Dual Field of View Spectrometer System for Characterizing Fusarium virguliforme Infection and Yield in Soybean.

Sudden death syndrome (SDS) is a disease of soybean caused by the soil-borne fungus Fusarium virguliforme (f.v.) and is among one of the top five most damaging pathogens in the Unites States today. The ability to spectrally characterize SDS infection and predict soybean seed yield can support breeding efforts as well as commercial production. The current study was aimed at spectral assessment of soybean yield by a tractor mounted Ocean Optics dual field of view spectrometer system (Piccolo) from an inoculated f.v. soybean experiment in Arlington, WI in 2016. The Piccolo collected near-simultaneous down and up welling light from which a target relative reflectance spectrum was calculated. The spectral data were collected in the range of 400-1600 nm from soybean canopy along the vegetative and reproductive (R) stages. The data were analyzed by partial least squares discriminant analyses (PLSDA) to classify infected and non-infected soybean plots. The model including plants in R1, R3 and R4 developmental stages resulted in the highest total accuracy calibration, cross-validation and validation values of 0.88, 0.79 and 0.83, respectively. The partial least squares regression (PLSR) model that included plants in both the R5 to R6 developmental stages resulted in the highest R² calibration, cross-validation and validation values of 0.71, 0.59 and 0.62 respectively, and the lowest root mean square error (RMSE) calibration, cross-validation and validation values of 0.35, 0.41 and 0.31 t ha^-1, respectively. Therefore, it was concluded that characterizing SDS infection as well as predicting yield with a piccolo system are possible.