Hyperspectral remote sensing of carbonatite hosted REE deposits in Namibia


Hyperspectral remote sensing of carbonatite hosted REE deposits in Namibia

Zimmermann, R.; Salati, S.; Andreani, L.; Brandmeier, M.; Gloaguen, R.

High spectral and spatial resolution of hyperspectral images allow mapping and determination of minerals on the earth surface. It also provide valuable information about ore deposits and their alteration zoning.
Carbonatites are well known for hosting economic concentrations of REE-bearing minerals like bastnäsite, monazite and apatite among others (WINTER 2001). They show signifcant spectral characteristics, even REE-absorption bands (ROWAN et al. 1986).
The aim of our study is to apply recent advances in hyperspectral imaging and to develop new tools in order to map these rocks. A refnement of the geological map by lithological mapping, image classifcation, mineral mapping and tectonic geomorphology is also done.
Our research mainly focuses on carbonatites from Namibia. We selected two sites with well known occurrences in Northern Namibia: the Epembe and the Lofdal dykes. Both yield signifcant REE- concentrations and being under exploration. In these areas LANDSAT 8, SRTM, EO-1 Hyperion and airborne hyperspectral (HyMap) data overlap giving an excellent coverage.
The data had been processed for atmospherical and geometric corrections frst. For the EO-1 Hyperion data a more carefully pre-processing had to be applied due to streaking and smiling effects. Results of LANDSAT 8 classifcation had been used for validation and classifcation purposes. Detailed spectral signatures and mineral maps are extracted by hyperspectral imaging. The results confrm former observations by BEDINI (2009) on the Sarfartoq carbonatite complex in West Greenland. Secondly the structural controls of emplacement of the selected carbonatite dykes had been carried out using tectonic geomorphology and feld observation. For this purpose the TecDEM-toolbox (SHAHZAD & GLOAGUEN 2011) was applied to the SRTM-data for drainage- network and stream-profle analysis.
Combining all these information with geological knowledge of carbonatites and their occurrences, it is possible to explore new, unknown deposits by remote sensing applications.

  • Poster
    GeoFrankfurt 2014, 22.-24.09.2014, Frankfurt/Main, Deutschland

Permalink: https://www.hzdr.de/publications/Publ-20851