Wind data recorded every 3 hours at Babolsar synoptic station in southern part of the Caspian Sea and also wave information measured by Neka buoy was used to investigate the wave climate in this region. For this purpose, monthly and seasonally wind roses of the region were drawn. The results reveal that most of the time the weather is calm. Besides, the dominant direction of the wind is northerly. Further, with the help of software provided in visual basic environment, the wind duration in different conditions was calculated. The results confirm that duration of winds of higher intensity is shorter.Then, Pierson-Moskowitz and JONSW AP spectral models together with the SMB method were used for wave hind casting from wind data. Based on the results, sea state is calm in most of conditions. Wave height often is less than 0.5m. The fully developed condition is rarely observed in the region. Comparing the predicted results with the actual values (Neka buoy information) revealed that the 5MB method has the least error. Error analysis also indicates that the Pierson-Moskowitz model, which is rather suitable for fully developed conditions, results in errors of more than one hundred percent. Because of using raw wind data, the JONSW AP model also results in more errors in comparison with the 5MB method. Finally, the SMB method is suggested as the most suitable wind wave predicting method in the Babolsar region.