Detection of crop herbicide injury through plant hyperspectral remote sensing of chlorophyll fluorescence

Crop injury from off-target herbicide application is of primary concern to farmers. Timely determination of crop injury is critical for farmers to take appropriate actions for remediation before it is too late. Traditionally, crop injury was determined by field sampling and measuring plant biological responses, making it difficult to provide timely information. We conducted studies to evaluate optical remote sensing techniques to detect the onset of herbicide crop injury. The techniques included measuring plant hyperspectral reflectance, PROSPECT model-based biological parameter sensitivity analysis, and biological parameter inversion, and measuring plant chlorophyll florescence. In our studies, we focused on the effect of glyphosate, a non-selective, systemic herbicide, highly toxic to sensitive plant species. It has been the most widely used herbicide and its use has seen a significant increase due to the increased utilization of genetically modified glyphosate-resistant crops in the last two decades. The data analysis indicated that the results using different techniques are varied but they basically outperform the method of vegetation indices directly derived from the remotely sensed data.