Agricultural soils under no tillage systems are affected by soil compaction caused by machinery traffic, which can restrict root growth and reduce crop yields. The purpose of this study was to evaluate the effect of different compaction levels on the physical quality of a Red Latosol (Oxisol), sandy-loam, in the Jaboticabal county, São Paulo state, Brazil, under corn, using multivariate analysis. The experimental design was completely randomized with six compaction levels and four repetitions. Undisturbed soil samples were collected in the 0.02-0.05, 0.08-0.11 and 0.15-0.18 m layers to determine the bulk density. The following plant traits were evaluated: plant height, ear height, stem diameter, plant dry mass, root density, root diameter and root dry matter. By the clustering and principal component analysis three groups were identified of high, intermediate and low corn yield, based on variables of the soil, root system and aerial plant part. The groups were formed by the method of hierarchical clustering, k-means and principal components analysis. The principal component analysis evidenced that high corn yields are correlated with a good development of the shoot under conditions of low bulk density, which in turn favored a high fine roots biomass. The physical quality of the Red Latosol was adequate for corn growth up to a bulk density of 1.38 Mg m-3.
bulk density; root system; crop yield; cluster analysis; principal components analysis
