Environmental DNA (eDNA) is rapidly becoming a favoured methodology for biodiversity monitoring in aquatic environments due to its ease of application, lack of requirements for taxonomic expertise, favourable cost, high sensitivity, and low ethical and environmental costs. The rise in popularity of eDNA is replacing more traditional sampling methods for species monitoring and in some cases is being promoted as the most efficient tool for everything from biodiversity monitoring to invasive species biosurveillance. eDNA sampling is often described as a low-tech solution to species monitoring and therefore amenable to use by unskilled staff or community groups. Whereas traditional fishing techniques may necessitate multiple fishing methods due to species dependent gear selectivity, eDNA can, in theory, detect all species present within a water body, even those extremely rare or cryptic species difficult to detect by standard methods. Is eDNA therefore a long-awaited Panacea for species monitoring?
The extreme sensitivity of and highly variable detection by eDNA can be problematic. eDNA is not highly localised, is not quantitative, can be plagued by false detections and contamination issues (both in situ and in vitro), can be swamped by excess DNA from non-target sources, and can fail in discriminating closely related species. Understanding some of these potential pitfalls can require more complex technical understanding that may pose challenges for unskilled staff or citizen scientists.
While some potential problems with eDNA, such as assigning reference DNA sequence data to species and applying suitably discriminative primer sets, can potentially be resolved, problems such as localisation of source target DNA, quantitation, accounting for exogenous DNA, contamination of sampling equipment, environmental patchiness of DNA, and cryptic taxonomic resolution present considerable challenges. Until some of these problems are adequately resolved, eDNA might therefore currently be considered something of a Pandora’s Box that has the potential to present more problems than solutions to biodiversity monitoring. It is certainly currently unable to match the type of population information possible with more traditional methods.