Integrated drone-borne and ground-based hyperspectral imaging for exploration targeting


Integrated drone-borne and ground-based hyperspectral imaging for exploration targeting

Herrmann, E.; Zimmermann, R.; Gloaguen, R.

The Iberian Pyrite Belt (IPB) in Southwest Spain and South Portugal is best known for its numerous massive sulphide deposits. With their 4000 years history of mining, the mines of Rio Tinto are of important economic importance in the IPB. Orebodies and related alteration zones were mined for copper, gold, silver and many other pre- cious metals. We selected this site as a case study for the exploration of massive sulphide deposits due to its excellent data basis, economic importance and excellent outcrops. Ground-based and drone-borne hyperspectral data were acquired during 3 field campaigns in 2017 and 2017. Drone-borne hyperspectral images are acquired by a RIKOLA Hyperspectral Imager in the wavelength range from 500 to 900 nm. Not only spectral information can be gathered that way, overlapping images can also be used to calculate 3D surface models using Structure-from-Motion photogrammetry. Resulting point cloud can be basis for 3D data integration and correlation of spatial orientation of surfaces to spectral characteristics. Ground-based data are acquired with a Specim AisaFenix hyperspectral camera with a wavelength range from 400 to 2500 nm.
A complex geological history lead to the appearance of numerous regional geochem- ical and structural characteristics correlated to the orebodies that comprise different types of syn- and post-mineralisation alteration zones. These features are hardly re- cognisable using classical geological fieldwork techniques. Until now, time consum- ing and expensive geological interpretations rely almost exclusively on geochemical data. Instead, we argue that by using in hyperspectral data, those alteration zones can easily be determined, due to characteristic features in the VNIR and SWIR range. This provides a powerful tool for a fast, low-cost and spatially precise altera- tion mapping to guide the exploration process. The result of this data integration is comparable to an early stage geological surface model. It can provide a very precise tool for exploration and mining at all stages.
We chose the Corta Atalaya in Riotinto/Spain, one of the most profitable mines in the Spanish part of the Iberian pyrite belt, to demonstrate this approach of modern, re- mote sensing based geological exploration methods. Methods include traditional al- gorithms like Spectral Feature Fitting and Mapping absorption wavelength, but also non-linear unsupervised classification algorithms. Different kind of alteration zones are discriminated and interpreted in terms of their spectral-spatial distribution.

  • Poster
    10th EARSeL SIG Imaging Spectroscopy Workshop, 19.-21.04.2017, Zürich, Schweiz

Permalink: https://www.hzdr.de/publications/Publ-24906
Publ.-Id: 24906