Atmospheric factors, namely, time of wetness of corroding metal panels, panel temperature, and atmospheric sulfur dioxide and atmospheric chloride content, were measured at four inland and three coastal North American test sites while corrosion data for steel, copper, and zinc were being developed. Corrosion losses experienced by panels of these metals which were exposed at different times of the year for similar periods of time showed considerable variation at all sites.
Statistical analyses show conclusively that the atmospheric factors measured completely control the rates of corrosion at all sites for at least the initial month. For longer periods of time control of the corrosion process remains with the atmospheric factors in some cases, and in others it is gradually transferred to factors related to the changing surface conditions resulting from accumulation of corrosion products and foreign agents.
At the sites where the atmospheric factors control corrosion, the empirical equations developed enable one to predict the corrosion losses of steel, copper, and zinc from a knowledge of the atmospheric factors. They also now make it possible to account for variations in observed corrosion losses experienced by panels exposed at different times of the year.