High Replicability in Strong Correlation Between Goldfish Abundance, eDNA Detection Probability, and eDNA Concentration in Urban Ponds

ABSTRACT Goldfish ( Carassius auratus ) is a globally invasive species commonly found in urban environments due to releases associated with the aquarium trade. Environmental DNA (eDNA) represents a novel, cost‐efficient, and sensitive technique to indirectly detect aquatic organisms by analyzing residual genetic material in environmental samples. We utilized eDNA to monitor invasive goldfish in 29 urban ponds near Hamilton, Ontario, evaluating both detection sensitivity of eDNA in comparison to conventional fisheries survey methods (electrofishing and seining) and the quantitative relationship between eDNA concentration and population abundance and biomass, determined from population censuses obtained following complete draining of eight ponds. Water samples from each pond were collected to detect and estimate the concentration of goldfish eDNA using a targeted qPCR assay. Samples from a subset of ponds were also collected and processed independently by two different working groups to evaluate the replicability of findings. Estimated pond eDNA concentrations were highly replicable across working groups; individual samples collected from the same sampling station and mean pond concentration estimates exhibited extremely strong correlations ( R 2 = 0.95 and R 2 = 0.99, respectively). Detection limits using eDNA, in comparison to conventional surveys, were low; a single water sample from a pond, for example, would be needed to detect goldfish at a density corresponding to approximately a single fish per 10 min of electrofishing. A single sample was sufficient for detection in all but one (inhabited by a single goldfish) of the nine drained ponds. eDNA concentration in drained ponds was also positively correlated with allometrically scaled mass (mean R 2 = 0.72) and abundance (mean R 2 = 0.71), but not with total biomass (mean R 2 = 0.28). Collectively, we demonstrate that eDNA is a replicable and effective technique to detect and monitor aquatic invasive species in urban ponds and may prove suitable for inclusion in management programs.