Mining deforestation in Madre de Dios – Peru for years 2000 and 2017

• Autores: Cerón C., Lasmit, Miranda S., Natalia, Rubin-de-Celis L., Ethel
• Formato: artículo
• Fecha de Publicación: 2021
• Institución: Universidad Nacional Agraria La Molina
• Repositorio: Revistas - Universidad Nacional Agraria La Molina
• Lenguaje: español
• OAI Identifier: oai:revistas.lamolina.edu.pe:article/1748
• Enlace del recurso: https://revistas.lamolina.edu.pe/index.php/acu/article/view/1748
• Nivel de acceso: acceso abierto
• Materia: Minería aurífera; Landsat; Árboles de decisión; Categorías territoriales; Amazonía peruana.; Gold mining; Decision trees algorithms; Territorial categories; Peruvian Amazon

Since the 1980s gold mining in Madre de Dios, Perú has been the main deforestation driver generating severe socioeconomic and ecological impacts. The present research quantifies this activity in 2000 and 2017 by decision tree algorithms applied on a supervised classification of satellite images from LANDSAT 5 and LANDSAT 8 sensors. Results reveal an increase of 47 657,43 ha of deforestation due to gold mining between 2000 and 2017. Categorized areas showed an increase of 42,52% (35 183,45 ha), with the greatest impact on the native communities, concessions for reforestation and buffer zones of Protected Natural Areas. It is worth mentioning that there were 1 139,77 ha by 2000 that presented mining activity and were recovered by 2017. Validation by experts confirmed up to 93% of the detections with Kappa index. Decision tree methodology is direct, with quick processes and accurate results.