Environmental DNA Metabarcoding Read Numbers Reveal Consistent Patterns of Microhabitat Use Among Fish Species Within Multiple Lakes and Tidal Creeks

ABSTRACT Environmental DNA (eDNA) metabarcoding has emerged as a powerful tool for biodiversity assessments in aquatic ecosystems, yet its utility to resolve fine‐scale spatial patterns of community structure within well‐connected habitats remains unclear. We investigated whether eDNA metabarcoding can detect microhabitat‐specific community structure by comparing fish assemblages across distinct zones within two ecosystem types: littoral, pelagic, and benthic zones in freshwater reservoirs, and inlets versus main channels in oligohaline tidal creeks. We collected 80 water samples across eight sites in the Virginia Peninsula region in Virginia USA and used mitochondrial 12S metabarcoding (MiFish‐U) to characterize fish communities. While per‐sample species richness did not differ significantly among microhabitats, multivariate analyses revealed distinct compositional differences that aligned with known species ecology. In lakes, microhabitat identity explained 12.9% of compositional variation (PERMANOVA R 2 ), with Lepomis macrochirus strongly associated with littoral zones and Dorosoma cepedianum with benthic‐pelagic areas. Tidal creek systems showed even stronger microhabitat structuring (18.9% of variation), with Fundulus heteroclitus serving as a clear indicator of inlet habitats while species including Brevoortia tyrannus and Morone americana characterized main channels. These findings demonstrate that eDNA maintains sufficient spatial fidelity to capture ecologically meaningful community patterns at scales of tens of meters, despite hydrodynamic mixing processes. Comparative analysis showed that analysis of read numbers identified species‐habitat relationships missed by analysis of presence‐absence data alone. Metabarcoding of eDNA is valuable for habitat‐specific conservation monitoring and restoration assessment even within well‐mixed aquatic habitats, opening new avenues for high‐resolution biodiversity monitoring.