Ice accretion for in-flight icing conditions often yields complicated shapes which can significantly affect the performance of an aircraft. This has been a primary motivation for numerous studies to understand the physics of ice formation on aircraft and develop prediction methodologies. During icing experiments, many of the complex shapes found in later stages of ice accretion have been observed to their origins in ice roughnesses which form early in the ice accretion process.

The present work focuses on the formation of roughnesses due to the inherent interfacial instabilities of a wetted ice surface. Solutions are found using a multiple scales model for the stagnation region near the leading edge of a wing. In the computations for both unswept and swept wing icing, the roughness sizes are found to be about the same size as the air boundary layer thickness. These roughnesses qualitatively agree with the typical geometries and properties of the roughnesses seen in unswept and swept wing icing experiments.

Using the multiple scales model for the stagnation region as a guide, a preliminary simplified engineering linear stability analysis is developed for glaze icing. The roughness diameters predicted using this model generally agree with the typical characteristics of the ice roughnesses seen in experiments providing that the airfoil surface temperature is sufficiently close to freezing. Smooth zones devoid of roughnesses, which have a sharp demarcation between the smooth and rough regions, can be present near the stagnation line. However, the solutions of the linear stability model are found to be overly sensitive to the temperature of the airfoil, which suggests the need for further modeling to incorporate the transient changes in the airfoil skin temperature and ice thickness during the ice accretion process.