Water molecules on the surface of an air-water interface have preferred orientations that lower their energy. This implies that these molecules are not uniformly distributed in orientation space, implying that the entropy is somewhat reduced when molecules are on the surface.
How much of water's surface tension is entropic? IE, the surface tension is the free energy change per unit area of surface created. How important is the $TS$ term compared to the $U$ term in this free energy? Can this be extracted from the surface tension as a function of temperature?