Mapping and future prediction of land use and land cover dynamics using Google Earth Engine and an artificial neural network model in the Jucusbamba River Basin, Amazonas (NW-Peru)

• Autores: Zabaleta Santisteban, J.A., Cachay Reynaga, R., Rojas Briceño, N.B., Silva López, J.O., Medina Medina, A.J., Tuesta Trauco, K.M., Rivera Fernandez, A.S., Sánchez Vega, J.A., Silva Melendez, T.B., Grandez Alberca, M.A., Salas López, R., Oliva Cruz, M., Gómez Fernández, Darwin, Barboza, E.
• Formato: artículo
• Fecha de Publicación: 2026
• Institución: Instituto Nacional de Innovación Agraria
• Repositorio: INIA-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.inia.gob.pe:20.500.12955/3200
• Enlace del recurso: http://hdl.handle.net/20.500.12955/3200; http://dx.doi.org/10.15666/aeer/2403_49654993
• Nivel de acceso: acceso abierto
• Materia: Spatial modeling; Modelamiento espacial; Deforestation dynamics; Dinámica de deforestación; Remote sensing classification; Clasificación por teledetección; Sustainable landscape planning; Planificación sostenible del paisaje; Andean-Amazon transition zone; Zona de transición andino-amazónica; https://purl.org/pe-repo/ocde/ford#4.01.04; Utilización de la tierra, Land use; Cobertura de suelos, Land cover; Teledetección, Remote sensing; Deforestación, Deforestation; Sistema de información geográfica, Geographical information systems; Cuencia hidrográfica, Watersheds, Cambio climático, Climate change.

Monitoring land use and land cover (LULC) changes is essential due to its close relationship with ecological processes, land-use planning, and environmental sustainability. In the Jucusbamba River sub-basin (Amazonas, Peru), knowledge of spatial transitions and cover change dynamics remains limited. This study analyzed LULC changes from 1992 to 2022 using Landsat and Sentinel satellite imagery classified with the Random Forest algorithm on the Google Earth Engine (GEE) platform. Additionally, future scenarios for 2037 and 2052 were simulated using the MOLUSCE plugin along with Artificial Neural Networks (ANN). Five main land cover classes were the followings: urban areas, pasture and cropland mosaics, forests, grasslands, and secondary shrub/herbaceous vegetation. Between 1992 and 2022, pasture/cropland mosaics increased by 24.36% and urban areas by 0.76%, while forests and secondary vegetation decreased by 8.89% and 16.25%, respectively. Projections to 2052 indicate further expansion of agricultural (4.74%) and urban (0.07%) land use, along with additional losses in forest cover (-1.47%) and secondary vegetation (-3.35%). The classification achieved an overall accuracy of 89.8% and a Kappa coefficient of 0.86. These findings provide a robust foundation for evidence-based decision-making in land management, ecological zoning, and natural resource conservation within the Andean-Amazonian region.