Abstract: This study addresses the demand for risk management of flood and drought disasters and reveals the complex network characteristics and resilience-driving mechanisms in water resource systems. Specifically, a five-dimensional, i.e., 'meteorology-hydrology-socioeconomic-ecology-engineering' water resource system resilience evaluation index is developed, which is applied to the Fenhe River basin, acting as the research object. Furthermore, this work quantitatively analyzes the interaction intensity between systems and their resilience-driving mechanisms by establishing a causality network of subsystems to clarify the logical relationships among elements and comprehensively employing the coupling degree model and the geographical detector method. The research findings are: ① The water resource system exhibits distinct complex network characteristics, with high-intensity coupling relationships among its subsystems. Among them, the coupling degree between meteorological subsystem and hydrological subsystem is the highest (0.933). ② In the series of 2010—2022, annual precipitation, surface water resources, urbanization level, forest cover and per capita water supply were the dominant drivers of water resource systems resilience in the Fenhe River basin. Among them, the annual precipitation driver is the strongest (0.812). ③ The interactive effects among different driving factors are consistently more significant than the individual effects of single factors. Notably, the interactions among various influencing factors have become increasingly complex, with multifactorial synergies becoming more pronounced. Exploring the complex network characteristics of water resource systems and revealing the resilience-driving mechanisms in water resource systems are of great significance for reducing disaster risk, optimizing the spatial and temporal pattern of water resources allocation and supporting the high development quality of the basin.