GRSG 35th Conference 2024 Abstract

Title: Multiscale Mapping of Mineralogy in the Memi Tailing (Cyprus) Using Spectral Remote Sensing Technology

Author: Saeid Asadzadeh

Organisation: German Research Center for Geosciences (GFZ)

The intensified oxidation of sulfide minerals in tailing and mining environments can lead to the development of acid rock drainage (ARD). This acidic environment promotes the formation of various secondary iron oxides/hydroxides/sulfates minerals including jarosite, copiapite, schwertmannite, ferrihydrite, goethite, and eventually hematite. Most of these mineral assemblages are spectrally active, making them detectable through imaging spectroscopic methods. Hyperspectral imaging has been used to determine acid discharge zones and assess their environmental impacts.

To demonstrate the advantages of state-of-the-art sensing technology for this aim, we conducted a multi-sensor and multi-scale remote sensing study as part of the EU-funded M4Mining project, collecting spectral remote sensing data over the Memi mine tailing site, a legacy pyrite mine in the Republic of Cyprus. The data collection included EnMAP hyperspectral data with 224 spectral bands at a 30-meter spatial resolution, WorldView-3 multispectral data with 16 bands and sub-meter spatial resolution, and Sentinel-2 data with moderate spatial (i.e., 20 m) and high temporal resolution.

Additionally, centimeter-scale hyperspectral data were acquired on the ground and from the air using the HySpex Mjolnir imaging system, mounted on a tripod and a UAV platform, respectively. Ground-based surveys of anomalous and non-anomalous outcrops were also conducted, involving measurements with a handheld point-spectrometer (ASD FieldSpec-3) and sample collection for geochemical and mineralogical analysis. All collected data were converted to surface reflectance, in the range of roughly 400 – 2500nm (Visible to near- and Shortwave infrared) and processed using a spectral matching method.

In this approach, a spectral library consisting of jarosite, copiapite, goethite, and hematite and their mixtures was matched against the imaging spectroscopic data. The coefficient of determination was used to assess the degree of match between the spectral library and each image pixel, allowing us to systematically map the occurrence of these minerals in the imaging data. Special emphasis was placed on mapping jarosite, as a key mineral for identifying acid discharge zones. We showed that EnMAP hyperspectral data can effectively screen large areas for signs of acidity in the environment.

High-resolution data from WorldView-3/UAVs were able to delineate the spatial extent of potentially acidic zones, with Sentinel-2 data enabling monitoring of changes over time. Ground spectroscopy provided insights into the maturity of jarosite in the tailing, allowing us to spectrally identify different species of jarosite in the environment. By integrating the results from multiple scales, we developed a comprehensive understanding of acid rock drainage around the Memi mine and tailing site. These maps, combined with a digital elevation model of the area, helped to predict the locations of acidic discharge zones contributing to surface runoff. This study highlights the significant role of imaging spectroscopy in risk assessment and remediation efforts for mining environments impacted by ARD.