In this work, a machine learning approach has been used to predict ammonia volatilization in Mediterranean cropping systems based on 234 experimental trial treatments. The “random forest” model had good predictive capacity. Fertilization dose was the main predictor. Unlike in the case of N2O, Mediterranean conditions do not have a particular effect on ammonia emissions. However, there is an indirect effect since Mediterranean systems, mainly rainfed ones, have fertilization doses lower than in temperate systems.