Research on multiple improvement pathways of agricultural resilience in China from the perspective of dynamic configuration

【Objective】 This study aimed to elucidate the importance of China’s agricultural resilience level and its influencing factors， and discuss the multiple configuration pathways to enhance agricultural resilience through dynamic fuzzy set qualitative comparative analysis （fsQCA）， so as to provide references for local governmental departments to improve agricultural resilience level and promote high-quality agricultural development. 【Method】 Based on system theory perspective， an evaluation index system for agricultural resilience level was established from four dimensions： agricultural economic resilience， agricultural social resilience， agricultural production resilience， and agricultural ecological resilience. The time-space range entropy weight method was adopted to measure the agricultural resilience levels of 31 provincial administrative regions in China （Hong Kong， Macao， and Taiwan were excluded from the study due to lack of data）. The random forest regression was used to identify the importance of influencing factors for agricultural resilience， and dynamic fsQCA was employed to explore the multiple configuration pathways for achieving high agricultural resi-lience level. 【Result】 The agricultural resilience level of China increased from 0.226 in 2011 to 0.320 in 2022， showing an increasing trend year by year， with an average annual growth rate of 3.191%. In addition， the agricultural resilience level of China presented a distribution pattern of eastern region （0.299）>central region （0.296）>northeastern region （0.288）>western region （0.232）， with their corresponding annual growth rates of 2.656%， 3.133%， 4.018%， and 2.434%， respectively. Whether on the basis of the increase in mean square error or the increase in node purity， the importance ranking of the influencing factors for agricultural resilience was basically similar： the leading influencing factors of importance were rural road density， rural internet penetration rate， science and technology investment level， industrial structure optimization， urban-rural income balance index， and agricultural planting structure. There were four conditional configurations for achieving the high agricultural resilience level， which can be categorized into three typical pathways based on core conditional compositions： agro-affluent and grain-stabilized archetype， grain-stabilized and infrastructure-enhanced archetype， and agro-affluent and production-diversified archetype. Different regions of China exhibited certain configurational preferences in the process of achieving the high agricultural resilience level： the eastern region preferred agro-affluent and production-diversified archetype； the central region preferred the grain-stabilized and infrastructure-enhanced archetype； northeastern region preferred the agro-affluent and grain-stabilized archetype； the coverage of configurations in the western region exhibited little differentiation， so the pathways to achieving the high agricultural resilience level presented diverse characteristics. 【Suggestion】（1） Cross-regional collaborative mechanisms should be innovated to drive balanced development of agricultural resilience； （2） Investment in core elements should be coordinated to cultivate endogenous driving forces for resilience generation； （3） Region-specific strategies should be implemented to address spatial development imbalances.