Soil color for the identification of areas with different yield potential and coffee quality

Abstract

The objective of this work was to verify soil color effectiveness for the identification of areas with different yield potentials and coffee quality, in an Ultisol developed from sandstone, in Planalto Ocidental Paulista, in the state of Sao Paulo, Brazil. One hundred seventy-three samples were collected from 39 ha, for the determination of hue, value, chroma, and the soil redness index, based on data of diffuse reflectance spectroscopy (DRS). Productivity and quality of coffee fruit were also evaluated, and the leaf diagnosis index (DRIS) was calculated. The highest spatial correlation, observed in the cross variogram, occurred between hue and DRIS in points up to 497 m apart. The lowest spatial correlation was observed for chroma and production, in points spaced by up to 207 m. The areas with the highest yield potential (20 to 33 bags per hectare) showed hue between 6.99 and 7.06, croma between 5.00 and 5.08, and value between 4.40 and 4.45. The areas with the lowest productivity potential (3 to 7 bags per hectare) showed hue between 7.27 and 7.68, chroma between 5.20 and 5.28, and value between 4.51 and 4.67. In these places, clay content was of 16 g kg(-1), P levels varied between 65 and 75 mg dm(-3), and the sum of bases was between 56 and 58 mmol(c) dm(-3). The use of the color determined by DRS is effective to identify areas with different productive potentials and coffee quality, with 61 to 97% accuracy.