Road networks facilitate invasive dominance by squeezing native mesopredators into niche margins

Transport infrastructure is expanding rapidly, creating escalating biodiversity impacts that demand management solutions beyond conventional road engineering. However, road mitigation is still commonly designed around barrier and connectivity paradigms, with limited attention to how roads may restructure biotic interactions and thereby amplify invasion impacts. Here, we quantify how road networks reshape competitive space use between invasive free-ranging cats (Felis catus) and native mesopredators across China, and translate the results into actionable mitigation priorities. Using a nationwide dataset of >12,000 verified vertebrate roadkill records, we reconstructed guild-specific "mortality niches" through a spatially explicit machine-learning framework that integrates prey suitability, human modification, and landscape structure. We show that roads act as selective filters rather than indiscriminate hazards. Resource-rich road corridors attract both guilds, yet space use is strongly asymmetric: free-ranging cats concentrate in the most human-modified, prey-suitable portions of the road network, whereas native mesopredators, despite tracking the same prey gradient, are under-represented in this human-modified, cat-associated core and remain instead in the less-modified margins of these corridors. Native mesopredators are therefore not excluded from prey-suitable road environments, but experience a realised-niche "squeeze" along the human-modification axis. These findings have direct implications for environmental management and infrastructure planning. Mitigation focused solely on structural connectivity (e.g., crossing structures) may be insufficient-and can be counterproductive-if it ignores interaction landscapes that favour synanthropic invaders. We recommend an integrated management approach that (i) reduces anthropogenic food subsidies along verges (e.g., refuse and carcasses) to lower predator attraction, (ii) prioritises targeted free-ranging cat control in high-biodiversity transport corridors, and (iii) incorporates invasion-competition risk surfaces when siting and managing crossing structures to avoid creating predator-enhanced hotspots.