GRSG 34th Conference 2023

Title: Preliminary observations of Sardinian Historical mine tailings using Google Earth Engine

Author: Susanna Grita

Abstract:

Sardinia is an important historical mining district in Europe and used to be the most important mining region in Italy. Extracted minerals include Barium, Fluorspar, Bauxite and polymetallic minerals bearing Pb, Ag, Cu and Zn. Today, almost all the mines present in Sardinia have ended their extractive activity, and their residues are now a potential source of pollution for soil, water, and air, but also represent a health hazard for the communities living close to the dumpsites.

Nevertheless, in recent years, due to the need for Europe to diversify its supply of Critical Raw Materials (CRMs), mine wastes have newly gained attention as a potential source of CRMs. Recent studies show that some mines in Sardinia (such as Silius mine and Buddusò granite quarries), may contain interesting amounts of REE as by-products. Geochemical studies on the Ingurtosu mine also suggest the presence of such elements in the tailings. The use of satellite and aerial data has proved effective for characterizing mine wastes from a geochemical point of view, allowing for an estimation of the pH level or the concentration of heavy metals.

In many studies the objective is to assess the environmental hazard these residues represent, or to map the Acid Mine Drainage (AMD) deriving from them. However, recent research projects have used remote sensing techniques to map minerals with a potential economic value such as in Kasmaeeyazdi et al. (2021, 2022) and Ogen et al. (2022), which yielded promising results. Yet, this research domain remains understudied. In addition, there is almost no literature on the characterization of historical mine residues in Sardinia using multispectral and hyperspectral satellite data. A more in-depth understanding of the spectral response of these sites could positively contribute to conducting a systematic assessment on the potential mineral resources available on the island. In this context, the aim of the present study is to characterize the mineralogical and geochemical composition of some mine waste sites located in Sardinia by using multispectral and hyperspectral remote sensing techniques. Sentinel-2, EnMap and PRISMA data are used to assess the extent of mine residues of the chosen test sites.

The mineralogical composition of its constituents is estimated through the application of band ratios and a semi-automatic classification is carried out. The research project investigates whether such a technique is suitable for distinguishing different mineral phases in the test areas. This is a preliminary analysis which aims to pave the way for further research on the detection of REE in those test sites using hyperspectral satellite and aerial images. Such research contributes to the development of a rapid, non-invasive technique for the identification of re-extractable mine tailings in Italy and abroad.