Graduation Date

Summer 2021

Document Type



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

Committee Chair Name

Andrew Kinziger

Committee Chair Affiliation

HSU Faculty or Staff

Second Committee Member Name

Nicholas Som

Second Committee Member Affiliation

HSU Faculty or Staff

Third Committee Member Name

Andre Buchheister

Third Committee Member Affiliation

HSU Faculty or Staff

Subject Categories



Evaluating abundance of juvenile salmonids is critical to conservation and management. Current abundance estimation involves use of rotary screw traps and mark-recapture studies. Use of environmental DNA (eDNA) in water samples offers a noninvasive and less expensive approach that may potentially improve or eventually replace traditional monitoring. Here I evaluate the utility of eDNA to predict weekly abundance estimates of outmigrating Chinook salmon (Oncorhynchus tshawytscha) smolts in the Klamath River, California. A total of 15 water samples were collected per week over the 17-week smolt outmigration in both 2019 and 2020. Chinook salmon eDNA concentration in each water sample was determined using a species-specific quantitative PCR assay. I compared the response of traditional abundance estimates from a rotary screw trap to the weekly eDNA rate, a metric corrected for variable flow (stream flow (cfs) x eDNA concentration). The best fit model included eDNA rate and temperature and explained 76% of the variation in weekly abundance. In an assessment of eDNA methods, I found that filtering 1-liter of water with 0.45 µm filters captured more eDNA than filtering 2-liters with 3 µm polycarbonate filters and that surface grab samples yielded more eDNA on average than 24-hour composite samples taken using an automated sampling device. A resampling analysis revealed that collecting more days per week and taking more water samples per day both resulted in improved deviance explained; however, if resources are limited, collecting more days per week provided greater improvements than taking more water samples per day. These results suggest that eDNA may potentially be a useful predictor of weekly abundance of out-migrating Chinook salmon smolts. However, when estimates of abundance were less than 13,500 smolts per week, the concentration of eDNA was too low to accurately estimate concentrations. Furthermore, the majority of eDNA measurements for this study were below the threshold determined for reliable quantification with defined precision. The study system evaluated here, where Chinook salmon smolt abundances ranged from 10,000 to over 1 million per week, may have provided an ideal setting for evaluating the utility of using eDNA concentrations to predict abundance. However, to utilize eDNA for standardized monitoring, further improvements of eDNA methods are needed for estimating smolt abundance.

Citation Style

Transactions of the American Fisheries Society