The wind speed data represented in the form of frequency curves show the shape of a potential model. The Weibull and Lognormal models are used for this purpose, with hourly mean wind speed data. This study deals with the estimation of the annual Weibull and Lognormal parameters from 20 locations in Navarre. The suitability of both distributions is judged from the R-2 coefficient with a linear regression for the Weibull distribution and a nonlinear regression for the Lognormal distribution. Both approaches give a good fit, giving better results for the Weibull distribution. A comparison between the estimation and the production for a wind farm is offered.