AbstractIn river systems worldwide, land cover changes have been identified as major drivers of biodiversity change. Quantifying how terrestrial land cover impacts riverine diversity requires local biodiversity assessments. In this study, we investigated the association of terrestrial land cover and the corresponding riverine fish species communities using eDNA‐metabarcoding in the Chinese Shaying River basin. This basin is home to about 37 million people and is largely dominated by a mix of intense agriculture and urban areas, creating a relatively homogeneous, intensely used landscape. We investigated whether the homogeneous landscape is mirrored in the composition and structure of fish communities in the river network. We sampled eDNA in spring and fall of 2018, amplified it with a primer designed for local fish species and used operational taxonomic units (OTU) assigned to fish as proxy for diversity. Furthermore, we used redundancy analysis, general linear models, and distance decay curves to assess the effects of land cover on fish communities. We found that the Shaying River showed relatively high basin‐wide richness (63 OTU) and seasonal differences in local richness, but limited community differentiation. Variations in alpha‐ and beta‐diversity, measured as local OTU richness and pairwise distance‐decay across the basin were low. Redundancy analysis showed only a weak association between observed aquatic communities and their terrestrial surroundings in a 10 km buffer upstream. The lack of community differentiation assessed by eDNA metabarcoding reflects the homogeneous and intense land‐use in this basin.