The Colombian Andes faces severe anthropogenic pressures from deforestation, agricultural expansion, mining, and urban development. Given its status as one of the world’s biodiversity hotspot, land cover monitoring for effective conservation strategies and sustainable development planning is essential. While early research relied on coarse or medium-resolution satellite imagery for limited temporal coverage, recent initiatives like MapBiomas Colombia have improved national-scale mapping capabilities. However, mapping the complex spectral-temporal patterns of heterogeneous tropical mountain environments benefits from advanced methods that can track long-term surface trends. This study presents comprehensive annual land cover maps for the Colombian Andes and Sierra Nevada of Santa Marta spanning 1997-2024, representing the longest and most detailed temporal analysis of this critical region to date. We employed the Continuous Change Detection and Classification (CCDC) algorithm applied to Landsat data, coupled with machine learning models trained on a custom dataset designed to address existing limitations in temporal depth and spatial detail. Our time series approach leverages spectral-temporal signatures to distinguish land cover types with similar spectral characteristics but distinct temporal behaviors, providing a unique method for monitoring forest dynamics, habitat changes, and anthropogenic impacts across the Colombian Andes.
Arévalo et al. (Wed,) studied this question.