Mapping Bauxite Mining Residues Using Remote Sensing Techniques

Abstract

Bauxite residuals from abandoned mining sites are both an environmental challenge and a possible source of secondary raw materials. Processing of multispectral and hyperspectral images with the best available techniques can help to produce multiscale spatial maps of elements inside and around the mining sites. The authors propose a procedure for mapping elements concentration using multiple data sets at different scales and resolutions. A comparison between multispectral Sentinel-2 images and hyperspectral PRISMA processing is performed over some case studies of bauxite residues in the Mediterranean area. Specifically, a case study from Italy is composed regarding artificial canyons created by past artisanal mining activities and by stockpiles of extracted bauxite. Hyperspectral punctual measurements (spectroradiometer surveys) were taken in various zones of the bauxite site, where infield topsoil samples were also taken for X-ray fluorescence chemical analysis. Final concentration maps were estimated by performing geostatistical techniques.

Keywords

• bauxite mapping
• hyperspectral images
• band ratio