Environmental DNA-Based Detection and Molecular Monitoring of Biodiversity in Freshwater and Terrestrial Ecosystems

Abstract

Environmental DNA (eDNA) metabarcoding has emerged as a powerful approach for biodiversity assessment because DNA released by organisms can be recovered from environmental matrices without direct observation or capture. However, eDNA-based biodiversity estimates are affected by imperfect detection at multiple stages, including environmental DNA capture, amplification, sequencing, and spatial representativeness. These uncertainties are particularly important when sampling resources are limited and must be divided between independent environmental samples and molecular replication. The present study developed a Monte Carlo simulation framework to quantify how this allocation affects species detection, occupancy estimation, false-positive accumulation, and biodiversity richness in freshwater and terrestrial ecosystems. A hierarchical detection structure was implemented in which true taxon occupancy was followed by environmental DNA capture and repeated PCR detection. Four occupancy levels (0.05, 0.10, 0.25, and 0.50), three field-level DNA capture probabilities for each ecosystem, three PCR detection probabilities, four molecular false-positive rates, and five sampling allocations were evaluated under a fixed budget of 120 PCR reactions. The sampling strategies were 40 environmental samples × 3 PCR replicates, 30 × 4, 20 × 6, 15 × 8, and 10 × 12. Each factorial scenario was simulated for 500 independent Monte Carlo realizations involving 100 taxa. Across 1,440 scenario–strategy combinations, increasing environmental replication consistently improved regional detection probability and reduced loss of observed richness, whereas increasing PCR replication primarily improved per-sample detection probability. Under a low terrestrial DNA-capture scenario (pF = 0.10; pP = 0.70), regional detection of an occupied taxon increased from 0.6513 with 10 × 12 sampling to 0.9833 with 40 × 3 sampling. In a representative freshwater scenario (pF = 0.40; pP = 0.80), regional detection remained very high across strategies, but observed richness declined as the number of independent environmental samples decreased. False-positive risk accumulated rapidly with the total number of PCR reactions: at a per-reaction false-positive probability of 0.001, the probability of at least one false-positive reaction among 120 reactions was 0.1131, increasing to 0.4520 at α = 0.005. The findings demonstrate that environmental and molecular replication address different components of eDNA uncertainty and should not be treated as interchangeable. Under a fixed molecular budget, allocation toward independent environmental samples substantially improved regional biodiversity recovery, particularly under low field-level DNA capture.