AbstractA better understanding of corrosion processes during the last years is leading to an increased demand for numerical corrosion models. Corrosion models become increasingly relevant toward simulations, lifetime predictions, and the optimization of corrosion prevention. Also, the improvements in computational power and capacity give rise to an increased number of research projects in this field. Totally different modeling approaches are applied to a wide range of different corrosion processes. This leads to a dispersion of information in the literature. The aim of this paper is to give an overview of the present approaches in atmospheric corrosion modeling. In the past, atmospheric corrosion has been modeled based on empirical, historical data. These models provide fitted functions of the corrosion rate or damage as a function of time for different environmental parameters. These methods give no or little information about the underlying physicochemical phenomena that determine the corrosion processes. During the last decade, more and more work has been done on causal approaches. Therefore, the focus will be on these recent advances in atmospheric corrosion modeling. The links will be made to the microscopic models of isolated corrosion phenomena. The examples are crevice corrosion, corrosion under porous layers, and the modeling of local electrochemical methods. Also, some more general, macroscopic approaches will be discussed. These macroscopic approaches will be compared to each other and positioned into the scope of multiscale modeling. The first goal of atmospheric corrosion modeling is to provide tools in the understanding and quantification of the processes under the influence of external atmospheric conditions. These models could reduce the current corrosion assessment methods that essentially rely on empirical models. On a longer term, modeling would aid in optimizing the material selection, structural design, and maintenance management.