Survival and Migration Dynamics of Juvenile Chinook Salmon in the Sacramento-San Joaquin River Delta

Population dynamics of migrating fish depend on how they use space over time.  Populations may traverse different migratory routes, and survival rates may vary among routes due to differences in migration timing, food resources, environmental conditions, etc.  Thus, understanding spatial variation in survival can provide important insights about population dynamics.  In California, the Sacramento-San Joaquin River Delta is a complex network of natural and man-made channels through which juvenile salmon must navigate on their journey to the ocean.  Juvenile salmon disperse among the Delta’s complex channel network where they are subject to channel-specific processes that affect their rate of migration, vulnerability to predation, feeding success, growth rates, and ultimately, survival.  Eventually, alternative migration routes converge at the exit of the Delta and the population once again comes together to migrate through San Francisco Bay.  Here, we describe a statistical model that explicitly links population-level survival to dispersal among routes and survival within routes.  Specifically, we developed a multistate mark-recapture model to estimate 1) survival of juvenile salmon using different migration routes, and 2) the probability of fish using each route.  Since these parameters determine population-level survival, simultaneous estimation of both allowed us to quantify the relative contribution of each migration route to population-level survival.  We applied this model to three years of acoustic telemetry data (2007-2009) of juvenile Chinook salmon emigrating from the Sacramento River.  We identified consistent patterns in survival, with some migration routes exhibiting consistently lower survival than others.  Because of spatial variation in survival, the fraction of fish using each migration route affects population-level survival.  In turn, water management actions influence both movement among, and survival within migration routes.  Our framework can be incorporated directly into population models to understand how water management actions influence specific components of the population and propagate through the life cycle to affect population dynamics.