The early detection of, and discrimination between, the causes of different plant stress symptoms is of paramount importance for enhancing crop yield and productivity. This paper examines the potential of thermal and spectral responses of crops for pre-visual stress detection, and additionally, in the discrimination between stresses induced by oil pollution and water deficit. Maize and bean were treated with oil, a water deficit and a combination of the two. Leaf physiological, thermal and spectral responses were measured prior and after treatment and responses followed through to leaf senescence. While changes in absolute leaf temperature were not significant for oil-induced stress in maize, a thermal index (IG) showed significant responses to water deficit. Indices such as IG, red-edge position (REP) and a normalized difference ratio combining wavebands in the red-edge and near infrared were useful in the early detection of both stresses. However, while the last two indices were not efficient in the early detection of stress in bean, IG and absolute temperature showed strong potential. The results indicate that the shape of the first derivative in the red-edge region could be used to discriminate between crop responses to different causes of stress and this will be explored in future work. We conclude that a combination of thermal and hyper spectral remote sensing can enhance the early detection and discrimination of oil and water deficit as different causes of stress in crops.