The local management of crop areas, which is the basic concept of precision agriculture, is essential for increasing crop yield. In this context, diffuse reflectance spectroscopy (DRS) and digital elevation modelling (DEM) appears as an important technique for determining soil properties, on an adequate scale to agricultural management, enabling faster and less costly evaluations in soil studies. The objective of this work was to evaluate the use of DRS together with topographic parameters for quantification of soil physical and chemical properties and its relationship with the precision agriculture. For that, 34 soil samples were collected at 0-0.20 m depth, in an area of 100 ha, belonging to the Santa Fé mill, in Tabatinga, State of São Paulo, Brazil. After the soil collection, the samples were dried in a forced-air oven at 45 °C for a period of 24 hours, sieved through a 2 mm mesh, and sent to a soil-testing laboratory. Soil spectra were measured using a commercially available spectrophotometer FieldSpec 4, (Analytical Spectral Devices, Inc., ASD, Boulder, Colorado, USA) in the range of 350 2500 nm (Vis - NIR -SWIR), with three replicates for each sample. The topographic data were obtained from the DEM. Then, using radiometric information, regression models were generated by partial least squares (PLSR) to estimate the soil attributes. DRS shows a good correlation with copper, clay and H+Al (R² of 0.94, 0.93 and 0.65 and Relative Percent Deviation (RPD) of 4.80, 3.47 and 2.15, respectively) and with intermediate performance to predict Mn, sum of bases (SB), OM and base saturation (V%) (R² of 0.70, 0.65, 0.59 and 0.58, RPD of 1.65, 1.59, 1.58, 1.46, respectively). The physical and chemical soil properties vary along the slope, this differentiation was detected via electromagnetic spectrum. It appears that the DRS can assist in determining soil properties and knowledge of the soil spatial variability, adding new information to management practices in precision agriculture.