The demand for “green” metals such as lithium is increasing as the world works to reduce its reliance on fossil fuels. More than half of the world’s lithium resources are contained in lithium-brine deposits, including the salt flats, or “salars”, of the Andean region of South America, also known as the Lithium Triangle. The genesis of lithium-brine deposits is largely driven by the leaching of lithium from source rocks in watersheds, transport via groundwater systems to salars, and evaporative concentration in salars. The goal of this research is to create a consistent and seamless methodology for tracking lithium mass from its source in the watershed to its greatest concentration in the nucleus. The area of interest is in and around Bolivia’s Salar de Uyuni, the world’s largest salt flat. We explore how Li-brine deposits form, where the water and solute come from, how the brines are formed, and how abstraction affects the mass balance inside the salar. To support the entire system, open-source Earth observation (EO) data are analysed. We found that by constructing a flexible and repeatable workflow, the question of how lithium reaches the Salar de Uyuni can be addressed. The work demonstrated the importance of groundwater flow to the river network and highlighted the need for flow data for the main river supplying the salar with both water inflow and lithium mass.

There is a critical need for a global monitoring capability for Tailings Storage Facilities (TSFs), to help protect the surrounding communities and the environment. Satellite Synthetic Aperture Radar Interferometry (InSAR) shows much promise towards this ambition. However, extracting meaningful information and interpreting the deformation patterns from InSAR data can be a challenging task. One approach to address this challenge is through the use of data science techniques. In this study, the representation of InSAR metadata as Entity Embeddings within a Deep Learning framework (EE-DL) is investigated for modelling the spatio-temporal deformation response. Entity embeddings are commonly used in natural-language-processing tasks. They represent discrete objects, such as words, as continuous, low-dimensional vectors that can be manipulated mathematically. We demonstrate that EE-DL can be used to predict anomalous patterns in the InSAR time series. To evaluate the performance of the EE-DL approach in SAR interferometry, we conducted experiments over a mining test site (Cadia, Australia), which has been subject to a TSF failure. This study demonstrated that EE-DL can detect and predict the fine spatial movement patterns that eventually resulted in the failure. We also compared the results with deformation predictions from common baseline models, the Random Forest model and Gaussian Process Regression (GPR). Both EE-DL and GPR greatly outperform Random Forest. While GPR is also able to predict displacement patterns with millimetric accuracy, it detects a significantly lower number of anomalies compared to EE-DL. Overall, our study showed that EE-DL is a promising approach for building early-warning systems for critical infrastructures that use InSAR to predict ground deformations.