Answers are grounded in this paper and cite their sources.
Soil environmental DNA (eDNA) metabarcoding is an increasingly used tool for non-invasive plant community monitoring, but how consistently different molecular markers recover ecological patterns remains insufficiently tested. We compared three widely used plant markers (rbcL, trnL, and ITS2) for characterizing plant communities across wetland and upland habitats in two temperate forest locations in South Korea. When analysed separately, rbcL, trnL, and ITS2 each indicated lower Shannon diversity in wetland than in upland soils, and distinct community compositions by habitat type and location. Agreement between markers in community structure was supported by Mantel (Spearman r = 0.52–0.61) and Procrustes (r = 0.87–0.92 on all ordination axes) correlations, with comparable patterns at both locations. Taxonomic outputs nonetheless differed: intragenomic variation within Pinopsida (mainly Pinus) contributed 69% of ITS2 OTUs but only 19% of reads, an artifact of OTU counting that created an apparent habitat difference in OTU richness; ferns were detected almost exclusively with the chloroplast markers; and 45 of 73 genera appeared in only one marker dataset. Indicator taxa aligned with known wetland and upland vegetation. At both locations, one appropriately selected marker captured the principal wetland–upland contrast, while the combined marker results provided broader taxonomic representation and exposed marker-specific biases.
No takes yet. Share an insight, caveat, or question.
Oh et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: