Soil organic carbon monitoring in an irrigated agriculture area

Carregando...
Imagem de Miniatura

Data

2021-03-04

Autores

Minhoni, Renata Teixeira De Almeida [UNESP]

Título da Revista

ISSN da Revista

Título de Volume

Editor

Universidade Estadual Paulista (Unesp)

Resumo

There is a global growing interest in agricultural practices to increase soil organic carbon (SOC) to promote soil fertility, climate change mitigation, and food security. Therefore, SOC measuring and monitoring is a key step in agricultural and environmental projects. The first chapter of this study aimed to identify potential proxies for topsoil SOC in an agricultural area, characterized by the adoption of central pivot irrigation, reduced cultivation, crop rotation, and use of crop residues. Nine spectral indices were extracted from time-series Sentinel-2 satellite images of bare soil. Additionally, three topographic indicators were extracted from a digital elevation model (DEM). Linear regression models were generated between the spectral indexes and SOC, considering the soil sampled in 2019, and between topographic indicators and SOC. The spectral indices with the best performance were EVI2 (Enhanced Vegetation Index), NDVI (Normalized Difference Vegetation Index), MSAVI2 (Modified Soil-Adjusted Vegetation Index), and RVI (Ratio Vegetation Index), and the elevation was the best topographic indicator. After a seven-year study period (from 2012 to 2019), the area showed a significant increase in the topsoil SOC. The results of this research consolidate that the use of a multitemporal dataset can lead to more accurate identification of topsoil SOC proxies, instead of the use of a single date image. Two orchards were selected in the region of San Joaquin Valley, California (USA), having as the main objective of the second chapter characterize the multi-scale (tree scale and orchard scale) distribution of SOC and soil total nitrogen (STN) in micro-irrigated citrus orchards, grown under the same soil conservation practice. The first with Page mandarin (Citrus reticulata) in Strathmore, and the second with Washington navel orange (Citrus sinensis) in Ivanhoe. Both orchards are irrigated with micro-sprinklers, and pruning residues are disposed on the inter-rows. Soil samples were collected along transects, and each transect was divided into three sections: the first (α) located 0.6 m from the tree trunk, the second (β) 1.2 m from the trunk (approximately below the canopy projection), and the third (γ) located in the center of the inter-row. In each section single samples were collected at two soil layers, 0-0.05 m and 0.05-0.4 m. From 0 to 0.05 m in Strathmore, the average SOC concentrations in the β and γ sections and the STN concentration in the γ section showed no statistical difference with the orchard. Whereas, from 0.05 to 0.40 m and from 0 to 0.40 m there was no statistical difference between the sections and the orchard. In Ivanhoe, where the soil is more homogeneous, there was no significant difference between the sections and the orchard. The results of this study can support the elaboration of future SOC and STN monitoring projects in the region.
Há um crescente interesse global em práticas agrícolas que aumentem o carbono orgânico do solo (COS) para promover a fertilidade, a mitigação das mudanças climáticas e a segurança alimentar. Dessa maneira, a medição e monitoramento do COS é uma etapa muito útil em projetos agrícolas e ambientais. O primeiro capítulo deste estudo teve como objetivo principal a identificação de potenciais indicadores para o carbono orgânico presente na camada superficial do solo de uma área agrícola, caracterizada pela adoção de irrigação por pivô central, cultivo reduzido, rotação de culturas e utilização de resíduos de culturas. A partir de uma série temporal de imagens de solo exposto obtidas com o satélite Sentinel-2 foram extraídos nove índices espectrais. Também foram extraídos três indicadores topográficos a partir de um modelo digital de elevação (MDE). Foram gerados modelos de regressão linear entre os índices espectrais e COS, considerando o solo amostrado em 2019, e entre indicadores topográficos e o COS. Os índices espectrais de melhor desempenho foram EVI2 (Enhanced Vegetation Index), NDVI (Normalized Difference Vegetation Index), MSAVI2 (Modified Soil-Adjusted Vegetation Index) e RVI (Ratio Vegetation Index), e a elevação foi melhor indicador topográfico. Após um período de sete anos (2012-2019) de estudo, a área apresentou um aumento significativo no COS na camada superficial. Os resultados desta pesquisa consolidam que o uso de um conjunto de dados multitemporais pode levar a uma identificação mais precisa de indicadores de COS da camada superior do solo, ao invés do uso de imagem de uma única data. Foram selecionados dois pomares na região do Vale de San Joaquin na Califórnia (EUA), tendo como objetivo principal do segundo capítulo caracterizar a distribuição multi-escalar (escala de árvore e escala de pomar) de COS e nitrogênio total do solo (NTS) em pomares cítricos irrigados por microaspersão, cultivados sob a mesma prática de conservação do solo. O primeiro com Page mandarin (Citrus reticulata) em Strathmore, e o segundo com Washington navel orange (Citrus sinensis) em Ivanhoe. Ambos os pomares são irrigados com microaspersores e dispõem resíduos de poda nas entrelinhas. Foram coletadas amostras de solo ao longo de transectos, e cada transecto foi dividido em três seções: a primeira (α) localizada a 0,6 m do tronco da árvore, a segunda (β) a 1,2 m do tronco (aproximadamente abaixo da projeção da copa das árvores) e a terceira (γ) localizada no centro da entrelinha. Em cada seção foram coletadas amostras simples em duas camadas, em 0-0,05 m e em 0,05-0,4 m. De 0 a 0,05 m em Strathmore, as concentrações médias de COS nas seções β e γ e a concentração de NTS na seção γ não apresentaram diferença estatística em relação ao pomar. Enquanto que, de 0,05 a 0,40 m e de 0 a 0,40 m não houve diferença estatística entre as seções e o pomar. Em Ivanhoe, onde o solo é mais homogêneo, não houve diferença significativa entre as seções e o pomar. Os resultados deste estudo poderão subsidiar na elaboração de projetos futuros de monitoramento de COS e NTS na região.

Descrição

Palavras-chave

Citrus orchard, Conservation practices, Remote sensing, Soil sampling, Spectral indices, Pomar de citros, Práticas conservacionistas, Sensoriamento remoto, Amostragem de solo, Índices espectrais

Como citar