Skip to main content
2023 ESA Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back

Latebreaking Poster

Session: Conservation Management (Latebreaking)

LB 11-107 - CANCELLED - Scale mismatch between salmon recovery plans and drivers of population dynamics

Thursday, August 10, 2023
5:00 PM – 6:30 PM PDT
Location: ESA Exhibit Hall
Understanding the biological, environmental, and human factors that shape population trajectories for threatened and endangered species is a global priority. Yet species are embedded in complex social-ecological systems in which the number of interacting variables that mediate population dynamics can make the problem of designing effective conservation interventions intractable. The challenge of uncovering key population drivers is especially acute for species like chinook salmon (Oncorhynchus tshawytscha) whose life history, food web interactions, and importance to humans place them at the nexus of myriad forces acting on their population dynamics. Moreover, neither salmon population dynamics nor the forces shaping them are uniform in space or time. In the face of such complexity, traditional modeling techniques often struggle to accurately capture dynamics due to (1) uncertainty regarding relationships among model components and (2) causal variables whose effects vary through time. Traditional models are particularly ill-equipped to characterize interactions in nonlinear systems where the strength of interactions are state-dependent, a hallmark of complex, natural systems. Here, to elucidate the environmental, biological, and human variables factors shaping Snake River Spring-run chinook salmon (SRSS), we take an equation-free, empirical dynamic modeling approach, avoiding a priori assumptions regarding system interactions and allowing population dynamics and the identity of important drivers to emerge from time series. Specifically, we employ convergent cross-mapping techniques (CCM) to establish which of the potential drivers evince a causal relationship with salmon recruitment, as shown by ability to recover driver time series from salmon recruitment data. We then estimate interaction strength of causal variables on SRSS recruitment in a pairwise (univariate) manner using the partial differentials from the best-performing multivariate models. Importantly, we estimate interaction strengths at each point in the time series, illuminating how the effects of each driver have changed through time and across gradients of environmental conditions. We find that, when considering the largest spatial scales (basin-wide), only ocean conditions emerge as important drivers. However, our subbasin- and stock-scale analyses revealed strong, but time-varying effects of hatcheries and marine mammal predation, both of which were masked at the basin-scale. Our results highlight critical differences in the effects of species interactions and human activities on individual salmon population segments and underline the importance of managing recovery at scales that match population responses.