Lanterna Rally

Author: Eric R. Buhle

Publisher:

Published: 2018

Total Pages: 35

ISBN-13:

Age- or stage-structured population models, also known as life cycle models, are a mainstay of applied ecology and conservation, particularly in fisheries management. The data available to inform parameters in such models are, however, often limited and variable in quality. Ad-hoc, piecemeal approaches to parameter estimation can lead to biased inference about key processes, such as the strength of density dependence and the magnitude of environmental variability in recruitment. Recent statistical advances have facilitated a more rigorous, comprehensive approach to fitting life cycle models by combining all relevant data into a joint likelihood function. Such integrated population models (IPMs) have been widely applied in marine fisheries stock assessment, but are less familiar in salmonid management. We developed a multipopulation IPM for Pacific salmon (Oncorhynchus spp.) that accounts for spatial and temporal variability in adult recruitment and age structure, the presence of hatchery-origin spawners, and observation error in abundance, age-composition, and hatchery-fraction data. The method is analogous to traditional spawner–recruit modeling based on brood-table reconstruction, but the model is fitted to the “raw” data and distinguishes between process and observation error. We applied the model to 29 populations of spring/summer Chinook salmon in the Snake River and Upper Columbia River Evolutionarily Significant Units (ESUs), and used the estimated parameters and states to simulate the impact of fishery exploitation rate on future abundance and quasi-extinction risk. As expected, predicted abundance declined and quasi-extinction risk increased across a range of fixed harvest rates from 0–0.3. The slope of the decline in abundance, relative to population-specific carrying capacity, was inversely related to intrinsic productivity. Large-scale environmental fluctuations (e.g., ocean conditions and hydrosystem operations, represented by the shared process error) were at least as important as harvest in determining long-term population viability. If future environmental conditions are relatively poor, and especially if they are assumed to have undergone a persistent state shift at some point in the last 60 years, then quasi-extinction risks are dramatically elevated even in the absence of harvest. We see potential for the further development of IPMs (e.g., the inclusion of more detailed stage structure) and their application to salmon conservation problems throughout the Pacific Northwest.