Comprehensive acquisition of aquatic organism data is fundamental for the effective conservation and restoration of aquatic ecosystems. However, efficient monitoring of aquatic biodiversity dynamics in large river systems remains highly challenging. Environmental DNA (eDNA) technology, as an emerging monitoring approach, offers advantages of rapidity and high efficiency. Nevertheless, conventional cross-sectional small-volume eDNA sampling methods exhibit clear limitations in detecting rare fish species. To address this issue, the present study developed a novel mobile large-volume eDNA sampling method, which substantially increases the water volume filtered per sample through the use of capsule filtration and integrates a mobile sampling strategy to improve the efficiency and representativeness of eDNA metabarcoding surveys. In October 2022, comparative surveys were conducted at two representative cross-sections of the lower Yangtze River to evaluate differences in fish diversity detection between the mobile large-volume method and the conventional small-volume cross-sectional method. All eDNA samples were amplified and sequenced using the Tele02 primer set. The results demonstrated that: (1) Each large-volume mobile sample detected an average of 38 fish species, representing a 216.7% increase compared with the small-volume cross-sectional method (12 species); (2) The mobile large-volume method exhibited higher accuracy in biodiversity detection and greater consistency among parallel replicates; (3) The mobile large-volume sampling method detected 24 fish species of primary concern, including rare and endangered species listed in the IUCN Red List of Threatened Species, fish species listed in the Catalogue of National Key Protected Economic Aquatic Animals and Plants of China, and Chinese endemic species. This represents a 50% increase compared with the 16 species detected by the cross-sectional small-volume sampling method, with 14 species shared between the two approaches. Overall, the study confirms that the mobile large-volume eDNA sampling method demonstrates significant advantages in species detection rate, detection precision and stability, and the identification of fish species of primary conservation concern in large river systems.