Integrating an abandoned farmland simulation model (AFSM) using system dynamics and CLUE-S for sustainable agriculture.

Addressing the challenges of achieving United Nations Sustainable Development Goal SDG2 is complex, especially considering the profound impacts of population growth, environmental degradation, and disruptions from the COVID-19 pandemic and armed conflicts on the global food supply system. Unjustifiable farmland abandonment poses a critical obstacle to food security, demanding a comprehensive knowledge framework for sensible decision-making in abandoned farmland reclamation. However, current research lacks systematic solutions. This research seeks to address the challenges of farmland abandonment through a comprehensive knowledge framework. Integrating remote sensing surveillance, mechanism analysis, and scenario projection, the primary objective is to develop the abandoned farmland simulation model (AFSM) by combining abandoned farmland identification, the system dynamics (SD), and CLUE-S. The AFSM facilitates explicit spatiotemporal abandonment simulation, contributing to a deeper understanding of the dynamic evolution and mechanisms involved in farmland abandonment. The AFSM is formulated through a logical sequence of "spatiotemporal data acquisition – quantity prognostication – spatial simulation." The process begins by tracking annual land use changes, identifying spatio-temporal changes in abandoned farmland. Subsequently, the research employs the SD model to establish an elucidative framework for abandonment mechanisms, facilitating the quantitative analysis of factors influencing farmers' decisions regarding farmland abandonment. Finally, the CLUE-S model is utilized to prognosticate the spatial abandonment trend. All modules of the model have passed the precision test and inspection. The AFSM outcomes reveal that the abandonment rate in the city of Jingdezhen fluctuated between 5.03% and 13.21% from 2003 to 2020, averaging 8.37%. The model systematically quantifies the multiple socio-economic variables that impact abandonment decisions. Key parameter groups delineate distinct development scenarios and make predictions. By 2032, the abandonment rate under the scenario of farmland protection is projected to be 10.88%, markedly lower than the inertial development and economic priority scenarios by 15.24% and 13.43%, respectively. The concentration of abandoned farmland is primarily observed in elevated, steep-sloped areas, exhibiting a "climbing mountain and slope" trend. The AFSM emerges as a crucial tool for policymakers, farmland managers, and researchers, empowering them to make well-informed decisions concerning abandoned farmland reclamation. The explicit spatio-temporal simulation, complemented by a clarifying framework for abandonment mechanisms, enables the quantification of the influence of various factors on farmland abandonment decision-making and realizes the coupling of quantitative and spatial simulation under various scenarios. [Display omitted] • Global unjustifiable abandonment threatens food security, necessitating a systematic knowledge framework and solutions. • The abandoned farmland simulation model (AFSM) achieves the coupling of quantitative and spatial simulations. • The quantitative simulation accuracy ranges from 89% to 92%, while the spatial simulation accuracy is 80.75%. • The farmland protection scenario is most conducive to curbing abandonment, but it limits the income growth rate of farmers. • Enhancing the cost-effectiveness of farming and developing new types of agricultural entities can curb abandonment. [ABSTRACT FROM AUTHOR]