Agent-based assessment of land management strategies
Land use in southern Amazonia has undergone rapid changes in recent years, especially in Central Mato Grosso where favorable soybean prices have led to both massive deforestation and expansion of the soybean area into former grazing lands. Soybeans have been promoted by the Brazilian Government as part of its efforts to increase biodiesel production. With the pavement of the Cuiabá-Santarém Highway, the soybean area is expected to increase by another 25%, which will have mixed effects on the overall balance of greenhouse gas (GHG) emissions, the profitability of farming systems, and the sustainability of ecosystem functions and services. This project will explore options to adapt land management to climate change, considering the soybean value chain and overall carbon balance (including the mitigation potential through increased bioenergy supply). The project will take an agro-economic perspective and analyze incentives for farm enterprises to adopt GHG reducing land management strategies. Finally, it will assess the trade-offs between various land uses and policy interventions in terms of GHG reduction, profitability and ecological sustainability.
To achieve these objectives, the project will use the MP-MAS software (standing for mathematical programming based multi-agent system) that the German applicant started developing in 1995, building on the pioneering work of Balman. MP-MAS uses mathematical programming to replicate the actual decision-making of farmers and other land users. Each real-world farm household is represented by a single agent in the model; that is, there is a one-to-one correspondence between real-world farmers and modeled agents. MP-MAS is spatially explicit and employs a cell-based data representation where each grid cell corresponds to one farm plot held by a single landowner. Sub-models of, for example, water run-off and crop growth are linked to this cell-based spatial framework. Since MP-MAS is specified at a very fine resolution, it offers potential for interactive validation and simulation together with farmers and other local stakeholders.
Prof. Dr. Thomas Berger
Dr. Marcus Vinicius Alves Finco
Dr. Matthias Siebold
Institute for Agricultural Economics and Social Sciences in the Tropics & Subtropics, Universität Hohenheim