Yield Adjustment Using GPR-Derived Spatial Covariance Structure in Cassava Field: A Preliminary Investigation

Many processes concerning below-ground plant performance are not fully understood, such as spatial and temporal dynamics their relation to environmental factors. Accounting for these patterns is very important they may be used adjust the estimation of cassava fresh root yield masked by field heterogeneity. The an characteristic that every breeder seeks maintain in germplasm. Ground-Penetrating Radar (GPR) has proven effective tool studying characteristics developing plants, but it yet been explored with respect its utility normalizing heterogeneity agricultural experiments. In this study, use GPR purpose was evaluated a trial conducted Momil, Colombia. Using signal amplitude radargram from each plot, we constructed plot error structure using variance developed GPR-based autoregressive (AR) models adjustment. comparison based on average standard (SE) Best Linear Unbiased Estimator (BLUE) through majority voting (MV) SE genotype across models. Our results show AR model outperformed other models, yielding 9.57 MV score 88.33%, while AR1 × IID had SEs 10.15 10.56% scores 17.37 0.00%, respectively. suggest can serve dual non-destructive normalization global tuber crop programs, presenting great potential adoption many applications.