The crew of the research vessel Chasina gets ready to drop an acoustic telemetry receiver 300 feet down into Puget Sound. The device will record tagged steelhead as they swim out of their spawning rivers. Credit: Ashley Ahearn
OPB radio recently provided a look into the research conducted by Megan Moore (NWFSC – NOAA) and colleagues on an early marine survival project. If you didn’t catch the program, it describes the sources and locations of steelhead smolt mortality upon leaving their natal river. The researchers are tagging wild and hatchery smolts, then tracking their movements (and survival) through Hood Canal in Washington to study where mortality is occurring and if differences between hatchery and wild fish are present.
Their most recent steelhead stalker project had the following 4 objectives:
- Get a baseline survival estimate for two wild populations;
- determine if conservation hatchery smolts had similar survival rates compared to wild smolts;
- evaluate if wild and conservation hatchery smolts from the same population have similar behavioral traits, and;
- locate hot spots of mortality during smolt migration.
The study was repeated over three years (2008-2010). Wild smolts were captured from two Hood Canal rivers and tagged with VEMCO acoustic tags. These tags emit an active signal (think along the lines of sonar) that is picked up by receivers positioned in migration corridors. Each tag has a unique ID code so individual fish can be tracked through their path of migration. The conservation hatchery fish were offspring of wild spawners and collected as recently fertilized embryos from the river (no artificial spawning involved).
Two different hatcheries were used to rear the fish. One hatchery (Duckabush population), was low-density with circular tanks and the second (Skokomish population), was higher-density (within conservation program guidelines) using raceways for rearing. The same tags were implanted in the hatchery fish and wild fish so there wasn’t a difference in tag effects.
To track the fish, four lines of receivers were deployed during the study. The first was at the river mouth of each river smolts were tagged in. The second was at the Hood Canal bridge, a third was made across Admiralty Inlet and the final array was spanning the Juan De Fuca Strait.
Program MARK, using a Cormack Jolly Seber survival model, was used to determine if any of the following five factors influenced survival;
- population (river differences);
- rearing type (hatchery vs wild);
- hatchery (differences between two hatcheries used);
- Skokomish (combined Duckabush wild and hatchery vs Skokomish hatchery), and;
- Duckabush (combined Skokomish wild and hatchery vs Duckabush).
Three covariates (length, condition factor and release date) were included in all models as well. If you want more details on the analysis check out the pdf (I dare you…).
For survival probabilities of each population in each section of the migration click here. Wild fish had the highest survival probability, but the Duckabush hatchery smolts were not statistically different from wild fish. The Skokomish hatchery fish had lower survival in every migration segment except from the Hood Canal bridge to Admiralty Inlet. The Hood Canal bridge to Admiralty Inlet segment was the area of highest mortality for all populations. Release date has a slight positive effect on survival, meaning hatchery fish released later in the season had a slightly higher probability of survival. There were no differences between wild and hatchery populations in freshwater or saltwater migration behavior.
Let’s put this one in perspective. Some hatchery fish had indistinguishable survival rates from wild fish…score one for the conservation hatchery program! The other hatchery population (the Skokomish) had lower survival probabilities, but they were raised at higher densities in a raceway environment. Potentially higher-rearing density, or something unique to the raceway, had an effect on post-release survival.
A mortality “hot spot” was found in the migration pathway. Between Hood Canal bridge and Admiralty Inlet, all smolts had mortality rates up to 15 times higher than any other migration segment. The authors suggested that the subsurface bridge pontoons could be inhibiting the surface-oriented smolts during their migration. And just in the past year Moore, et al. have published another paper on the effects of the bridge on smolt survival. Stay tuned for that article in my next post.
Neil Thompson is a P.h.D. student at Oregon State University studying the drivers of domestication selection in hatchery culture of Pacific salmon and steelhead. Find out more info on the lab where I work here.