Assinatura magnética e espectral na estimativa de elementos potencialmente tóxicos em solos tropicais
Carregando...
Arquivos
Data
2019-08-01
Autores
França, Ana Beatriz Coelho
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A contaminação dos solos causada pela ação antrópica representa uma preocupação mundial no escopo da segurança alimentar. Dessa forma, torna-se necessário mapear de maneira rápida e não poluente os teores dos elementos potencialmente tóxicos (EPTs) dos solos, como Ba, Co, Cr, Cu, Ni, Pb e Cd, a fim de mitigar a ação danosa destes elementos no ambiente. A falta de informações sobre estes EPTs e a necessidade de um grande número de amostras dificultam as avaliações de risco em áreas contaminadas e sua espacialização em grandes áreas agrícolas. Nesse sentido, a Suscetibilidade magnética (SM) e a Espectroscopia de reflectância difusa (ERD) podem ser técnicas indiretas promissoras para a predição dos teores dos EPTs por estar relacionada com atributos pedoindicadores do solo, tais como a mineralogia. Por isso, os objetivos com este trabalho foram: (a) avaliar os teores de EPTs em solos sob cultivo de cana-de-açúcar e compreender a influência antrópica na presença destes elementos nos solos, e (b) estimar os teores dos EPTs (Ba, Co, Cr, Cu, Ni, Pb e Cd) nos solos com o auxílio das medidas de SM e ERD. As amostras de solo foram coletadas em uma área de transição de solos originários de basalto, arenito Botucatu e Depósito Colúvio Eluvionar. Foram realizadas análises granulométricas, químicas, mineralógicas, espectrais e medidas de SM. Os dados foram analisados por estatística descritiva, correlação de Pearson, regressão linear múltipla (RLM), geoestatísica e funções de pedotranferência para a estimativa dos teores dos EPTs, utilizando a SM e ERD. Modelos de predição dos teores dos EPTs foram calibrados, utilizando a SM e argila para os elementos Co, Cr, Cu, Ni e Pb (R2 > 0,59). A predição dos teores dos EPTs, por meio da ERD, teve coeficientes de determinação (R2) maiores que 0,72, exceto para o Cu (R2 = 0,47). O Cu e o Ni apresentaram os melhores modelos, utilizando a SM. Por outro lado, Ba, Cr e Pb foram preditos com mais assertividade, utilizando a ERD. O Co pode ser estimado pelas duas técnicas de quantificação indireta (SM e ERD). Os EPTs apresentaram dependência espacial e correlacionaram-se positivamente com os teores dos óxidos de ferro e SM. Co, Cu e Ni apresentaram teores acima dos valores de referência de qualidade nos solos de basalto, inclusive nos pontos analisados na mata nativa sobre basalto. A presença de EPTs nos solos está fortemente relacionada às suas concentrações no material de origem evidenciada pela análise dos componentes principais.
Contamination of soil caused by anthropic action is a worldwide concern for food safety. Thus, it is necessary to map the potentially toxic elements (PTEs) as Ba, Co, Cr, Cu, Ni, Pb and Cd, in a quick and non-polluting way in order to mitigate the damaging action of these elements in the environment. In addition, the lack of information on these elements prevents risk assessments in contaminated areas and its spatialization in large agricultural areas requires many samples. In this sense, magnetic susceptibility (MS) and the diffuse reflectance spectroscopy (DRS) may be promising indirect techniques for the prediction of PTEs because it is related to soil pedoindicator attributes such as mineralogy. Therefore, the objectives with this work were: (a) to evaluate the levels of PTEs in soils under sugar cane cultivation and to understand the anthropic influence on these elements in the soils and (b) to estimate the content of PTEs (Ba, Co, Cr, Cu, Ni, Pb and Cd) in soils through MS and DRS. Soil samples were collected in a transition area of soils originating from Basalt, Botucatu sandstone and Eluvionar Collution Deposit. Sieve analysis, chemical, mineralogical, spectral and MS measurements were performed. Data were analyzed by descriptive statistics, Pearson's correlation, linear multiple regression (LMR), geostatistics and pedotransfer functions were used in the estimation the content PTEs using MS and DRS. The prediction models of the PTEs were calibrated using MS and clay for the elements Co, Cr, Cu, Ni and Pb (R2> 0.59). For the prediction the content PTEs by DRS, had coefficients of determination (R2) greater than 0.72, except for Cu (R2 = 0.47). The Cu and Ni presented the best models using MS, while Ba, Cr and Pb were predicted with more assertiveness using DRS. The Co can be estimated by the two techniques of indirect quantification (MS and DRS). The PTEs presented spatial dependence and were positively correlated with iron oxides content and MS. The Co, Cu and Ni presented levels above the background values of quality in the basalt soils, including in the points analyzed in the native forest on basalt. The presence of PTEs in soils is strongly related to their concentrations in the source material evidenced by the analysis of the main components.
Contamination of soil caused by anthropic action is a worldwide concern for food safety. Thus, it is necessary to map the potentially toxic elements (PTEs) as Ba, Co, Cr, Cu, Ni, Pb and Cd, in a quick and non-polluting way in order to mitigate the damaging action of these elements in the environment. In addition, the lack of information on these elements prevents risk assessments in contaminated areas and its spatialization in large agricultural areas requires many samples. In this sense, magnetic susceptibility (MS) and the diffuse reflectance spectroscopy (DRS) may be promising indirect techniques for the prediction of PTEs because it is related to soil pedoindicator attributes such as mineralogy. Therefore, the objectives with this work were: (a) to evaluate the levels of PTEs in soils under sugar cane cultivation and to understand the anthropic influence on these elements in the soils and (b) to estimate the content of PTEs (Ba, Co, Cr, Cu, Ni, Pb and Cd) in soils through MS and DRS. Soil samples were collected in a transition area of soils originating from Basalt, Botucatu sandstone and Eluvionar Collution Deposit. Sieve analysis, chemical, mineralogical, spectral and MS measurements were performed. Data were analyzed by descriptive statistics, Pearson's correlation, linear multiple regression (LMR), geostatistics and pedotransfer functions were used in the estimation the content PTEs using MS and DRS. The prediction models of the PTEs were calibrated using MS and clay for the elements Co, Cr, Cu, Ni and Pb (R2> 0.59). For the prediction the content PTEs by DRS, had coefficients of determination (R2) greater than 0.72, except for Cu (R2 = 0.47). The Cu and Ni presented the best models using MS, while Ba, Cr and Pb were predicted with more assertiveness using DRS. The Co can be estimated by the two techniques of indirect quantification (MS and DRS). The PTEs presented spatial dependence and were positively correlated with iron oxides content and MS. The Co, Cu and Ni presented levels above the background values of quality in the basalt soils, including in the points analyzed in the native forest on basalt. The presence of PTEs in soils is strongly related to their concentrations in the source material evidenced by the analysis of the main components.
Descrição
Palavras-chave
Espectroscopia de reflectância difusa, Magnetismo no solo, Segurança alimentar, Pedometria, Óxidos de ferro, Metais pesados, Heavy metals, Iron oxides, Pedometric