Early detection model of waterlogging stress in grasses (Urochloa spp.) using hyperspectral imaging

Grasslands cover approximately 40% of the Earth's land surface and almost 70% of the world's agricultural land (Reinermann et al., 2020). These ecosystems provide a variety of benefits, such as supporting biodiversity, grazing livestock, and food production.
Water stress in the soil causes a drastic decrease in the rate of leaf expansion and photosynthesis, inhibiting plant development and reducing biomass production in forage grasses (Fariaszewska et al., 2020). Tolerance to water stress refers to the degree to which a plant adapts to low soil water availability or drought conditions (Basú et al., 2016).
The use of new technologies in precision agriculture allows for the identification of morphological and biochemical changes at early stages, not observable to the human eye (Calderón et al., 2018). Among these tools, hyperspectral imaging stands out, which, by capturing the reflectance of objects across a wide range of spectra, makes it possible to identify small biochemical changes in plants, providing.