Limestone mining area mapping and assessment at the Cement Industrial area using Spectral Index and Machine Learning Algorithms

Abstract

Abstract Limestone mining contributes significantly to the Gross Domestic Product (GDP) of any country but it comes with some adverse impacts on the environment. The objective of this study is to determine the spatial distribution area of limestone mines using remote sensing, spectral index, and machine learning algorithms and compare their area estimation with industrial data for the financial year 2019. The study area includes a limestone mining area, of approximately 2226.16 ha with an excavation mining area of 487.10 ha at the Yerraguntla cement industrial region, YSR Kadapa district, Andhra Pradesh, India. In this study, we used the normalized vegetation index (NDVI), iterative self organizing data analysis technique (ISODATA), K-Nearest Neighbors (KNN), and random forest (RF) algorithms to analyze multispectral Sentinel-2A satellite data in QGIS 3.18 software tool. The RF classifier estimated a limestone mine area of 379.57 ha with best user accuracy (UA) 97.25% and producer accuracy (PA) 99.18% with a kappa coefficient value of 0.957. The mine area estimated 417.47 ha with UA of 98.99% and PA of 99.10% and kappa value 0.947 of the KNN method, The NDVI method estimated 469.92 ha with UA of 93.63% and PA of 92.04% and kappa value 0.685. This study confirmed that RF classifier well performed in classification with overall accuracy (OA) of 95.79% than KNN (OA of 94.78%), and NDVI (OA of 79.84%) classifiers, and ISODATA is poor in classification with OA of 64.16%. This study supports environmentally sustainable decisions, eco-friendly mine planning and monitoring for limestone mine owners and environmental engineers.