Multivariate analysis of physical–chemical and biological parameters along wheat growth (Triticum aestivum L.) and the effects of a phosphorus-solubilizing bioinoculant

Abstract

Two plots (10 ha each) of Triticum aestivum L. wheat crop were analysed, one with a phosphorus-solubilizing bioinoculant (bioinput) and another without the bioinput (control). Traditional physical–chemical variables and four enzymes were analysed: arylsulfatase (AS), β-glucosidase (β-G), acid phosphatase (AP), and N-acetyl-β-d-glucosaminidase (NAG). These variables were considered by Hierarchical Cluster Analysis (HCA) and partial least squares-discriminant analysis (PLS-DA). The HCA separated 100% correctly the control and bioinput samples along the wheat phenological stages (Feekes-Large scale). The PLS-DA model was constructed with seven latent variables and obtained 100% accuracy, explaining a variance of 97.05% in calibration and 93.13% in cross-validation. The control soil had a productivity of 57.8 bags per hectare, while the soil with bioinput had a productivity of 60.8 bags per hectare. Furthermore, soil with the bioinput presented higher concentrations of the enzymes AS, β-G, and AP in the final phenological stages.