Dados do Trabalho

Título

PROXIMAL AND ORBITAL SENSING TO ASSIST IN CLAY CONTENT AND YIELD POTENTIAL MAPPING IN AGRICULTURAL FIELDS

Resumo

The advancement in sensing techniques that provide information about field variability for site-specific management purposes leads to questioning which is better related to soil variability and crop productivity. Thus, this study aimed to explore the potential of some soil sensing techniques for mapping an agricultural area's yield potential and clay content. A 106-ha field in Cosmópolis-SP, Brazil, was adopted for the study. For clay content determination, soil samples were collected at 244 geolocated points in the 0-0.20 m depth layer. These same points were used to extract soil apparent electrical conductivity (ECa) measurements at four depths (0.5m, 0.7m, 0.9m, and 1.1m) using the TSM sensor from Geoprospectors, magnetic susceptibility (MS) measured at two depths (0.37m and 0.75m) using the EM38 sensor from Geonics, and soil brightness index (SBI) and SWIR2 band, both obtained via the Sentinel-2 MSI sensor. Normalized average yield (NY) of 4 crops (soybean, sorghum, soybean, and oats) between 2021 and 2022 was also extracted using a yield monitor on the harvester. Regarding the relationship with clay content, the correlations of MS in the two depths (r = -0.57 and -0.56) and of ECa in the first depth (r = 0.53) stood out among the other variables. Multiple regression analysis revealed that the significant variables (p < 0.05) explaining the variability of NY were the ECa in the first and second depth, MS at 0.37m, SWIR2, and SBI, transcribing an adjusted R² of 0.65. These variables showed individual correlations of -0.62, -0.65, 0.55, -0.39, and 0.20 with NY, respectively. Therefore, although ECa, MSa, SWIR2, and SBI are related data layers, they are complementary in determining agricultural area's yield potential.

Palavras-chave

Proximal sensing, Precision Agriculture, Bare Soil, Soil Health, Carbon

Instituição financiadora

FAPESP, process 2023/02532-1, 2020/02223-0 and 2022/03160-8

Agradecimentos

To Dr. Maiara Pusch for collecting and sharing part of the dataset.