Zooplankton as indicators of coastal water quality: a comparative study of PlanktonScope imaging and DNA metabarcoding of seawater and mesh-screened samples

Zooplankton serve as a vital link between primary producers and higher consumers and act as key indicators of aquatic environmental health. Traditional zooplankton monitoring methods are often time-consuming and labour-intensive, while eDNA metabarcoding and automated imaging systems offer promising alternatives. However, few studies have compared these methods, particularly regarding their sensitivity to small-scale spatial changes in zooplankton communities. Here, we compared the performance of PlanktonScope-an imaging system enabling in situ monitoring and identification of dominant plankton taxa-with eDNA metabarcoding of water and mesh-screened samples along a 20-km pollution gradient in Hong Kong. All three methods consistently detected a spatial gradient in the relative abundance of the dinoflagellate Noctiluca scintillans, jellyfish, and copepods. However, eDNA metabarcoding, especially from net samples, produced very few jellyfish reads, likely due to the fragility of their bodies. Within the eDNA results, Temora sp. And Parvocalanus crassirostris were dominant in the inner harbour, while Centropages typicus and N. scintillans prevailed in the outer harbour. Turbidity and orthophosphate phosphorus significantly influenced the abundance of several key taxa. Overall, our findings show that water- and net-based eDNA approaches provide comparable information on zooplankton diversity, capturing a broader range of taxa at finer taxonomic resolution, while PlanktonScope delivers high-resolution spatial and temporal data. Given the considerably lower workload involved, we recommend using automated imaging for the rapid assessment and monitoring of zooplankton taxa, and water-based eDNA metabarcoding for a deeper understanding of coastal zooplankton community dynamics.