This dissertation has two key objectives: the first objective is to develop a method of predicting and quantifying the amount of water that can enter into a pavement system by vapor transport; the second objective is to identify to which extent the fatigue crack growth of pavement would result from such moisture accumulation. To fulfill these two objectives, a diffusion model was first established to illustrate the wetting process of the surface asphalt layer due to the vapor migration from subgrade soil into the upper layer. Secondly, in order to quantify the degree of moisture damage induced by water vapor diffusion, fine aggregate mixture specimens were fabricated and conditioned at different levels of relative humidity in closed vacuum desiccators that allows little temperature fluctuation. Moreover, the moisture conditioned specimens were tested using a newly developed repeated direct tension test method to evaluate the fatigue crack growth. The RDT test greatly reduced the stress state complexity within the specimens by evenly distributing stress over the cross section area of the cylindrical specimen. Compared to the previous torsional test, the newly proposed test protocol was more efficient in characterizing the moisture susceptibility of the asphalt mixture. A major finding in this dissertation is that the higher level of RH in as asphalt surface layer will induce significantly higher crack growth rates.
