Estimating Spatial Distribution of Fish Species Derived from Relationships Among Environmental DNA Sampling Data, Satellite-based Spectral Vegetation Indices, and Water Quality in Estuarine Protected Area

An estuarine ecosystem is considered one of the most productive ecosystems, providing critical foraging and breeding grounds. The use of eDNA technology has emerged to detect fish diversity in terms of objectivity and persistence monitoring. However, when we use eDNA technology, results show a list of species and DNA total read in point data at the sampling spot. We tried to expand the point data to area data through a prediction model that uses satellite image data as an environmental variable. The study site, an estuary of Suncheon-si, Korea, is vital for biodiversity and conservation, including the Ramsar wetlands, wetland protection area, and water supply protection area. We conducted eDNA sampling at 10 locations on September 12 and 13, 2020. We selected a metabarcoding with MiFish primers for multispecies detection. We selected the Maxcent for single species prediction and the regression model for multispecies and DNA total read. For environmental variables, we measured water temperature, salinity, and dissolved oxygen simultaneously with eDNA sampling. We derived vegetation-related spectral indices such as NDVI, NDMI, NDWI, and MSAVI using Sentinel-2 imageries in August 2020. The eDNA methods captured 12 orders, 39 families, 92 fish species. The DNA total read was detected at 10 points with a maximum of 661,826 and a minimum of 105,158. Among the environment variables, DEM and NDMI were strongly linearly related to the number of DNA fragments (R square = 0.954). The distribution map described that the upper ten percentile values of the DNA fragments were estimated in the reservoir areas in the upper reaches. Further works will be focused on mapping the distribution model of classified fish species (e.g., marine, freshwater, and brackish) by the time series. These works will show the applicability of combining the eDNA sampling method and remote sensing monitoring for the fish species. This work was conducted with the support of the Korea Environment Industry & Technology Institute (KEITI) through its Urban Ecological Health Promotion Technology Development Project and funded by the Korea Ministry of Environment (MOE) (2019002760001) This work is financialy supported by Korea Ministry of Land, Infrastructure and Transport(MOLIT) as Innovative Talent Education Program for Smart City