It is of great importance to investigate the mass transfer mechanism across the air-sea interface in order to improve the reliability of predictions for global warming. However, previous sub-models used in general circulation model for predicting mass transfer velocity across the air-sea interface have been based on the simple assumption that the transfer velocity is proportional to wind velocity over the ocean surface. This rough assumption reduces the reliability of the sub-models. To develop a reliable model for the mass transfer velocity, some factors that truly reflect the physical processes involved should be investigated. We experimentally investigated the effects of ocean surface physical processes on the mass transfer between the atmosphere and the ocean, including phenomena such as wave breaking, swells, density stratifications, surface contamination and rain. The results show that the conventional model of the mass transfer velocity has a lot of shortcomings.