Evaluación de actividades mineras sobre el uso y cobertura de suelo en la Amazonía ecuatoriana

Abstract

CONDITION FOR PUBLICATION OF PROJECT. The Ecuadorian Amazon faces multiple environmental pressures derived from human activities such as oil exploitation, deforestation, and mining. Mining, in particular, induces rapid transformation of land use and land cover (LULC), biodiversity, and the hydrological dynamics of rivers. This study aims to evaluate mining activities on land use and land cover in the Ecuadorian Amazon, for the identification of affected areas in the Zamora and Nangaritza river basins during the 2019-2024 period. The methodology was divided into three stages: i) acquisition and processing of Sentinel-2 satellite imagery in Google Earth Engine (GEE), ii) supervised classification of LULC through the application of the Random Forest algorithm for each scenario, and iii) identification of hotspots through the generation of Kernel density maps. The results revealed a 2.22% increase in mining activity, with the greatest losses occurring in the grassland class. The Kernel density analysis identified a transformation from three nuclei (2019) to a continuous corridor of high concentration of mining activities in 2024. These findings provide critical information for the implementation of control and monitoring policies for mining activities in the region. Keywords: Google Earth Engine, LULC, Mining, HotSpot