Scenarios of land use and land cover change for NW Amazonia: Impact on forest intactness

Changes in land use and land cover (LULC) are important drivers of environmental changes in the tropics. Amazônia Legal has showed some declines in deforestation rates, but NW Amazonia has become one of the most recent/new hotspots of forest loss in the Amazon basin. Regional models of change are tools that support the analysis of causes and consequences in the dynamics of land use and are critical in the construction of scenarios for planning the future of the territory and conserving Amazonian forests. The neural networks implemented in the Terrset Land Change Modeler (LCM) module were used to predict and analyze the recent (2007–2016) and future business as usual (BAU2030) and two alternative scenarios for 2030 (PAST50_2030 AND TECH_2030) of the Colombian Amazon and prognostics for intact forests of the future. We used 5 main categories (forests, pastures, secondary vegetation, fragmented forests, and others) and grouped the main transitions of categories into three submodels of change (conversion to pastures, degradation and regeneration). The simulation foresees a substantial loss of forests in the business as usual scenario of 7.92% for 2030 (−3,387,898 ha), followed by a projected increase in pastures of 2,012,087 ha (52.5%) respectively. On the contrary the alternative scenarios show more reduced loss of forest for 2030 of 1,998,299 ha (4.6%) for the PAST50_2030 scenario and a 612,989 ha (1.4%) in the TECH_2030 scenario, accompanied of an increase of only 362,966 ha of pastures in the first case and a reduction of pastures of 316,705 ha in the TECH_2030 case. The most worrying projection is the reduction of intact forest from an average size of forest fragments of 5724 ha in 2016 to 1668 ha (BAU2030), 1744 ha (PAST50_2030) and a minor reduction to 2397 ha for the TECH_2030 scenario. There is also an increased isolation and a reduction of total forest core that only in the TECH_2030 scenario is less pronounced. The protection of those intact patches is key to the future of the NW Amazonia, the reduction in the rate of conversion to pastures PAST50_2030 indicates an opportunity to have a less dramatic loss of forest intact patches but the TECH_2030 scenario is where those areas remain less affected and connectivity will be less affected. These findings signal the urgent need to find mechanisms to implement effective planning in the region in order to prevent uncontrolled expansion, slow deforestation and fragmentation rates and to ensure the integrity of the protected areas and the connectivity of such key landscapes.