Seasonal trends in eDNA detection and occupancy of bigheaded carps, Wabash River, IN

Abstract

Invasive species may vary in their seasonal distribution and abundance due to environmental conditions such as precipitation and temperature. Bigheaded carps, which include silver and bighead carp, are one such taxon of invasive species that appear to change habitats seasonally. Seasonal changes in bigheaded carp distribution may occur because of hydrological changes, water temperature changes, and spawning activities. Monitoring this seasonal dispersal and migration is important for management to control the population size and spread of the species. We examined if environmental DNA (eDNA) approaches could detect seasonal changes in the occurrence of DNA in water samples and used these approaches to calculate the probability of detection of DNA within samples. To do this, we developed a novel genetic marker that was able to both detect and differentiate bighead and silver carp DNA. We implemented the genetic markers to study the occurrence of bigheaded carps at 3 sites on the Wabash River over the course of a year. These sites were chosen because they exist along a spatial gradient of bigheaded carps densities. Furthermore, we studied the Wabash River because of concerns that carps may be able to use the system to invade the Great Lakes via a connection at Eagle Marsh between the Wabash River's watershed and the Great Lakes' watershed. The Wabash River is a tributary of the Ohio River, which is a tributary of the Mississippi River and the source of bigheaded carps in the Wabash. We used occupancy modeling to estimate both the probability that a water sample contained DNA as well as the probability of detecting the DNA within a sample for both species. We found seasonal trends in the probability of detection and occupancy that varied across sites. These findings demonstrate that eDNA methods can detect seasonal changes in bigheaded carps densities and suggest the amount of DNA present changes seasonally. Both the probability of detection and the probability of a sample containing DNA were generally similar for both species. The site that was farthest upstream and had the lowest carp densities exhibited the strongest seasonal trends for both detection probabilities and sample occupancy probabilities. Furthermore, other observations suggest that carps seasonally leave this site, and we were able to detect this with our eDNA approach. More broadly, our research offers support for using eDNA to detect seasonal trends in abundances of aquatic invasive species.