Abstract A mathematical model is presented for oxygen flow, heat transfer, and yeast growth at the exposed face of bunker silos. The model considers the diffusion of oxygen and conduction of heat within the silage mass; volume flow is neglected. A single comparison of yeast counts and temperatures with measurements by previous authors shows areas of good agreement, but much more experimental work is needed to determine the conditions in which diffusion is the primary mechanism of oxygen infiltration. The model predicts a rapid influx of oxygen early in aerobic exposure followed by a decrease in oxygen levels deep within the silage. Higher silage densities and a faster removal rate from the exposed face are predicted to mitigate aerobic deterioration substantially, while treatment with propionic acid in the outermost layer still results in extensive deterioration under the treated layer.