Science activity

The Sacramento River in California’s Central Valley has been highly modified over the past 150 years due to mining, urbanization, and impoundment/diversion of river flow to provide water for municipal, industrial, and agricultural needs. Land use changes combined with high levels of harvest have been accompanied by drastic declines in native salmon populations, including the once abundant Chinook salmon (Oncorhynchus tshawytscha). Further, the region has been subject to the introduction and widespread establishment of non-native fish species, some of which are predators of juvenile salmon. Of the four historic ecotypes of Chinook salmon (fall, late-fall, winter, and spring runs), winter- and spring-runs have been most impacted and are currently listed as endangered and threatened respectively under the US Endangered Species act. Past research has illustrated how smoltification of juvenile salmon and outmigration from freshwater to the ocean is a time of increased mortality, and reduced survival at this life stage can impact the number of reproducing adults returning to the system in subsequent years. While these studies have provided valuable information on how habitat and environmental conditions experienced by migrating Chinook salmon smolts can affect survival, they have primarily focused on individual ecotypes during the portion of the year where downstream migrations occur. However, variation in smolt size and migration timing among ecotypes can expose migrating fish to differing environmental conditions and levels of exposure to predation, which can present distinct risks for outmigration survival. To identify the areas and environmental conditions which have the greatest relative impact on juvenile survival for each ecotype, this project will use over ten years of data (2012-2022) from acoustically tagged smolts representing all four Chinook salmon ecotypes in the Sacramento River/Central Valley. Combining these data will increase sample size relative to previous studies, the range of environmental conditions (e.g., temperature, flow, and predator abundance) modeled, the range of fish sizes, and thus, the statistical power of our analyses. We hypothesize that each ecotype will have different factors that will be the primary drivers of mortality experienced during outmigration. To test our hypotheses, we will implement Cormack-Jolly-Seber (CJS) mark-recapture models to estimate both the probability of survival through reaches of the Sacramento River delineated by acoustic receivers, and the detection probability in each reach. Survival will be modeled as a function of individual, release group, reach-specific, and time-varying covariates. Further, to examine the relative impact of predation on smolt survival, we will include an additional covariate representing predator-prey encounter rates using the Mean Free-path Length model. Finally, model selection will be applied to a series of CJS models to assess the relative impact of each covariate on smolt survival for each of the four Chinook ecotypes.