Development and application of environmental DNA surveillance for the threatened crucian carp (Carassius carassius)

Abstract

The crucian carp (Carassius carassius) is one of few fish species associated with small ponds in the UK. These populations contain genetic diversity not found in Europe and are important to conservation efforts for the species which has declined across its range in Europe. Detection and monitoring of extant crucian carp populations are crucial for conservation success. Environmental DNA (eDNA) analysis could be very useful in this respect as a rapid, cost‐efficient monitoring tool. We developed a species‐specific quantitative polymerase chain reaction (qPCR) assay for eDNA surveillance of crucian carp to enable non‐invasive, large‐scale distribution monitoring. We compared fyke netting and eDNA analysis at ponds with (n = 10) and without (n = 10) crucian carp for presence–absence detection. We examined biotic (crucian carp density represented by catch‐per‐unit‐effort [CPUE] estimate) and abiotic influences on eDNA detection probability using a hierarchical occupancy model, and eDNA quantification using a mixed‐effects model. eDNA analysis achieved 90% detection for crucian carp (n = 10), failing in only one pond where presence was known. CPUE estimate and conductivity had positive and negative influences on eDNA detection probability in qPCR replicates respectively. Similarly, conductivity had a negative effect on DNA copy number, whereas copy number increased with CPUE estimate. Our results demonstrate that eDNA analysis could enable detection of crucian carp populations in ponds and benefit ongoing conservation efforts, but imperfect species detection in relation to biotic and abiotic factors and eDNA workflow requires further investigation. Nonetheless, we have established an eDNA framework for the crucian carp as well as sources of imperfect detection which future investigations can build upon.

False‐negative detections from environmental DNA collected in the presence of large numbers of killer whales (Orcinus orca)

“…We support current recommendations that advise pilot studies be performed on new systems/organisms before implementing eDNA monitoring (Hansen, Bekkevold, Clausen, & Nielsen, 2018;Harper et al, 2018), and furthermore, we encourage the publication of unsuccessful eDNA studies to better inform the eDNA research community, reduce financial and time losses due to ineffective study designs, and thereby aid the continued success of eDNA applications in future studies. Undoubtedly, eDNA has great potential in the context of cetacean monitoring and management, but as a complimentary additional tool rather than an outright replacement of tried and tested techniques.…”

“…We collected between 900 and 1,080 ml at each site but sampling a larger volume of water may have improved our detection probability (Harper et al, 2018;Hunter, Ferrante, Meigs-Friend, & Ulmer, 2019;Schultz & Lance, 2015). Foote et al (2012) also highlighted that small sample volume may have affected their ability to successfully detect harbor porpoises in seawater samples collected around acoustic dataloggers, with just one sample from eight sites amplifying porpoise DNA despite positive acoustic detections at four of the sites.…”

“…Values well above or below 100% could indicate inhibition, sub-optimal primers or reaction conditions, pipetting errors, or formation of non-specific products or primer dimers (Collins et al, 2018). The probability of eDNA detection also depends on the number of samples, volume of water collected, timing of sampling (e.g., breeding/spawning season), sample concentration, preservation methods, number of PCR replicates, and amplification methodologies used (Alberdi et al, 2018;Harper et al, 2018;Schultz & Lance, 2015;Spens et al, 2017;Stewart, 2019).…”

“…Much of the published eDNA literature highlights the use and applications of eDNA as an important conservation tool for the monitoring of animal populations and provides compelling evidence for eDNA analysis as a potential replacement for traditional genetic sampling methods, for example, physical handling, biopsying and tagging of individuals (Deiner et al, 2017;Harper et al, 2018;Minamoto et al, 2017;Stewart, Ma, Zheng, & Zhao, 2017;Thomsen & Willerslev, 2015). While there have been many reviews and studies on best practices for eDNA studies (Alberdi et al, 2018;Spens et al, 2017) and on the influence of abiotic and biotic factors on the persistence of eDNA in aquatic systems (Barnes et al, 2014;Coble et al, 2018), very little of the eDNA published literature contain studies with false-negative results for target DNA, and thus, there is a gap in the literature.…”

A probabilistic model for designing and assessing the performance of eDNA sampling protocols

“…Environmental DNA (eDNA) sampling, the collection and detection of a target species' DNA in water samples, is an emerging method for monitoring the spread of aquatic invasive species (Dejean et al, 2012;Jerde, Mahon, Chadderton, & Lodge, 2011;Takahara, Minamoto, & Doi, 2013). One of the potential advantages of eDNA sampling compared to conventional survey techniques, such as electrofishing and netting, is its superior species detection sensitivity (Dejean et al, 2012;Dougherty et al, 2016;Harper et al, 2019;Jerde et al, 2011;Simmons et al, 2016;Smart, Tingley, Weeks, van Rooyen, & McCarthy, 2015;Takahara et al, 2013;Wilcox et al, 2016;Wilson et al, 2014). However, despite its high sensitivity, eDNA sampling can still suffer from false negative errors.…”