Journal of Animal Ecology · 2021 · 49 citations · 64 references
Life SciencesEcologyZoologyFreshwater ConservationFreshwater EcologyFreshwater ZoologyAnimal EcologyBiodiversityEnvironmental Dna SurveyingFreshwater BiodiversitySpecies DiversitySpecies RichnessEnvironmental Dna (Edna)Environmental Dna MetabarcodingLine Transect SurveysProbability Of Detection AnalysisReplicate SamplingSpecies Abundance AnalysisSpecies Abundance AssessmentSpecies Abundance Estimation
In some situations, environmental DNA (eDNA) metabarcoding is a more accurate tool for measuring species richness of a taxon than conventional methods. Whether this tool can reliably estimate the abundance of a taxon remains unresolved. We examined the reliability of metabarcoding for measuring anuran diversity compared to a commonly used traditional line transect method (TLTM) through the replicate sampling of three visits across 71 waterbodies (ponds or reservoirs) in Liuheng, China. We also investigated the relative contributions of species-specific characteristics and the physiochemical properties of a waterbody on the relative read count across species and waterbodies. We found that eDNA metabarcoding had a higher detection probability for each of seven anuran species found in the sampling region than TLTM. Furthermore, the relative read count estimated by metabarcoding was positively correlated with the density or relative density of individuals identified with the TLTM across waterbodies for every species. Species-specific characteristics of anurans, such as density, relative density, body mass, biomass and relative biomass, accounted for substantial variations in the read count across species and waterbodies, while physiochemical factors, including pH, temperature, water volume, vegetation and elevation, had little effect on the read count. Our results based on robust sampling suggest that metabarcoding enables more reliable and efficient measurements of anuran occurrence at a large scale during a short-term survey (within 15 days) than that obtained by the TLTM, and offers an alternative tool for quantifications of anuran abundance. Density or biomass is better and more reliable indicator of anuran abundance associated with read count than relative density or relative biomass.
64
R: A Language and Environment for Statistical Computing
R Core Team · 2000 · 353K citations · Full text
Natural Sciences, Computer Science, Statistical Computing +10
Model Selection and Multimodel Inference
Technometrics · 2003 · 5.6K citations