Spatial Modelling of D ynamics of Land Use Land Cover Due to Mining Activities in Taita Taveta County
Taita Taveta County, Kenya’s coastal mineral belt rich in a variety of minerals, faces environmental challenges to resources management and conservation. Widespread small-scale mining activities pose a significant threat to the environment, which has led to changes in Land Use and Land Cover (LULC), therefore adversely affecting the environment. This study aims at spatial modeling of the dynamics of LULC, evaluating the accuracy in different time epochs and detecting changes in the mined and mining areas at a temporal scale. The modeling and analysis was done using spatial-temporal remote sensed data and digital image processing techniques utilizing machine learning algorithms in GIS Software and R Studio. Interpretation of the processed data led to the delineation of LULC categories and classes. It was observed that the mined/ mined and developed areas increased by 19% and 12%, respectively, between 2011 and 2019. Also, the area with vegetation land was decreased by 38%, and waste dumps increased significantly. Normalized differential vegetation index (NDVI) was also done to correlate the state of the healthy vegetation. The overall accuracy of classified images and kappa statistics was 83.393% and 0.7591, respectively. This study revealed the declining nature of the vegetation and the significance of using remotely sensed data to model LULC. The modeling showed that the key drivers for LULC changes resulting in environmental degradation in the study area are iron ore mining and mineral exploration.