Desarrollo e implementación de propuesta metodológica para la determinación de zonas con cobertura vegetal deficiente bajo el marco del Plan de Gestión Integral de Residuos Sólidos del Distrito Capital
Fecha
Autores
Autor corporativo
Título de la revista
ISSN de la revista
Título del volumen
Editor
Compartir
Director
Altmetric
Resumen
This document compiles the main aspects of the methodology for the determination of areas with poor plant cover based on a supervised classification of high-resolution aerial images. The process begins with the determination of the public areas targeted by the program of lawn cutting and tree pruning and sweeping and cleaning of roads and public areas. Once the zones have been identified, a first supervised classification is performed using the Google Earth Engine application, based on the selection of representative areas of different categories of coverages present in the zones. Subsequently, a second supervised classification is carried out on areas classified as plant coverings and areas defined as parks by the IDRD, Areas representative of the different states of the vegetation cover are selected and categorized from the value of digital or pixel levels to determine if it is abundant, medium or uncovered soil. Once the algorithm is structured and tested on the images of the capital city, a raster image is obtained whose pixel values reflect a value related to the defined categories (abundant vegetation cover, average vegetation cover, bare soil and shadows). The accuracy of the classification is assessed by validation data. Subsequently, a geospatial analysis is performed on the areas classified as medium vegetation cover and uncovered soil, which reflect a deficiency of vegetation cover. In addition to the classification, the relationship between areas with poor vegetation cover and relevant factors, such as territorial dynamics, frequency of sweeping and road cleaning, and concentration of air pollutants, is evaluated, as the particulate material of 10 microns (PM10). The combination of remote sensing techniques, supervised classification and geospatial analysis provides a comprehensive approach for the identification of specific areas requiring intervention to improve vegetation coverage in the capital city’s public space and contribute to greening processes for the mitigation of atmospheric pollutants in addition to contributing to the identification of areas with differential needs in terms of the frequency of tree sweeping and pruning activity