Impacts of socio-environmental policy mix on mitigating agricultural abandonment: An empirical agent-based modeling

• Published in: Ecological informatics Vol. 80; p. 102491
• Main Authors: Estacio, Ian, Sianipar, Corinthias P.M., Onitsuka, Kenichiro, Hoshino, Satoshi
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.05.2024
• Subjects: Farm succession; Genetic algorithm; Geographic information system; Logistic regression; Open-source software; Patternoriented modeling; Logistic regression; Open-source software; Patternoriented modeling; Geographic information system; Genetic algorithm; Farm succession
• ISSN: 1574-9541
• DOI: 10.1016/j.ecoinf.2024.102491

The complexity of socio-ecological systems in agricultural landscapes has been a subject of previous agent-based modeling studies in order to understand the impacts of agricultural policies. However, there is still a lack of models that incorporate the processes of farm succession within farm households and how these lead to agricultural abandonment. This study aims to simulate the effects of socio-environmental policy mixes on the mitigation of agricultural abandonment in the heritage landscape of Ifugao rice terraces, Philippines by developing an empirically-grounded Agent-Based Model (ABM). This ABM utilizes spatially explicit data and a logistic model to model how socio-ecological processes such as changing environmental conditions, cultivation of land, and succession of farmlands emerge to spatial patterns of paddy fields. Model validation showed that simulated paddy field maps achieved a minimum Fuzzy Kappa statistic of 0.7780, thus the model was deemed suitable for prediction of future scenarios. Simulation of the impacts of policy mixes from 2020 to 2050 showed that the provision of aid in restoring eroded terraces is effective in mitigating agricultural abandonment, preventing almost half of the agricultural abandonment compared to the Business-as-usual (BAU) scenario. Meanwhile, promoting the heritage value of the terraces to the youth and provision of monthly subsidies to farm owners did not exhibit significant mitigation of agricultural abandonment, preventing only 3% of the total abandonment compared to the BAU scenario. The findings suggest that policies directly affecting the farmlands will have a quicker mitigating effect on abandonment than policies affecting the farm owners or children that may take time to manifest. •Simulated the effects of policy mixes on the mitigation of agricultural abandonment.•Developed an ABM that incorporates the behavior of potential successors of farmlands.•Devised a framework for simulating accurate land cover maps using non-spatial agents.•Adopted pattern-oriented modeling through a Genetic Algorithm.•Policies that motivate farmland succession have low mitigating effect on abandonment.