Plant Phenomics / 2019 / Article / Fig 3

Research Article

Development of Optimized Phenomic Predictors for Efficient Plant Breeding Decisions Using Phenomic-Assisted Selection in Soybean

Figure 3

Analysis of hyperspectral canopy reflectance wavebands (average reflectance per 10 nm) and relationship with seed yield using 292 soybean genotypes grown in six environments (replication per environment = 2). (a) Genetic correlation () between seed yield and waveband, (b) SNP-based heritability () across waveband, and (c) feature importance for predictor variables (i.e., waveband) for SY estimation using the random forest algorithm. Hyperspectral canopy reflectance data were collected in six environments across central Iowa by recording two measurements by positioning the sensor 1 m above the canopy in the nadir position.