This paper presents a study of the vertical profiles of horizontal mean winds calculated from data obtained with the University of Wisconsin volume imaging lidar (VIL) during the 1989 First International Satellite Land Surface Climatology Project (ISLSCP) Field Experiment program. Wind profiles are determined from cross correlation functions between subsequent constant altitude plan position indicator planes calculated from volume scans of the convective boundary layer. An objective technique to identify unreliable measurements is developed. An error analysis based on determining the errors due to random noise in the data is performed. The VIL wind profiles are compared with radiosonde, aircraft-based, and ground-based measurements. On the basis of internal consistency tests we estimate that in the convective boundary layer the hourly averaged VIL wind estimates, which have passed the reliability analysis, have root-mean-square errors of 0.03--0.1 ms-1 in speed and 0.4--2.0¿ in direction. ¿ American Geophysical Union 1995