Graduation Date

Spring 2018

Document Type



Master of Science degree with a major in Natural Resources, option Fisheries

Committee Chair Name

Dr. Andrew Kinziger

Committee Chair Affiliation

HSU Faculty or Staff

Second Committee Member Name

Dr. Andre Buchheister

Second Committee Member Affiliation

HSU Faculty or Staff

Third Committee Member Name

Dr. Darren Ward

Third Committee Member Affiliation

HSU Faculty or Staff

Subject Categories



Rangewide monitoring of fish species is critical for determining status and trends in abundance and distribution; however, implementations of large-scale distribution surveys have generally been constrained by time and cost. This study uses environmental DNA (eDNA) to monitor the presence or absence of two endangered tidewater goby species, the northern tidewater goby (Eucyclogobius newberryi) and the southern tidewater goby (Eucyclogobius kristinae), across their combined geographic range that encompasses the entire California coast (1,350 km).A multi-scale occupancy model designed specifically for eDNA methods was used to account for imperfect detection and to estimate true site occupancy. A total of 209 sites were surveyed in coastal California from Del Norte to San Diego counties between May and September 2016. Among these sites, 12 were dry during the survey and assigned a status of non-detection. Among the 197 sites with water present, a total of 430 water samples were collected, filtered, and tested for the presence/absence of northern and southern tidewater goby, using species-specific quantitative PCR (qPCR) assays. The number of water samples collected per site ranged from one to six. Northern tidewater goby were detected at 81 out of 175 sites and southern tidewater goby were detected at 4 out of 22 sites, resulting in a combined naïve occupancy of 0.43. In contrast, the multi-scale occupancy model estimated site occupancy at 0.55 (95% CRI 0.46–0.64), indicating that tidewater goby were present but not detected at 23 additional sites. Even though eDNA typically has higher detection probabilities than traditional field approaches, these findings indicate that imperfect detection needs to be accounted for in eDNA surveys. Tidewater goby were detected at seven sites where they have previously not been detected or were thought to be extirpated, including one site in San Francisco Bay. As a covariate, salinity was found to have a strong negative effect on qPCR detection probability and tidewater goby DNA availability in a water sample. This finding implies that when using eDNA methods for species detection, more water samples and qPCR replicates might be needed at high salinity sites to achieve the desired level of detection. This study illustrates the power of eDNA for generating point-in-time snapshots of a species’ entire geographic distribution. The distributional information generated herein is critical for management as it will serve as a baseline for determining site occupancy and if tidewater goby are expanding or contracting in the number of sites occupied.

Citation Style