Accurately quantifying soil moisture content (SMC) and soil organic carbon (SOC) at high spatial and temporal resolutions is critical for management of vineyard inputs such as irrigation and fertiliser. In recent years, remote sensing has been shown to be valuable for characterisation of soil parameters, including SM and SOC. However, the labile fraction of SOC, a key determinant of soil microbial activity and soil health, has not been previously quantified using remote sensing. 

My proposed PhD research will use multi-source remote sensing techniques coupled with machine learning and/or deep learning models to predict SMC and SOC (inert and labile fractions).  Besides, the project is evaluate the effectiveness of machine-and-deep-learning models for viticultural practices, providing contextual improvement of SMC/SOC prediction accuracy and to provide greater insights on vineyard behaviour and performance, a first of its kind anywhere.