Authors:
Refilwe Moeletsi
1
and
Solomon Tesfamichael
2
Affiliations:
1
University of Johannesburg and Mintek, South Africa
;
2
University of Johannesburg, South Africa
Keyword(s):
Remote Sensing, Land Cover Changes, Granite Quarries, Landsat, Supervised Classification.
Abstract:
Environmental monitoring is an important aspect in sustainable development. The use of remote sensing in the mining industry has evolved significantly and allows for improved mapping and monitoring environmental impacts related to mining activities. The aim of this study was to measure land cover changes caused by granite quarrying activities located between Rustenburg and Brits towns, North West Province, South Africa using Landsat time series data. Landsat data used in the study were acquired in the years 1973, 1986, 1998 and 2015. Each image was classified using supervised classification and change detection was subsequently applied to measure land cover changes. Furthermore, the normalized difference vegetation index (NDVI) was used to highlight the dynamics in vegetation in the quarries. Accuracy assessment of the classification resulted in an overall accuracy and Kappa coefficient of 75% and 0.71, respectively. The results of post –classification change detection revealed a s
ignificant increase of 907.4 ha in granite quarries between 1973 and 2015. The expansion in granite quarries resulted in development of water bodies (2.07 ha) within the quarries. Correspondingly, there were significant losses in vegetation (782.1 ha) and bare land (119 ha). NDVI results showed variability in mean NDVI values within the digitized quarries. The overall mean NDVI values trends showed that most granite quarries had the highest vegetation in 1998, while the least vegetation cover was observed 1986.
(More)