UTILIZING EMIT’S IMAGING SPECTROSCOPY SATELLITE FOR GEOLOGIC CLASSIFICATION AND LITHIUM DETECTION IN THE MCDERMITT CALDERA

Recent advances in remote sensing technology have introduced new possibilities for the application of imaging spectroscopy to expand geologic mapping capabilities. Successful integration of this technology with remote sensing systems (e.g., satellites) would reduce the numerous barriers associated with physical fieldwork. One sector that has the protentional to greatly benefit from the development of this technology is renewable energy. Lithium (Li) is a geologically concentrated metal with increasing demand, needed for producing renewable energy technologies (e.g., Li-ion batteries for electric cars). Detecting Li sources is therefore crucial for transitioning into an eco-efficient energy society, however, the ability to do so from remote sensing technology is still in its nascent stages. Here we take advantage of NASA’s EMIT (Earth Surface Mineral Dust Source Investigation) mission and proximity to the McDermitt Caldera, one of the largest Li sources in the world, to develop and test a framework for remotely sensing Li enrichment. The McDermitt Caldera, located in SE Oregon and northern Nevada, is an area of increasing interest due to its abundance of Li found in clays. Understanding the variation in Li concentrations within the caldera, coupled with the storage center for that Li (rock vs. clay) allows for more precise Li extraction, reducing waste and environmental hazards associated with mining. EMIT is a space-borne imaging spectrometer collecting data from the ISS since 2022. We developed a PLS-DA (partial least square discriminant analysis) model to discriminate the geologic units from EMIT’s spectral image data, in accordance with the geologic map of the McDermitt Caldera. In addition, a PLS-R (regression) model will be developed to couple the concentration of Li in accordance with spectral data. The results will allow us to better define the strengths and limitations for using remote sensing to understand geologic units. The methods and models developed here are transferable to other disciplines, advancing the application of imaging spectroscopy research across different fields.