Visual and radiometric assessments for yield losses caused by ray blight in pyrethrum

The potential of remote sensing to nondestructive measure relationships between ray blight disease (caused by Phoma ligulicola), plant measurements and components of pyrethrum [Tanacetum cinerariifolium (Trevir.) Sch. Bip.] biomass and yield using a hand-held multispectral radiometer was examined. A range of disease intensities were generated using fungicides in three fields over two years. Nondestructive assessments were obtained by measuring the percentage of sunlight reflected from canopies with a radiometer equipped with five wavelength bands. Combinations of wavelength ratios and four vegetation indices were calculated. Relationships between reflectance and biomass were investigated by removing foliage from the canopy and periodically measuring reflectance. Measurements such as stem height and the number of flowers in October consistently had significant linear relationships with relative pyrethrin and flower yield. The best predictors of relative flower and pyrethrin yield were found using either percentage reflectance in the near infrared (830 nm) and the difference vegetative index (DVI). Sev eral measures had significant linear relationships with fresh weight of foliage, including the near infrared bandwidth and the DVI, which explained 95 to 97% of the biomass variation. This study demonstrated that plant measurements and disease intensity are strongly related to pyrethrin yield, and that remote sensing has great potential to nondestructively obtain preharvest yield and biomass estimates.