Towards a Global Carbonatite Atlas: A Satellite Remote Sensing Approach to Mapping and Characterization

Rare Earth Elements (REEs) have become critical for global technological advancements and, consequently, economic growth. Ensuring supply requires significant future exploration, potentially including the use of space-borne hyperspectral data for direct mapping of REEs. While space-borne detection of REEs has been demonstrated (e.g., Asadzadeh et al., 2024), this approach has limited application. Low concentrations of these valuable resources in most carbonatite host rocks and small sizes of ore zones represent a  major hurdle and complicate reliable detection and mapping efforts. 

We propose a comprehensive approach to remotely characterise carbonatites, which are known to host REEs, with the aim of improving our overall understanding of these unusual rocks and better identifying potentially fertile systems. Carbonatites are typically classified into three types: calcio-carbonatites, magnesio-carbonatites, and ferro-carbonatites. However, recent studies, such as Mitchell & Gittins (2022), suggest additional variants that don't fit these categories, indicating the current classification system may require further refinement. Regardless of classification complexities, the composite mineralogical phases of carbonatites are spectrally active and exhibit distinctive absorption features in hyperspectral data. Furthermore, the presence of alteration halos and the structural controls commonly associated with carbonatite structures make these sites well-suited for optical remote sensing studies by both hyperspectral and multispectral datasets. This paves the way for the development of a global carbonatite atlas based on remote sensing data.

We demonstrate the feasibility of the approach using two REE-bearing carbonatite complexes in Namibia—Lofdal and Marinkas-Quellen. We selected EnMAP provided by the German Aerospace Center (DLR) hyperspectral data as they are the most accurate to this date (Chakraborty et,al., 2024). We employed different processing techniques such as minimum wavelength mapping and spectral abundance analysis to map the carbonatite lithologies in each of the two sites individually. We then streamlined the workflow to identify common parameters and trained a decision tree to map the broader carbonatite footprints across both sites. In parallel, Sentinel-2 multispectral data was used to map geological structures (e.g., dykes, faults, and bedding) aiming to understand controls on carbonatite emplacement. A fusion-based resolution enhancement algorithm was also applied to integrate EnMAP with Sentinel-2 data, providing a more spatially detailed understanding of the targets. 

We aim to expand this study to include a wider range of carbonatite complexes, with the goal of creating a global carbonatite atlas. By covering diverse geological settings and ages, this atlas will capture the full spectrum of mineralogical variation and structural features, enhancing our understanding of carbonatite bodies. This atlas not only will promote the applications of remote sensing techniques in carbonatite studies but also provide a valuable basis for future exploration of REEs in carbonatite settings. 

1. Asadzadeh, S., Koellner, N., & Chabrillat, S. (2024). Detecting rare earth elements using EnMAP hyperspectral satellite data: a case study from Mountain Pass, California. Scientific Reports

2. Mitchell, R. H., & Gittins, J. (2022). Carbonatites and carbothermalites: A revised classification. Lithos

3. Chakraborty, R., Rachdi, I., Thiele, S., Booysen, R., Kirsch, M., Lorenz, S., ... & Sebari, I. (2024). A Spectral and Spatial Comparison of Satellite-Based Hyperspectral Data for Geological Mapping. Remote Sensing