Assessment of fish biodiversity in Korean rivers using the environmental DNA metabarcoding technique

Abstract

Environmental DNA (eDNA) metabarcoding is a cost-effective novel approach to estimate the biodiversity in an ecosystem. We here adopted the MiFish pipeline to know if the system is reliable to estimate fish biodiversity in the Korean rivers. Total 125 unique haplotypes and 73 confirmed fish species were identified from 16 water samples collected from a single survey of four Korean rivers (Hyeongsan, Taehwa, Seomjin, and Nakdong) indicating MiFish pipeline is a useful tool to estimate the fish biodiversity with relatively low cost and labors. However, low 12S sequences of endemic species in the database and low resolution of MiFish region for differentiating several taxa should be upgraded for their wide use. Among the four rivers, the highest species richness was identified in Seomjin river (52 species), followed by Taehwa river (42 species), Hyeongsan river (40 species). Nakdong river (26 species) showed the lowest species richness and endemic species numbers presumably due to its metropolitan location and anthropogenic impacts such as dams or weirs there. We were also able to know that five exotic species (Carassius cuvieri, Cyprinus carpio, Cyprinus megalophthalmus, Lepomis macrochirus, and Micropterus salmoides) are widely distributed in all surveyed rivers, which would be problematic in the Korean river ecosystem. These findings strongly support the idea that the eDNA metabarcoding technique would be one of the cost-effective and scientific tools in the management and conservation of fish resources among Korean rivers. In the second part of this dissertation, we compared the species abundance and seasonal variation for one year surveys in the Suyeong River. For this study, we have collected water samples from August 2017 to June 2018 at the four sampling stations of the Suyeong River, Korea. The MiFish universal primer set was used for eDNA metabarcoding analysis in this river. Here we identified 65 fish haplotypes from four sample collection stations of the Suyeong River, among those haplotypes, the highest 31 were identified from the family Cyprinidae, followed by Gobiidae (8), and the remaining 15 were from the other 11 families. The highest Shannon-Wiener (H') index was found at the station A (2.364), followed by station C (1.186) and station B (1.039), while the lowest was at the station D (0.976). The average Margalef index was also highest at the station A (3.406), followed by station C (3.073), station D (2.462), and the lowest was at the station B (1.963). Among all sampling stations, station A had the highest average species detection rate (17.33), while the lowest (9.33) was at station B. During the study period, the highest average number of species was found in June (24.25), while the lowest (8.5) was in February. We also identified five exotic species (Carassius cuvieri, Cyprinus carpio, Micropterus salmoides, Lepomis macrochirus, and Oreochromis niloticus) in the Suyeong River, and now they are widely distributed in Korean waters. Our findings suggest that eDNA metabarcoding required less time and taxonomic expertise, and it is better to understand fish distribution and biodiversity in rivers/streams than the traditional survey systems. Although MiFish metabarcoding successfully presented fish species inhabiting in Korean rivers, additional sequence data should be supplemented for better result. The accuracy for detection of endemic species, endangered species, invasive species, and fish distribution by eDNA analysis is possible very effectively to supplement the traditional monitoring approaches in different ecosystems. In conclusion, we expect that these findings would provide useful information for the effective management or conservation of fish resources in the inland open water of the Korean peninsula.