlescale Forecast Model (BFM) Target Area Wind Speed Validation Over WSMR, NM Initial Results

摘要

The Battlescale Forecast Model (BFM) was run on a 200 MHz Pentium PC using initialization and verification data collected during November and December 1974 at WSMR, NM. BFM target area artillery wind speed forecasts were verified at two locations on the northern pan of the range. For these cases, the BFM was initialized using data from up to seven southern range locations. Gridded BFM output at the two targets was convened to standard artillery computer meteorological messages consisting of ten levels of wind data from the surface to 4 km above ground level. Output from these BFM messages was compared against similar messages derived from balloon launches at the targets and treated as truth. Additionally, BFM output was compared against messages produced by the Default Met and Time Space Weighted model techniques for the same cases. Statistical validation calculations show that the BFM performed 75 percent better than the Default Met technique and 21 percent better than the Time Space Weighted model technique.