Neighborhood variations in urban green cover patterns in Bogotá (Colombia) estimated by NICFI-Planet images
Keywords: Urban vegetation, NDVI, Machine learning, K-means, Google Earth Engine, Environmental justice
Abstract. To achieve a more green city in the shortest possible time, joint work between government and society is necessary and this work intend to contribute with scientific basis to support the expansion of urban greens areas in Bogota DC. We analyzed the spatial and the social distribution of urban greens and their variations inside of Bogotá urban areas, between years 2019 and 2023, based on NICFI-Planet images. The Normalized Difference Vegetation Index (NDVI) was calculated for each monthly cloud-free mosaics over Bogota DC. and a NDVI Maximum Value Composite (NDVI-MVC) image was calculated for both years. The NDVI-MVC images were clustered with the k-means algorithm, generate a binary image with built-ups targets and other urban targets (vegetation and shadows). These images were analyzed with socio-economic data to a better understanding about the social distribution of urban greens. Were observed an environmental injustice, where the benefits of green areas of the human health are allowed for people from middle and upper classes.