MultiMiner - Multi Source and Multi Scale Earth Observation and Novel Machine Learning for Mineral Exploration and Mine Site Monitoring

Country / Region: EU

Begin of project: January 1, 2023

End of project: June 30, 2026

Status of project: August 4, 2023

View into the Weissenstein Mine/ RHI Magnesita Source: RHI Magnesita

MultiMiner is a pan-European project involving 13 partners from research and industry from 6 EU member states. The project is led by the Geological Survey of Finland (GTK). The project develops novel data processing algorithms for efficient utilization of Earth Observation (EO) technologies for mineral exploration for Critical Raw Materials (CRM) and mine site monitoring in the EU.

CRMs are mineral raw materials of high economic importance for the EU but have to be imported to a large extent. In order to strengthen the long-term autonomy in important strategic raw materials, it is necessary to identify and develop European deposits. Remote sensing data are a powerful source of information for the exploration of mineral resources and for the later stages of a mine's life to identify or prevent environmental impacts caused by mining activities. However, large-scale use of remote sensing data for exploration often fails due to the lack and transferability of automated analysis methods.

MultiMiner unlocks the potential of EO data, including Copernicus, commercial satellites, upcoming missions, airborne and UAV as well as in situ data, to support the entire mining life cycle including mineral exploration, operational, closure and post-closure stages. This is achieved by developing new data processing algorithms for the exploration of CRMs and the monitoring of related mining activities that are scalable and transferable. The focus is on Machine Learning (ML) and Deep Learning (DL) approaches that require little or no training data (Weakly Supervised and Unsupervised Learning) and synergistically use data from different sensors (e.g., EnMAP, Sentinel-1/-2, drone,) and different resolutions (spatial and spectral).

BGR’s Remote Sensing Working Group is contributing to the project with expertise in Hyperspectral, Multispectral, UAV-based and Radar Remote Sensing and is involved in all five work packages of the project. BGR is leading the work on ‘Scalable mineral prospectivity tools’. Within this WP, enhanced products and mineral maps for exploration are developed based on the automated classification of spectral information using novel Machine Learning approaches. Furthermore, BGR is leading the development of a 4D model formine tailings storage facility composition and volume development and InSAR analyses for mine site stability monitoring.