Discrimination of plant stress caused by oil pollution and waterlogging using hyperspectral and thermal remote sensing

Remote sensing of plant stress holds promise for detecting environmental pollution by oil. However, in oil-rich delta regions, waterlogging is a frequent source of plant stress that has similar physiological effects to oil pollution. This study investigated the capabilities of remote sensing for discriminating between these two sources of plant stress. Bean plants were subjected to oil pollution, waterlogging, and combined oil and waterlogging treatments. Canopy physiological, hyperspectral, and thermal measurements were taken every two to three days after treatment to follow the stress responses. For plants treated with oil, spectral and thermal responses were evident six days before symptoms could be observed visually. In waterlogged plants, only spectral responses were observed, but these were present up to eight days before visual symptoms. A narrowband reflectance ratio was efficient in detecting stress caused by oil and waterlogging. Canopy temperature and a thermal index were good indicators of oil and combined oil and waterlogging stress, but insensitive to waterlogging alone. Hence, this study provides evidence that combined hyperspectral and thermal remote sensing of vegetation has potential for monitoring oil pollution in environments that are also subjected to waterlogging. (C) 2013 Society of Photo-Optical Instrumentation Engineers (SPIE)