A spatial pattern analysis of gold mining-related forest change in the Madre de Dios region, Peru

Over the last twenty years, the Madre de Dios region of Peru, part of the western Amazon, has experienced a rapid increase of gold mining activities along with associated environmental and public health impacts. This study uses Spectral Mixture Analysis (SMA) in a cloud-computing platform to map deforestation within and outside indigenous territories, protected areas, and legal mining concessions within the Madre de Dios. The study area focus on the western part of the Tambopata National Reserve, the La Pampa region, and the Malinowski River. Landsat surface reflectance data were used to create annual cloud-free composites spanning 2013 to 2017. Spectral unmixing was performed to identify patterns of land cover change for each year. High-resolution data from Planet was used to validate the final maps. Preliminary results show large areas of gold mining-related forest loss, especially in La Pampa and in the buffer zones of the protected areas. Areas of small-scale and artisanal gold mining also appear to be expanding in the Kotsimba Native Community that encompasses part of the Malinowski River, which is also located in the buffer zone of protected areas. Another concern is that the gold mining activity is not well restricted to the legal mining concession areas. These results reiterate the applicability of a cloud-based platform not only for land use land cover change detection, but also for accessing and processing big data; the importance of monitoring gold mining activities in the Peruvian Amazon; and suggests the use of SMA as a reliable classification approach.