Modelo predictivo de la dinámica espacio-temporal del Bosque Altoandino en el Municipio de Gutiérrez, Cundinamarca en un escenario de 20 años

Abstract

The Andean High Forest ecosystem plays an important role in the supply of water resources due to its ability to capture, store and release water; they are the habitat of more than 32 endemic species of angiosperms and the refuge of a great variety of wild species. Despite the foregoing, less than 5% of the original forests persist in the country, being mainly restricted to fragments of different sizes and degrees of isolation, conditioning their prevalence over time. Geographic Information Systems (GIS) have become a valuable evaluation tool for decision-making in environmental studies, such as predictive analyzes in terms of biodiversity and their potential impact on ecosystems, analyzing change and potential of land cover transition. This study aims to build a predictive model that allows us to know in a 20-year scenario the spatio-temporal dynamics of the high Andean forest, taking as a case study the forests present in the municipality of Gutiérrez Cundinamarca, due to its proximity to the Sumapaz National Natural Park, area protected especially for the conservation of water resources. To achieve the predictive model, a historical analysis was performed quantifying the forest area for the years 1988, 1997, 2003, 2015 and 2018, interpreted from LANDSAT satellite images. The TerrSet - Land Change Modeler (LCM) software was used for the construction of the model, with the determining variables of change being the distances to the built area, to the areas in conflict of use, to the roads, and to the areas with mining titles. The model revealed that the proximity to roads, built areas and areas in conflict of use, determine the permanence in time of the Andean High Forest in the municipality, showing that areas with loss of forest cover are concentrated in the area Western of the municipality, unlike the eastern area that retains coverage over time, even showing areas with regeneration, favored by its slopes and the distance to the determining variables.