The use of landforms to predict the variability of soil and orange attributes

Relief may be considered an integrating factor that expresses the interaction of various soil and plant attributes. This work aimed to analyze the potential use of landforms to predict the variability of soil and orange attributes. The study area is located in the state of São Paulo, Brazil. The following soil attributes were analyzed: clay content, organic matter content, water content, aggregate stability, macropores, micropores, total pore volume, saturated soil hydraulic conductivity, soil density, and soil resistance to penetration at 0.00–0.20-m depth. The orange attributes analyzed are total soluble solids, total titratable acidity, ratio, production, concentrated juice yield, and fruit size, which were performed in three periods (July, August and September). The soil and fruit data were submitted to descriptive statistical, geostatistical, and canonical correlation (CCA) analyses. The mean soil and fruit attributes were significantly different for the landforms by Tukey's test at 5% probability. The analysis of the geostatistical results showed that the spatial variability of the soil and fruit attributes is influenced by landforms. We point out that the temporal variability of fruit attributes is also influenced by landforms, resulting in different ripening gradients for each of the relief compartments. The first canonical pair explained 77.00% of the attribute variance. The landforms were shown to be efficient in mapping the variability of the soil and orange attributes and contributed to the understanding of the soil–plant system.