Environmental DNA reveals landscape mosaic of wetland plant communities

Abstract

Globally, floodplain wetlands have been dramatically impacted by changes in water regimes. A key indicator of change in wetland state is riparian and aquatic plant community composition. Plant communities are traditionally assessed by visual monitoring programs that assess community composition. However, results from these monitoring programs may be effected by antecedent conditions that govern plant growth, such as season. In this study we used environmental DNA (eDNA) to assess the spatial composition of wetland vegetation communities from six floodplain complexes along the Murray River, Australia and compared results to those derived from tradition plant surveys. Our analyses of eDNA were conducted targeting two gene regions: 18S RNA and trnL at the species and operational taxonomic unit (OTU) levels. Analysis revealed significant similarity between the spatial patterns derived from the eDNA data and those derived from the traditional monitoring data. Moreover, using the eDNA data, analysis of endemicity revealed the majority of OTUs and species occurred only at a single wetland or wetland complex. Few genera were shared between the traditional survey data set and the molecular data sets. But for those that were shared, detectability varied between gene fragments and among genera, ranging from 11 to 100% in the trnL data to 14.3–100% in the18S data. Our results show that landscape patterns in plant communities can be reliably derived from eDNA analysis, thus providing a cost-effective way of undertaking vegetation surveys. Keywords: Murray river, 18S RNA gene, trnL, Metabarcoding, Landscape ecology