This paper presents results of an empirical study to estimate the measurement error in the peak wind speed at Shuttle Launch Complex 39 (LC-39) which results from the measurement being made by sensors 1,300 feet away. Quality controlled data taken at a height of 30 feet from an array of sensors at the Shuttle Landing Facility (SLF) were used to model differences of peak winds as a function of separation distance and time interval. The SLF data covered wind speeds from less than ten to more than 25 knots. Winds measured at the standard LC-39 site at the normal height of 60 feet were used to verify the applicability of the model to the LC-39 situation. The error in the peak wind speed resulting from separation of the sensor from the target site obeys a power law as a function of separation distance and varies linearly with mean wind speed. At large separation distances, the error becomes a constant fraction of the mean wind speed as the separation function reaches an asymptotic value. The asymptotic average of the mean of the absolute difference in the peak wind speed between the two locations is about twelve percent of the mean wind speed. The distribution of the normalized absolute differences is half-Gaussian. Merceret, Francis J. Kennedy Space Center NASA/TM-1999-208544, NAS 1.15:208544