Integrating habitat based eDNA metabarcoding and conventional surveys to characterize the lake-wide structure of fish assemblages

• eDNA is more effective in detecting fish taxa and traits than traditional methods. • eDNA can be effectively used in all habitat types. • Habitat-area-weighted analyses show congruency between sampling methods. • eDNA provides the most comprehensive, lake level assessment of community structure. • Integration of eDNA metabarcoding into long term fish monitoring is warranted. The combined use of traditional sampling methods and environmental DNA (eDNA) metabarcoding provides deeper insights into the organization of ecological communities at both habitat-specific and ecosystem-wide scales. However, direct comparisons of these methods across diverse habitats, and the extrapolation of results on assemblage structure to the entire ecosystem remain limited. This study examines the congruence between traditional and eDNA-based approaches in characterizing both the taxonomic and functional structure of fish assemblages across multiple habitat types in Lake Balaton, the largest shallow lake in Central Europe, highlighting the strengths and limitations of each method. Fish assemblages were assessed using electrofishing in shoreline habitats (reed, rip-rap, and harbors), gillnetting in offshore areas, and eDNA metabarcoding across all habitat types. At the lake-wide level, only minor differences were observed in species detection, suggesting both approaches provide representative sampling. However, eDNA metabarcoding proved particularly effective in detecting very rare species. Multivariate analyses revealed that eDNA metabarcoding captured a greater number of species and functional traits per sample and yielded a more even distribution across habitats. Notably, both benthic and large-bodied species were consistently overrepresented in eDNA samples compared to traditional survey methods, highlighting the potential of eDNA to detect taxa that are often underrepresented in traditional surveys. Habitat-area-weighted analyses showed strong correlations in species rank abundance between eDNA metabarcoding and traditional methods at the whole-lake scale, as well as in inshore and offshore habitats. However, notable differences also emerged in the relative dominance of certain species. While eDNA-based methods offer distinct advantages, challenges in taxonomic resolution persist, emphasizing the need for further methodological refinement. Overall, our findings support the integration of eDNA metabarcoding into long-term fish community monitoring. Its ability to provide a comprehensive, habitat-independent assessment of taxonomic and functional community structure enhances conservation and management efforts in lake ecosystems.