Surveying the Potential Auriferous Areas in Western Region of MeshkinShahr, using Hyperion Images

Nowadays, the geologists and mining scientists are benefited from remote sensing which acquires unique spectral properties of minerals. The advent of hyperspectral remote sensing has offered new possibilities for measuring specific mineral characteristics that were difficult to be measured, using conventional multispectral sensors. In this study, a Hyperion image from the EO-1 satellite which referred to 4th September 2004 for Meshkinshahr, Ardabil was used for surveying the potential auriferous (such as gold mines, golden streaks, etc.) Gold was elected in this study due to its rare existence and of course its high economic value and importance. The Hyperion images were analyzed by using Spectral Angular Mapper (SAM), Linear Spectral Unmixing (LSU) and the field measurements. The end-members were extracted; using the field measured data and also through the results obtained from using Pixel Purity Index (PPI). The overall accuracy of the derived potential distribution maps were examined using the field measured data and the earlier geological maps. The results showed that in comparison to LSU, a higher accuracy has been achieved when SAM method was applied and the end-members were selected using the PPI (overall acc. =94%). Probably this has been due to the rare existence of gold and relative homogeneity of minerals which exist in a pixel. The outcome of this study highlights the ability of SAM and hyperspectral images for surveying identifying the potential auriferous areas in the mentioned region.