The preliminary wind survey is normally limited by wind speed measurements and registration of information at very great height. This problem is particularly important specially when one intends to characterize preliminarily the relative quality of a determined place for the utilization of wind energy. For this kind of approximation it is neither practical, fast nor economically suitable to install structures supporting sensors at heights between 50 and 100 meters (the most frequently used height ranges for actual aerogenerators) if one doesn't have prior reliable information about general eolic conditions of the local region. This work intends to show a procedure which allows preliminary information by means of height extrapolation from experimental wind speed data and average surface density of wind power obtained at two or three different heights and with a proved model of extrapolation. The values are then extrapolated up to a height of 50 and 100 meters. After applying a power law model of extrapolation the results obtained are then compared among themselves as well as with the two experimental wind stations employed. The registration period covers a complete year of data for one experimental wind station (Punta Curaumilla) and eight months for the other one (Llo-Lleo Alto). Data were registered continuously and integrated every 5 and 10 minutes, and were measured at a height of 10 and 15 meters above the ground at Punta Curaumilla and at 10 and 15 meters at Llo-Lleo Alto. This procedure admits the carrying out and the backing up of preliminary studies for wind energy surveys in a shorter period of time with a reduced cost.