Mapping radioactive minerals using remote sensing: a case study of Mrima Hill Kwale County, Kenya

Abstract

Mrima hill is an elliptical carbonatite plug area of approximately 2 km across, a gazetted forest reserve in Kwale county, bounded by 4° 29'l0"S; 39° 15 '10'’ E coordinates, 750 feet above the sea level. The hill is covered by deeply weathered materials. Rare earth elements, niobium, Monazite minerals and associated carbonatite rocks are known to exist in the area. The area is classified as a high background radiation. Applications of geological remote sensing and GIS for radioactive mineral mapping have not been fully integrated into the mineral exploration activities of the Geological Survey of Kenya. This study employed remote sensing and Geographic Information System (GIS) to map minerals in Mrima hill region in Kwale County, specifically radioactive minerals, as the area has been classified a high background radiation area. In this study, the data used was obtained from Earth explorer (https://earthexplorer.usgs.gov/) website, with spatial resolution of 30 m, and was processed for mineral spectral signatures by using ENVI5.3 and Arc Map 10.3 software by means of the color composite, band rationing, principal component analysis and supervised classification.Landsat-8 OLI imagery of Mrima hill was processed to enhance the geological features and mineral potential of the area. Band ratios 6/7, 6/5, 4/2 were assigned to RGB. Band ratio 4/2 highlighted ferric ion minerals, 6/5 emphasized ferrous minerals, and 6/7 distinguished iron oxide minerals from carbonate minerals. In a second technique, band ratio 6/7 was replaced with 7/ 5 in order to accentuate clay minerals with high reflectance within band 7. Supervised classification training data were obtained using five classes for rocks associated with radioactive minerals (carbonatite, granites, sandstone, serpentine and shale). The classification using maximum likelihood classification was fairly accurate and matched the radiometric and geologic map of the area, also showing an alteration zone that coincided with the high dose rate areas. However, for areas covered by vegetation, botanical indicators of vegetation species associated with radioactive mineralization including, the Asparagus sp, Stanleya, Aster venustus, and Oryzopsisj species, from the Envi database, were used to map for the presence of radioactive minerals in the study area. The use of supervised classification method identified the following vegetation; big berry Manzanita, big sagebrush, Mormon tea, pynon pine, specifically as associated with radioactive minerals. The classified image was finally validated using existing radiometric data of the study area. In conclusion, this study demonstrated the usefulness of applications of remote sensing to map minerals in general, for application in mineral exploration.