FARMY128.99: Regional Atmospheric Conditions

• Main Author: Essenwanger, Oskar
• Other Authors: ALABAMA UNIV IN HUNTSVILLE
• Format: Text
• Language: English
• Published: 1999
• Subjects: Meteorology; *WEATHER FORECASTING; *WIND DIRECTION; *WIND VELOCITY; BOUNDARY LAYER; METEOROLOGICAL DATA; REGRESSION ANALYSIS; GERMANY; PILOT STUDIES; BERLIN; FOURIER ANALYSIS; YUGOSLAVIA; GREENLAND; Greenland; Thule
• Online Access: http://www.dtic.mil/docs/citations/ADA368554; http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADA368554

The scope of work required 4 tasks to be performed. This report is divided into 3 different research tasks as required by the scope of work. The three different tasks are briefly described in the introduction. The first task was an establishment of wind speed profiles for specified probabilities of exceedance. Four representative stations in the former area of Yugoslavia were selected and the profiles of the wind speed exceeding certain specified probabilities were given. For comparison the profiles for Berlin and Germany were included. We learn that the wind speed profiles of the 5 stations are similar except for the mean wind direction profiles. The second task required a study of whether the wind speed profile for 13-25 km could be predicted given the wind speed from surface to 12 km. Instead of a 26 by 26 matrix system, a 4 coefficient model was developed based on former work utilizing Fourier components which were also used for the probability profiles. This pilot study for Berlin and Thule resulted in an accounted variance of 70-74 %. Although this may be considered low in the first instance, however, the left error for the wind speed is lower than the measuring error, assumed to be 5 m/sec. The last investigation was a development of a multiple regression system for the prediction of veering or backing of the wind in the boundary layer. The study was performed for Berlin and Thule. Unfortunately the result using a matrix to calculate the coefficients contained very low correlations and was not successful at first. Thus an indirect approach was used, calculating the wind direction for 300 and 600 meters with an evaluation of the observed wind direction and the analytical direction by stating whether this was veering or backing. This system resulted in 60 to 80% correct answers. Again, Berlin was better than Thule.