Use este identificador para citar ou linkar para este item: http://repositorio.ufla.br/jspui/handle/1/59266
Título: Green tech proximal sensor analyses: advances for soil, parent material and geological exploration
Título(s) alternativo(s): Análise “verde” com sensores próximos: avanços exploração de solo, material de origem e geológica
Autores: Silva, Sérgio Henrique Godinho
Menezes , Michele Duarte de
Andrade, Renata
Serafim, Milson Evaldo
Couto, Eduardo Guimarães
Palavras-chave: Geoquímica do solo
Sensores proximais
Aprendizado de máquinas
Material de origem do solo
Exploração geológica
pXRF (Espectroscopia de fluorescência de raios X portátil)
Suscetibilidade magnética
Pegmatitos
Modelagem espacia
Soil geochemistry
Proximal sensors
Machine learning
Soil parent material
Geological exploration
pXRF (Portable X-ray fluorescence spectrometry)
Magnetic susceptibility
Pegmatites
Spatial modeling
Precision agriculture
Data do documento: 23-Ago-2024
Editor: Universidade Federal de Lavras
Citação: PIERANGELI, Luiza Maria Pereira. Green tech proximal sensor analyses: advances for soil, parent material and geological exploration. 2024. 118p. Tese (Doutorado em Ciência do Solo) - Universidade Federal de Lavras, Lavras, 2024.
Resumo: Soil parent material (PM) is proving to be a critical factor in understanding soil variability, with the complexity and inaccessibility of PM in deep soils posing a challenge for detailed assessments. Proximal sensor approaches, including portable X-ray fluorescence (pXRF) and magnetic susceptibility (MS), offer practical solutions for predicting soil PM. The global search for sustainable energy alternatives has increased significantly, with lithium (Li) emerging as a key element for rechargeable Li-ion batteries. The demand for Li requires the development of cost-effective and rapid exploration methods to improve the identification of new deposits, with lithium-cesium-tantalum (LCT) pegmatites as primary sources of Li. This dissertation is divided into four chapters: (I) a pilot study focused on the generation of spatial PM predictive models for three different rock types (charnockite, mudstone, and alluvial sediments) at the Palmital Experimental Farm (Brazil), using Random Forest (RF) algorithm combined with pXRF and MS data from A and B horizons; (II) the evaluation of the effectiveness of pXRF data and the RF algorithm in predicting Li content in soil samples and Li-rich soil PM using Li pathfinder elements; (III) the comparison pXRF vs. total digestion Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) for soils and rocks analysis. The findings in this dissertation propose alternative, cost-effective methods for assessing soil PM spatial variability. Paper (I) had a strong validation for PM prediction results (Kappa coefficient = 0.85 and overall accuracy = 0.93). Meanwhile, PM prediction model in paper (II) achieved a Kappa coefficient of 0.77 and an overall accuracy of 0.85. The Li prediction model tested in paper (II) achieved a coefficient of determination (R2) of 0.86, root mean square error (RMSE) of 68.5 mg×kg-1, and residual prediction deviation (RPD) of 1.78. Paper (III) evaluated the performance and comparability of two pXRF systems, Alpha and Beta. Both pXRFs systems produced similar results compared to reported concentrations of certified reference materials between systems and methods, showing tendencies of overestimate or underestimate elements. They could provide an alternative, pXRF-based method for more sustainable prospecting methods for PM and Li content determination and exploration. Furthermore, they help in complex and careful decisions and clarify some doubts regarding the use and comparability of results obtained from the same sample with different pXRF models.
URI: http://repositorio.ufla.br/jspui/handle/1/59266
Aparece nas coleções:Ciência do Solo - Doutorado (Teses)



Este item está licenciada sob uma Licença Creative Commons Creative Commons