Using Occupancy Models to Compare Environmental DNA to Traditional Seining Methods for Region-Wide Monitoring of an Endangered Species

Environmental DNA (eDNA) is a rapidly advancing genetic technique used for species monitoring. To date, few studies have applied eDNA techniques at large geographic scales and within a statistical framework that explicitly allows for comparison between eDNA and traditional field approaches. We compared eDNA to traditional seining for monitoring endangered tidewater goby (Eucyclogobius newberryi) across 400 kilometers of northern California’s coastline. At a total of 29 locations we collected paired samples including a traditional seine haul and a water sample for eDNA testing. Each water sample was tested for tidewater goby DNA using 6 replicate qPCR assays specific to tidewater goby. We used a multi-method occupancy modeling framework to estimate method specific detection probabilities. eDNA had a 40% higher detection probability than seining and detected tidewater goby at 19 sites compared to 14 sites with seining methods. eDNA detected tidewater goby at two locations where they have never been detected previously despite previous field surveys and at one population location thought to be locally extirpated. The results indicate that eDNA provides a much more powerful monitoring technique than traditional field sampling and can be used to increase management efficiency across a broad spatial range and within a diversity of habitats.