Articles
Representation of Opuntia ficus-indica (L.) Miller macro-nutrient compositions
Article number
1343_38
Pages
297 – 308
Language
English
Abstract
Using standard statistical methods with nutrient ratios is known to lead to spurious, nonsensical results.
Thenceforth, it appears desirable to use techniques that allow the visualization and analysis of orthonormal nutrient balances in subspaces of plant ionomes.
These tools are available in the framework of compositional data analysis.
This work aimed to demonstrate that 1-year-old cladodes as low- and high-yielders of fruits have different N, P, and K compositions (n=228), taking into account closed concentrations and centered log-ratio (clr) and isometric log-ratio (ilr) transformations.
A covariance biplot of clr transformations of closed concentrations, that is principal components constructed in ilr coordinates, allowed two principal components to explain 100% of data variability (PC1, 83.61%; and PC2, 16.39%). This biplot suggests that the ilr [N, P|K], and the ilr [N|P] could be important nutrient orthonormal balances in this subspace of the O. ficus-indica cultivar Rojo Pelón ionome.
Moreover, the ternary diagram and the balance-dendrogram suggest low-yielders tended to load more N than P, and more N plus P than K compared to the high-yielders when the target yield was 1166.67 g cladode‑1. Markedly, low-yielders tended to show P deficiencies and excessive consumption of N. Novel techniques to perform suitable recommendations should be developed to address the needed change objectively to enhance nutrient status by improving the diagnosed P deficiency and the N excess.
Thenceforth, it appears desirable to use techniques that allow the visualization and analysis of orthonormal nutrient balances in subspaces of plant ionomes.
These tools are available in the framework of compositional data analysis.
This work aimed to demonstrate that 1-year-old cladodes as low- and high-yielders of fruits have different N, P, and K compositions (n=228), taking into account closed concentrations and centered log-ratio (clr) and isometric log-ratio (ilr) transformations.
A covariance biplot of clr transformations of closed concentrations, that is principal components constructed in ilr coordinates, allowed two principal components to explain 100% of data variability (PC1, 83.61%; and PC2, 16.39%). This biplot suggests that the ilr [N, P|K], and the ilr [N|P] could be important nutrient orthonormal balances in this subspace of the O. ficus-indica cultivar Rojo Pelón ionome.
Moreover, the ternary diagram and the balance-dendrogram suggest low-yielders tended to load more N than P, and more N plus P than K compared to the high-yielders when the target yield was 1166.67 g cladode‑1. Markedly, low-yielders tended to show P deficiencies and excessive consumption of N. Novel techniques to perform suitable recommendations should be developed to address the needed change objectively to enhance nutrient status by improving the diagnosed P deficiency and the N excess.
Publication
Authors
F. Blanco-Macías, M. Márquez-Madrid, A. Lara-Herrera, R.D. Valdez-Cepeda
Keywords
macro-nutrient balances, log-ratio transformation, N, P, K
Groups involved
Online Articles (75)