Tonle Sap Lake is the largest freshwater lake in Southeast Asia. As a key ecological system of the Mekong Basin, it sustains highly productive fisheries and rich fish biodiversity across the surrounding floodplain through seasonal flood-pulse dynamics. However, most existing knowledge of fish biodiversity in Tonle Sap Lake is based on traditional capture-based surveys with limited spatial coverage. Despite its ecological and socio-economic importance, lake-wide patterns of fish biodiversity remain insufficiently understood.
In this study, we investigated lake-wide spatio-temporal patterns of fish communities with environmental DNA (eDNA) metabarcoding, which allowed us to explore fish community composition across the entire lake–floodplain system at broader spatial coverage and higher resolution. eDNA samples were collected from more than 200 sampling sites across the lake–floodplain system, including the Tonle Sap River (TSR) and Mekong River (MR). A total of seven surveys were conducted between 2024 and 2025, covering both dry and wet seasons.
Spatial and seasonal patterns in fish communities are analyzed based on species richness and community composition dissimilarity. Hierarchical modeling of species communities was also combined with eDNA-derived presence–absence data for analyzing species-specific responses to seasonality effect and area. We found there is pronounced spatial heterogeneity in fish community composition across the lake–floodplain system. In particular, fish diversity in floodplain habitats were higher than those in the open lake, and stronger community difference was observed at the northern and southern ends of the lake compared to the central region. Seasonal patterns of species richness and assemblage composition were comparatively stable at the community level, whereas species-level analyses indicated that individual fish species exhibited distinct seasonal variation.
Overall, our results revealed that environmental DNA can effectively capture fish community composition in Tonle Sap Lake, complementing traditional monitoring approaches and providing new support for ecological modeling for the whole system.