EVALUACION DEL USO DE LA TIERRA MEDIANTE UN ANALISIS MULTITEMPORAL UTILIZANDO IMÁGENES SATELITALES EN LA RESERVA DEL HOMBRE Y

The project operates in three areas of the Amazonian Andes: Caquetá, Colombia, Madre de Dios, Peru, and Napo, Ecuador. In Ecuador, the project focuses on the territory of the Kichwa community of Wamaní, where a mosaic of land uses, primarily associated with agroforestry systems, crops, and livestock, predominate. As a result, the NZDZ activities in Wamaní focus on promoting sustainable forest management practices, improving agroforestry and agricultural systems, among others. To estimate the impact of these practices on carbon dynamics at the landscape scale in Wamaní, there is a need for tools that can estimate the carbon content in the territory of the community. This document systematizes the generation of a land cover and land use map in Wamaní, which serves as a base input for estimating carbon contents in the community's territory. The objectives of the study are to elaborate a land cover and land use map (CUT) of the Wamaní community. The specific objectives are to review the most up-to-date and high-resolution remote sensing information available for the study area, generate and validate a CUT legend for Wamaní that represents the best possible thematic detail in relation to different carbon dynamics, and to elaborate a land use and land cover map of the Wamaní community, considering easily identifiable units in the available remote sensing information and representing different biomass classes. The study area is the Wamaní community, located in the Hatún Sumaku parish, Archidona canton, Napo province, Ecuador. The community is approximately 30 km from the intersection of the Hollín-Loreto road with the Baeza-Tena road. The community's surface area is 5,056.88 hectares, covering an elevation range of 764 to 1,498 meters. According to the Ministry of Environment (MAE, 2013), the predominant ecosystem in the study area is the Montane Evergreen Forest of the Northern-Central Cordillera of the Andes. The main land uses correspond to agricultural and livestock mosaics, with naranjilla (Solanum Quitoense) being the main commercial crop in the area. The methodology used was the visual interpretation of high-resolution remote sensing information available for the study area. The procedures followed were: collection and processing of remote sensing inputs, visual interpretation of land cover and land use patterns, collection of field information, and editing and generation of statistics. The main input used in the generation of the CUT map of Wamaní was the orthophoto at a scale of 1:5,000, which covers the area of the OIII_E2 topographic map at a scale of 1:50,000. This input is part of the SIGTIERRAS program, a national initiative for the collection of cadastral information that included the taking of aerial photographs at a scale of 1:20,000 with a GSD (Ground Sample Distance) of 50 cm for the Ecuadorian Amazon. The orthophoto used in Wamaní was taken in 2010. As a support for the interpretation process, a Landsat 7 ETM+ image acquired on September 18, 2010, corresponding to Path 9, Row 61, was used. The image was downloaded from the USGS server using the Earth Explorer platform. The base cartography at a scale of 1:50,000 for the study area was systematized in vector format. All procedures were implemented in the ArcGIS 10.1 platform. The method of land cover and land use interpretation was visual, delineating homogeneous polygons of land cover and land use on the screen. Automated interpretation methods were not implemented due to the different tones, textures, and radiometric scales for the same land cover class in the images covering the study area, resulting from differences in flight conditions and image processing of the mosaics of orthophotos (Ortiz and Peralvo, 2013). A visual analysis of the orthophoto mosaic allowed identifying that the different flight lines that covered the Wamaní community were taken at different times, generating different shadow patterns in the forest canopy. In the interpretation process, a macro-mapping of forest and non-forest classes was first performed. Based on this, an initial proposal for a thematic legend of CUT was constructed, incorporating land cover classes that presented greater interest from the perspective of carbon dynamics and/or were easier to interpret (Table 1). The proposed legend is hierarchical, with progressively more detailed levels nested in coarse land cover classes. The base used was the national CUT legend generated by the Ministry of Environment of Ecuador in the development of the deforestation baseline (MAE, 2012). The fieldwork calibration of the CUT interpretation was carried out by collecting georeferenced information in the field. GPS points and associated digital photos were taken during