Agent-Based Simulation of Land Use Governance (ABSOLUG) in Tropical Commodity Frontiers

Well-designed land use governance that involves multiple stakeholders is crucial to reducing defor-estation in tropical commodity frontiers. The effectiveness of different governance interventions is difficult to assess due to long implementation times and challenges to conducting real-world experiments. Here we introduce an agent-based simulation of land use governance (ABSOLUG) to examine the interactions among governments, commodity producers, and civil society and assess the impacts of different land use governance approaches on deforestation. The model represents a generic commodity producing landscape in the tropics with a central marketplace and features four groups of agents: largeholders, smallholders, NGOs, and a gov-ernment. The objective of largeholders and smallholders is to generate profits through the production of com-modity crops. Statistical evaluation through local and global sensitivity analyses shows that the model is ro-bust, and few parameters show threshold behaviors. We used three governance scenario to evaluate the model operationally. The hands-off scenario was inspired by high rates of tropical deforestation in the second half of the 20th century, the middle-ground scenario by compliance with international conservation treaties, and the proactive government scenario by a few recent cases of forest transition countries. The hands-off scenario led to quasi-complete deforestation of the landscape at the end of the simulation period, while deforestation in the middle-ground scenario leveled off and then stagnated. In the proactive government scenario, deforestation decreased and eventually stopped in the second half of the simulation period, followed by reforestation. We demonstrate that the ABSOLUG model can simulate key policy interventions and land use trajectories in tropi-cal systems, and as such can be a valuable contribution to the growing literature on environmental policy and governance evaluation.