Modelling farmland conversion integrating multi-agent systems and resource economics theory

Farmland conversion, i.e. conversion of farmland into non-agricultural construction land, is an unavoidable trend in global economic development(Fazal 2001; Tan et al. 2005; Bugri 2008). In China, which is relatively short of farmland resources and yet in the phase of accelerating urbanization, conversion of farmland is closely related to national food security, ecological security as well as sustainable land resource use(Liu et al. 2005). Therefore, a reasonable allocation of farmland conversion quota is of great importance. Compared with traditional methods such as empirical statistics, systems dynamics and Cellular Automata, Multi-Agent Systems (MAS), integrating macroscopic phenomena and micro decision-making, is more able to interpret the driving mechanisms of LUCC from the view of interaction relations of driving agents(Manson 2006; Li and Liu 2008; Tian et al. 2011). Since MAS is advantageous in the study of LUCC, this research aims to explore such issues as feedback between agents and their environment, and corporation and competition between agents. Based on the decision-making framework of multi agent systems with full consideration of agents’ heterogeneity and resource economics model, a mode for spatio-temporal allocation of farmland conversion quota is developed.