Spatiotemporal eDNA Monitoring of Marine Biodiversity in a Hyperurbanised Coastal Environment

ABSTRACT Environmental DNA (eDNA) provides a powerful means of monitoring biodiversity, offering high taxonomic resolution and broad spatial coverage beyond traditional methods. To characterize ecological communities, it is critical to understand shifts in species composition through time to potentially differentiate resident from transient species in the studied habitats. This study used eDNA metabarcoding to examine temporal and spatial patterns of α‐ and β‐diversity across three distinct habitat types (sandy, rocky, and mangrove) at four coastal sites in Singapore over 1 year. We targeted invertebrates using the cytochrome oxidase subunit I (COI) gene and vertebrates using the 16S rRNA gene. We recorded lower diversity at nature reserves, which harbor more rare species than unprotected habitats. β‐diversity differed significantly by site and time for both markers, though β‐dispersion generally remained consistent over time within sites for both invertebrate and vertebrate communities. The difference in marine metazoan communities was driven by high spatial and temporal turnover without strong directional trends across Singapore's coastal sites. These patterns reflect distinct, cohesive communities with limited seasonality, characteristic of equatorial climates. However, certain taxa showed monsoon‐associated distributions, except in mangrove habitats. Importantly, we suggest more mid‐ to long‐term surveys to elucidate the community of resident species. Our findings highlight the value of using eDNA methods to identify dynamic biodiversity patterns and support its use in long‐term ecological monitoring and conservation planning.