1 Introduction
Large river systems are essential for providing critical foraging, breeding ground, and nursery habitats for a variety of fauna and are considered among the most productive ecosystems worldwide (Wang et al., 2020a; Wang et al., 2021b). Fish, as consumers at high trophic levels in river food webs, represent the sum of a wide range of complex trophic interactions (Wang et al., 2018b). Linking the ecological indicators of fish communities to human interference remains an important goal for river managers (Zou et al., 2020). The distribution, composition and diversity of fish communities are commonly used as proxies to assess an ecosystem’s health and integrity. To appropriately manage and protect aquatic ecosystems, it is essential to develop effective and eco-friendly monitoring approaches to collect field data and obtain biological parameters (Kumar et al., 2022; Shu et al., 2021).
Traditional monitoring of fish diversity has depended largely on census methods such as electrofishing, gill/hoop/seine netting, and dredging/trawling (Wang et al., 2020a; Wang et al., 2019a). However, those methods have always been limited by their low sampling efficiencies, destructiveness to organisms, and strict reliance on taxonomic expertise (Sakata et al., 2020; Zhang et al., 2020). The application of environmental DNA (eDNA) metabarcoding for fish diversity analysis has emerged and offers a new avenue for characterizing fish diversity in river ecosystems (Pont et al., 2018). It provides cost-effective, dependable, rapid and continuous investigations for monitoring and assessing fish diversity, which is crucial for the timely and effective management and conservation of river and estuary ecosystems (Garlapati et al., 2019).
The metabarcoding approach coupled with the use of eDNA is a potentially powerful tool for surveying and assessing aquatic diversity. Numerous studies have demonstrated the utility of eDNA metabarcoding for assessing fish diversity (Rourke et al., 2022). Researchers have successfully applied eDNA metabarcoding to monitor fish diversity in freshwater and seawater samples from different habitats, especially in streams, reservoirs, estuaries and oceans (Civade et al., 2016; Stoeckle et al., 2017; Yao et al., 2022; Zou et al., 2020). The results from these studies have shown that eDNA metabarcoding is a sound biomonitoring tool for use in the conservation and management of aquatic ecosystems (Nguyen et al., 2020). Currently, the application of eDNA metabarcoding to monitor and assess biodiversity is at the forefront of the available methods used by ecologists and conservation scientists (Beng and Corlett, 2020; Bernos et al., 2023).
Many studies have compared eDNA results to those from traditional methods and have shown a correlation between the results from the two approaches (Lacoursière-Roussel et al., 2016; Port et al., 2016). In some studies, eDNA analysis was superior for characterizing fish biodiversity compared to traditional techniques such as electrofishing and hoop netting (Nguyen et al., 2020; Pont et al., 2018). In other studies, the results obtained from eDNA have been comparable to those of traditional methods in which fishes are caught through visual dive surveys and trawling (Port et al., 2016; Zou et al., 2020). Previous studies have shown that eDNA metabarcoding retains a higher diversity of taxa than the traditional method; however, the practical application of eDNA in evaluating the composition and structure of fish communities has been less explored, and whether eDNA could replace traditional monitoring is still unknown.
The river systems of southern China, in a typical subtropical monsoon climate zone, serve as reserves for biodiversity conservation. Due to disproportionate use of coastal wetland resources and intense anthropogenic activities (i.e., drainage, reclamation, and pollution), subtropical river ecosystems (e.g., the Pearl River) have been badly damaged, and their biodiversity and bioresources have seriously declined. To investigate the distribution and composition of fish communities in this area, eDNA metabarcoding studies combined with electrofishing surveys were conducted. In addition, the diversity of fluvial fishes observed by electrofishing with the characterization of eDNA collected concurrently from rivers by metabarcoding was compared. The objectives of our study include 1) using a metabarcoding protocol to assess the eDNA-based composition and diversity of fish communities; 2) analysing the response between eDNA OTU richness and fish amounts (e.g., individual number and biomass); and 3) exploring the application of eDNA in assessing environmental influences on fish diversity.