Oil pollution is a major source of environmental degradation, and requires accurate monitoring and timely detection for an effective control of its occurrence. This paper examines the potential of a remote sensing approach using the spectral and thermal responses of crops for the early detection of stress caused by oil pollution. In a glasshouse, pot-grown maize was treated with oil at sublethal and lethal applications. Thereafter, leaf thermal, spectral and physiological measurements were taken every two to three days to monitor the development of stress responses. Our results indicate that absolute leaf temperature was a poor indicator of developing stress. However, a derived thermal index () responded consistently in the early stages of physiological damage. Various spectral reflectance features were highly sensitive to oil-induced stress. A narrow-band index using wavelengths in the near-infrared and red-edge region, , was optimal for previsual detection of oil-induced stress. This index had a strong linear relationship with photosynthetic rate. This indicates that by detecting vegetation stress, thermal and hyperspectral remote sensing has considerable potential for the timely detection of oil pollution in the environment.