摘要：Ensuring the long-term sustainability of post-resettlement livelihoods remains a critical challenge in large-scale resettlement programmes, particularly in the context of climate adaptation and rural transformation. Focusing on resettled households in China's Yellow River Floodplain, this study examines the dynamic evolution and optimization of livelihood resilience and strategy choices in Xingfuhe Community, a typical large-scale floodplain resettlement site in Shandong Province. Drawing on survey data from 344 households, this research constructs a multidimensional livelihood resilience framework encompassing buffer capacity, self-organization, and learning ability. A multinomial logistic regression model identifies the key determinants of livelihood strategy choices, which are then integrated into an Agent-Based Model (ABM) to simulate household behavioural dynamics and policy impacts over a five-year period (2024-2029). The results reveal that land quality, skill training, policy awareness, and social trust significantly influence livelihood strategy selection. Simulation outcomes suggest that, within the modelled scenarios, interventions strengthening human capital (particularly skill training) and institutional responsiveness tend to be more effective in promoting livelihood diversification and resilience improvement than infrastructure investment alone. Households show a clear shift toward diversified and parttime livelihood strategies under targeted policy scenarios. This study contributes to the literature by bridging livelihood resilience theory with empirically calibrated dynamic modelling, offering a nuanced understanding of micro-level adaptation processes in resettlement contexts. It provides evidence-based policy insights that may inform livelihood support strategies in China's Yellow River Floodplain Resettlement Programme and similar large-scale relocation programmes in comparable institutional, governance, and biophysical contexts.

关键词：Yellow River floodplain resettlement；Livelihood resilience；Agent-based modelling；Sustainable development

文献来源：HABITAT INTERNATIONAL

WOS链接：https://webofscience.clarivate.cn/wos/alldb/full-record/WOS:001870658000001