Proximal and remote sensing on the soil processes: from punctual to spatial approaches

• Autor(a) principal: Souza, Arnaldo Barros e
• Data de Publicação: 2020
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: http://www.teses.usp.br/teses/disponiveis/11/11140/tde-04062020-101342/
• topic: Agrupamento; Clustering; Formation processes; Heterogeneity index; Homogeneidade de mapas; Horizonation; Horizontação; Índice de heterogeneidade; Maps homogeneity; Processos de formação

The characterization of soils has been highlighted due to the need to conserve natural resources in order to promote greater food and environmental security. The great influence of soils on water and climate resources has long been recognized by the scientific community. However, knowledge about the predominant processes and the variation of soil attributes in depth and throughout the landscape has been limited by the high cost of wet chemical analyzes, traditionally used. These analyzes were calibrated for routine use and mainly for agronomic purposes, leaving much to be desired in more comprehensive land use planning. As an alternative to this gap, the use of geophysical equipment associated with relief and satellite data has been shown to be very useful in characterizing soils, both on the surface and the subsurface. Techniques such as visible and infrared spectroscopy, X-ray fluorescence, magnetic susceptibility and apparent electrical conductivity are among the most used. These technologies provide information related to mineralogy, water and solute dynamics, granulometric distribution, the distribution of elements in the soil profile and the content of organic material, quickly and at low operating cost. Based on these data it is possible to make inferences about the geological formations on which the soil developed, the cycles of deposition and removal of particles as well as the translocation within the profile, which may also be associated with past climates. In addition, it is possible to identify the dynamics of the elements in the soil profile, which is conditioned by the water dynamics in the soil profile and in the landscape as a whole. However, the use of these equipment has often been applied only in the construction of models to estimate soil attributes determined by traditional analyzes. These analyzes contain errors considered acceptable for direct application, but which are maximized and propagated by the modeling. In this sense, the main proposal of this work is to provide a basis for the characterization of soil bodies, from the profile to the landscape, using only the information obtained with proximal and remote sensors. Thus, indices and forms of observation have been proposed in order to enable the extraction of information directly from the data obtained with the sensors. Chapter 1 deals with the study of soil horizonation with spectral sensors and the X-ray fluorescence used in trenches. Chapter 2 presents the approach with profiles simulated where it is suggested a Profile Heterogeneity Index based on sensors and their relationship with the landscape. The third one addresses the influence of the use of satellite reflectance compared with data from laboratory sensors, with and without sample preparation, as well as the quality of the models generated to estimate weathering indexes. In the fourth and last chapter, the integration of sensor data with satellite and relief data for the identification of soil environments was studied. In this chapter, an index is proposed to measure the homogeneity of the generated mapping units (soil environments) given a reference map.