Agricultural policies have impacts on land use, the economy, and the environment and their analysis requires disaggregated data at the local level with geographical references. Thus, this study proposes a model for disaggregating agricultural data, which develops a supervised classification of satellite images by using a survey and empirical knowledge. To ensure the consistency with multiple sources of information, a minimum cross-entropy process was used. The proposed model was applied using two supervised classification algorithms and a more informative set of biophysical information.

Bogotá, the rapidly growing center of an emerging economy in the northern part of South America, is located within a biodiversity hotspot in the tropical Andes. The surrounding mountains harbor the ecosystems Páramo and Bosque Altoandino whose high water retention capacity serves as a “natural water tower” for the city’s freshwater supply. Since Bogotá is steadily growing, the city spreads into its peri-urban area, thus threatening its proximal ecosystems.

Smallholder farming constitutes an important but marginalized sector, responsible for most of the world’s agricultural production. This has a significant influence in the land use/cover change process and agrobiodiversity conservation, especially in mountainous regions of the developing world. Thus, the maintenance of sustainable smallholder farming systems represents a key condition for sustainable land management and to safeguard the livelihoods of millions of rural households.