Temporal variation in environmental DNA detections within an urban estuary

Biodiversity in estuarine ecosystems can be difficult to survey on time scales that support effective management, even when conditions for sampling are ideal and monitoring locations are adjacent to populated urban centres. As such, identifying the strengths and weaknesses of multiple survey approaches, as well as if and how to combine these efforts to maximize biotic information, is necessary in these rapidly changing ecosystems. Our study therefore aimed to evaluate the temporal variability of environmental DNA (eDNA) based detections across two sites (Camp Cove and Parsley Bay) in a highly urbanized estuary (Sydney Harbour, Australia) to assess how seasonality and tidal cycles influenced biodiversity. To this end, we filtered two water samples at each of two sites monthly for one year to isolate eDNA and generate metabarcoding data from several taxonomic subgroups (bacteria, basal metazoans, algae and aquatic plants, fish, macroinvertebrates) using 15 metabarcoding assays. Biotic detections resulting from these data at the species level were compared to historical collections, citizen science observations, and local knowledge sourced from these same sites across much broader timescales (decades to centuries). Based on the temporally replicated eDNA metabarcoding detections, we found that richness was typically highest in summer and autumn, and lowest in winter across most taxonomic subgroups, particularly when water sampling occurred at low tide in the winter. We also observed significant differences in community structure between seasons and sites for all taxonomic subgroups. When comparing validated species-level checklists generated using all survey approaches, we found that each method detected different fractions of the fish and macroinvertebrate fauna, and there were proportionally fewer detections with eDNA, noting the differences in collection timescales, life history stage bias, and gaps in available DNA reference libraries. Our study highlights the role that eDNA metabarcoding can play in providing biotic information as part of long-term monitoring programs in modified estuaries, particularly in identifying environmental conditions that can influence these detections, and reinforces that there is value in integrating other biodiversity sources to inform lower taxonomic ranks.