Unveiling deep-sea benthic biodiversity in the northern South China Sea through environmental DNA metabarcoding

The northern South China Sea hosts highly heterogeneous deep-sea habitats, dotted with numerous active cold seeps and associated chemosynthetic ecosystems, while the Qiongdongnan Basin represents one of the most important natural gas hydrate accumulation regions in this area. Benthic fauna are core components sustaining deep-sea ecosystem stability and function, yet the biodiversity and community structure of benthic assemblages in this hydrate-rich basin remain poorly documented, creating a critical knowledge gap that hinders the ecological risk assessment and sustainable exploitation of local gas hydrate resources. Using environmental DNA (eDNA) metabarcoding, this study systematically investigated benthic biodiversity, community structure, and underlying assembly mechanisms across three representative habitat types in the Qiongdongnan Basin: cold seeps (CSs), gas hydrate stable zones (GHSZs), and non-gas hydrate stable zones (Non-GHSZs). A total of 334 benthic faunal species belonging to 275 genera, 167 families, 78 orders, 29 classes, and 19 phyla were recovered across all samples. Benthic genus richness peaked in GHSZs (191 genera), followed by CSs (184 genera), and was lowest in Non-GHSZs (176 genera). Community composition exhibited clear habitat-specific partitioning: GHSZ assemblages were dominated by Nematoda, Annelida, and Nemertea; Non-GHSZ assemblages were dominated by Nematoda, Echinodermata, and Annelida; and CS assemblages were dominated by Annelida, Echinodermata, and Mollusca. PERMANOVA confirmed significant overall differences in benthic community composition among the three habitats, with pairwise tests identifying CSs as the primary driver of inter-habitat dissimilarity. Integrated analyses using neutral community models and null models further revealed that deterministic processes. particularly variable selection, overwhelmingly governed benthic community assembly, while stochastic processes such as ecological drift and homogenizing dispersal played only a minor role. Environmental driver analysis indicated that benthic community variation in non-seep habitats (GHSZs and Non-GHSZs) was primarily modulated by sediment grain-size parameters and bottom-water salinity, whereas CS community structure was strongly correlated with sediment total organic carbon, sand content, and local meiobenthos density. This study provides a systematic eDNA-based assessment of benthic biodiversity across contrasting deep-sea habitats in the hydrate-bearing Qiongdongnan Basin and contributes a regional biodiversity baseline to support future ecological conservation, environmental impact assessment, and sustainable natural gas hydrate exploration and exploitation.