Early detection of potato leafhopper and cabbage aphid damage to crops using spectral reflectance

The cost and time associated with traditional pest scouting hamper the effective implementation of integrated pest management. Early detection of pest infestations is the key to minimizing their damage and reducing the over-application of pesticides. The goal of this project was to detect changes in leaf spectral reflectance patterns induced by potato leafhopper (Empoasca fabae (Harris)) and cabbage aphid (Brevicoryne brassicae L.) feeding. This information is key in order to develop remote-sensing monitoring tools in the form of unmanned aerial systems (UAS) that will save time and labor.
Analysis of spectral indices showed that normalized red-edge index (NDRE), photochemical reflectance index (PRI), and water balance index (WABI) are comparatively responsive to early stages of potato leafhopper attack. In the case of cabbage aphid, normalized red-edge index, narrow-band NDVI, and photochemical reflectance index are responsive to early infestation levels. Overall, these preliminary results show that NDRE could be a proxy to be utilized in UAS-based field scouting efforts. Other candidate indices such as leafhopper index-2 (LH-2) and WABI appeared sensitive to detection of potato leafhopper feeding. However, off-the-shelf, affordable (< $10,000) UAS sensors do not provide a continuous spectrum from visible to SWIR region and this limits the use of these narrow band indices. Further research on leaf spectral reflectance derived under multiple stresses – biotic and abiotic- will refine our understanding of vegetation indices for use with UAS-based reflectance data.