Environmental DNA sampling as a surveillance tool for cane toad Rhinella marina introductions on offshore islands
Containing the spread of established invasive species is critical for minimizing their ecological impact. Effective containment requires sensitive sampling methods capable of detecting new introductions when invaders are at low density. Here we explore whether environmental DNA (eDNA) sampling could be used as a surveillance tool to detect new incursions of aquatic invasive species on offshore islands. We develop an eDNA molecular assay for invasive cane toads (Rhinella marina) in Australia, validate our assay on the mainland, and apply it to an offshore island (Moreton Island) that is a target of ongoing cane toad surveillance. Our eDNA assay correctly identified four mainland sites at which cane toads were observed, as well as a fifth site within 1 km of known populations. Five additional sites outside the cane toad’s current distribution tested negative for cane toad eDNA. Site occupancy detection models indicated that two water samples and three qPCR replicates were sufficient to achieve a cumulate detection probability > 0.95. Applying our eDNA assay to samples from 19 sites on an offshore island over a 2-year period revealed the absence of cane toad eDNA, in line with our current understanding of cane toad distribution. Our results suggest that eDNA sampling could be strategically applied to meet the Australian Commonwealth’s objective of maintaining cane toad-free offshore islands.
KeywordseDNA Containment Detection probability Islands Sensitivity Surveillance
RT and ARW were funded by the Australian Research Council Linkage Scheme (LP140100731). MG would like to thank Sutherland Shire Council and Sydney Metropolitan Local Land Services for funding, and Trent McKenna and George Madani for assistance in the field.
- Commonwealth of Australia (2011) Threat abatement plan for the biological effects, including lethal toxic ingestion, caused by cane toads. Department of Sustainability, Environment, Water, Population and Communities, Canberra, AustraliaGoogle Scholar
- Plummer M (2003) JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. Proc Int Workshop Distrib stat Comput 124:125Google Scholar
- R Core Team (2016) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, AustriaGoogle Scholar
- Su Y-S, Yajima M (2015) R2jags: using R to Run ‘JAGS’. R package version 0.5-7. https://CRAN.R-project.org/package=R2jags