The soil mapping is necessary to achieve the goal of improvement in the field of production and environmental monitoring. By the use of sensors, such result can be reached faster and with more quality. The objective of this work will be to develop a methodology to assemble data from multi-source satellite images with different features, getting higher detail for digital mapping. The study area covers five municipalities located in the region of Piracicaba, São Paulo. Multi temporal satellite data from RapidEye, Sentinel-2, Landsat 5-8 and Planet Constellation will be used. For each one satellite sensor, bare soil locations in every image will be determined and assembled into a new image, designated as SYSI (Synthetic Soil Image). There will be several SYSIs, one for each satellite sensor, which would cover a particular region. These SYSIs will be assembled (overlay) into a single image (bare soil composite) based on data of all the satellites, giving priority, in the sequence, to those with better spatial and spectral resolution. The same methodological basis will be used for vegetation. Therefore, the final SYSI of the entire study area will be transformed to color and via color related to drainage and pedological classification issues. After that, an image in the reverse direction will be created; in other words, only with vegetation. Thus, the SYVEI (Synthetic Vegetation Image) will allow to verify the soil variations using the vegetation reflectance as a "sensor" of the soil properties. The union of soil and vegetation reflectance information will allow to specialize the color of the soil and other important attributes in the pedological classification, as well as in the evaluation of the productive potential.
News published in Agência FAPESP Newsletter about the scholarship: