Diversity of frontier processes in Amazonian subnational jurisdictions: Frontier metrics reveal major patterns of human–nature interactions

International audience

English. Deforestation, degradation and regrowth of the tropical forests of the Amazon clearly alter forest cover. These changes in space and over time generate diverse landscape use (archetypes). Identifying the differences and similarities between units and associated changes in forest cover due to deforestation, degradation and regrowth is crucial for context-specific management and planning. Methods for quantitatively characterizing this complexity across large agricultural frontiers are still underdeveloped. This article presents a new method to study the archetypes resulting from forest cover changes in Amazonian subnational jurisdictions by integrating spatial and temporal analysis techniques for deforestation, degradation and regrowth. The weighted k-means approach was linked to nine metrics covering the period 1990-2021 in three subnational jurisdictions of the Brazilian and Colombian Amazon: 1. baseline forest, 2. percentage forest loss, 3. remaining forest, 4. speed of forest loss, 5. active deforestation, 6. percentage forest degradation, 7. speed of forest degradation, 8. active degradation, and 9. percentage regrowth. Four optimal archetypes were chosen using k-means classification: a. consolidated frontier, b. vulnerable frontier, c. past gradual frontier and d. rampant frontier. Consolidated frontiers are areas with high and long term deforestation. Vulnerable frontiers have high forest cover but show signs of previous or recent deforestation and degradation. Past gradual and rampant frontiers show medium to high levels of deforestation associated with degradation. The importance and spatial distribution of each archetype varies at a territorial scale depending on colonization history and on the drivers of deforestation and degradation. This approach provides valuable insights for stakeholder to target interventions and policies adapted to each archetype, for example, payment for ecosystem services, command and control policies, land tenure regulations, land restoration strategies or land use intensification.