Frost formation and anti-icing methods are studied in a variety of areas for decades. In refrigeration systems, a number of studies are focused on the influence of surface treatment on delaying the time of frost formation. There are far fewer studies on the effect of surface wettability on the defrost process. Some of the literature show experimental results on defrost time for different surfaces, but the results could be different based on individual experimental conditions. In this study, the influences of surface wettability on the defrost process are investigated analytically. The melting process is divided into three stages based on the behaviors of the meltwater. The water saturation and meltwater draining velocity are formulated for absorption and drainage stages separately. The comparison among the melting rate, the permeation rate, and the draining velocity determines the meltwater behavior, which influences the defrost process and defrost mechanisms. The slumping condition is a special phenomenon in the melting process and can be a potential method for frost removal. Water accumulation at the surface decreases the adherence strength of the frost column to the surface. A slumping criterion is formulated based on the analysis of interfacial forces and the body force on the frost column. The slumping condition of the model depends on the contact angle on hydrophilic surfaces and on the contact angle hysteresis on hydrophobic surfaces. Experiments of frost and defrost process on vertical surfaces with different wetting conditions are conducted on a lab-build setup. The experimental results show that defrost time and efficiency are determined by the system design, the heating methods, the heat flux applied at the surface and the surface wettability. Defrost mechanisms vary with surface wettability. During the defrost process, a frost column detaches from a superhydrophobic surface and falls off as a whole piece. While on the superhydrophilic and plain surfaces, frost melts, and water film or retention evaporates. Defrost time and efficiency are not significantly different on the tested surfaces at the point that frost melts and water film/retention droplets remain on the surfaces. However, defrost time and efficiency are improved noticeably on the superhyrophobic surface for a complete defrost process in which the water evaporation time is included. This research serves as a supplement to current studies on the effect of surface wettability in refrigeration systems. In application, the preference to the different surface wettability depends on the practical operation requirements and the actual system design.