Assessment of food web dynamics in relation to biodiversity is crucial to understand the impact of climate change on marine ecosystems. However, due to practical limitations in conventional ecological survey methods, the available information is often insufficient to construct a complete food web structure. Here, we assessed biodiversity and food web structures in Marian Cove, Antarctica—an area undergoing rapid ecological change due to climate change—by analyzing multiple genetic loci from environmental DNA (eDNA) amplicon sequencing data. The biodiversity in Marian Cove was distinguishable between the inner and outer coves in correlation with the physicochemical properties of seawater, influenced by glacial retreat. The size composition of primary producers, represented by diatoms in the outer cove and nano- or picophytoplankton in the inner cove, were identified as core taxa in each food web as revealed by linear discriminant analysis (LDA). Topological analysis further revealed that the food web structure in inner cove was mainly driven by bottom-up regulation due to smaller phytoplankton size classes responding to environmental fluctuations, in contrast to the outer cove, where top-down regulation by predators dominated. Our study provides better understanding on food web dynamics in response to climate change.

eDNA, Marian Cove, food web, climate change, biodiversity