Soil degradation detected by temporal satellite image in Sao Paulo state, Brazil

Author(s)

Nascimento, C. M. ; Dematte, J. A. M. ; Mello, F. A. O. ; Rosas, J. T. F. ; Tayebi, M. ; Bellinaso, H. ; Greschuk, L. T. ; Albarracin, H. S. R. ; Ostovari, Yaser

• Document type: Journal article
• Source: Journal of South American Earth Sciences; v. 120, p. 13-pg., 2022-10-07.

Abstract

Soil degradation is a major challenge in the 21st Century. Tropical regions are having the strongest expansion in agricultural lands. Therefore, novel researches on the soil degradation process are imperative to prevent damage to social and environmental dynamics. The main goal of this research was to generate a Soil Degradation Index in a tropical region that includes the entire agricultural areas of the Sa similar to o Paulo State, Brazil, making use of the factors that can be directly related to it as different environmental indicators. The determination of the areas with exposed soil, based on Landsat time series data (1985-2019), was processed with the Geospatial Soil Sensing System methodology. Additionally, thematic maps of clay, cation exchange capacity and organic matter were generated from the calibration of pixels of Landsat images, taking into account surface soil samples (0-20 cm). The spatialization was performed using a random forest algorithm. The average precipitation was obtained for the period of analysis, using the CHIRPS dataset to generate the historical mean information about the years of study. The surface temperature was determined based on the Landsat 5 and 8 thermal bands. Using the elevation model, other terrain data were obtained, such as LS factor and Slope. The land use information was acquired from the Mapbiomas platform and reclassified into five categories of use. The k-means clustering al-gorithm was used to generate the Soil Degradation Index (SDI), which classified the values of the variables into five degradation categories: from 1, very low, to 5, very high. The model was validated using the OM infor-mation. There was an important relationship between the SDI and the spectral surface reflectance obtained by Landsat. Locations with less OM presented a higher degradation level. Integrating multitemporal remote sensing data and environmental variables proved to be effective to assist the SDI, which can monitor and improve the land use decision-making and public policies, in order to prevent economic and environmental issues. (AU)

• FAPESP's process: 14/22262-0 - Geotechnologies on a detailed digital soil mapping and the Brazilian soil spectral library: development and applications
• Grantee: José Alexandre Melo Demattê
• Support Opportunities: Research Projects - Thematic Grants