For the advancement of marker assisted selection and improving the viability of rapid field phenotyping, more efficient tools are needed to estimate grain yield potential and plant physiological status. Currently screening large populations over multiple environments is expensive and time consuming but it is the only way to increase the heritability of more complex traits. The expense is due in part to the inefficiency of combine harvesting and the need to measure and record phenotypic data manually. One potential tool for improving the efficiency of field phenotyping is canopy spectral reflectance.

The objectives of this research were to test whether canopy spectral reflectance can be used to remotely estimate potential grain yield, growth stage, and plant health.

Spectral and physiological data where obtained on 301 and 256 genotypes of winter and spring wheat, respectively, over the 2012 and 2013 growing seasons in Manhattan, KS. The experiment was an Augmented Row-Column design with two replications and two checks as irrigated or rain fed. Reflectance was measured with an ASD Inc FieldSpec HandHeld 2 Pro with single nanometer bands from 325-1075 nm. Over both seasons spectra were recorded multiple times over the winter and spring wheat life cycle as well as SPAD, relative water content, chlorophyll content, chlorophyll fluorescence (both light and dark adapted), leaf nitrogen, growth stages, canopy temperature, grain yield, and yield components. Correlations between spectra and physiological measurement where conducted using SAS 9.1.3 to isolate spectral regions of importance and to reduce dimensionality of the spectral data. Data were compared using hyperspectral and multispectral bands or regions and used to develop models to predict grain yield, growth stage, and plant health. Results of these correlations and models will be discussed.