Monitoring Uranium Mining and Processing Sites Under Decommissioning Using Hyperspectral Imagery

Previous safeguards-related studies using hyperspectral imagery focused on the possibilities of extracting unique features of uranium mining and milling, and of verifying States’ declarations, such as movement of material. If these unique features existed, they could be used for identifying potential clandestine activities as to uranium mining and processing. However, no specific “uranium signature” was found so far. The objective of the given study was to assess the potential of hyperspectral imagery analysis for detecting environmental contaminations due to uranium mining and processing. Though intended for environmental monitoring purposes, the study was supposed to give also some implications on monitoring uranium mining and processing sites from a nuclear safeguards perspective. Hyperspectral data were acquired near Zwickau/Saxony over Wismut uranium mining legacies under rehabilitation in the context of an airborne hyperspectral campaign using the HyMap Imaging Spectrometer. Simultaneously to image acquisition, the surface reflectance was measured in the area of interest using the ASD Fieldspec Pro spectrometer, and soil and vegetation samples were collected and subsequently analyzed by mass spectrometry (ICP-MS). The first project phase was intended to evaluate new hyperspectral data classifiers and change detection methods. The second project phase aimed at integrating hyperspectral remote sensing and the in-situ measurements by applying traditional and new supervised hyperspectral classifiers. The given paper reports on the second project phase and summarizes the overall findings from either phase