To solve the problem of sea and land waveform classification for domestic airborne lidar bathymetric system, support vector machine method is used to build the model for classification of sea and land waveform. This method is based on the features of multi-channel ocean lidar waveform data, and characteristic parameters of multi-channel waveform are extracted. The validation results indicate that overall accuracy and Kappa coefficient of the classification method are 99.03% and 0.9805, respectively. The accuracy of the sea and land waveform classification of the proposed method is good enough for the engineering application. And it is suit for the waveform classification treatment of domestic airborne lidar bathymetric system. Which can lay the foundation for correcting the velocity of light in the water medium during the depth calculation procedure, tide and wave correction.