A Bayesian life-cycle model to estimate escapement at maximum sustained yield in salmon based on limited information

Life-cycle models combine several strengths for estimating population parameters and biological reference points of harvested species and are particularly useful for those exhibiting distinct habitat shifts and experiencing contrasting environments. Unfortunately, time series data are often limited to counts of adult abundance and harvest. By incorporating data from other populations and by dynamically linking the life-history stages, Bayesian life-cycle models can be used to estimate stage-specific productivities and capacities as well as abundance of breeders that produce maximum sustained yield (MSY). Using coho salmon (Oncorhynchus kisutch) as our case study, we show that incorporating information on marine survival variability from nearby populations can improve model estimates and affect management parameters such as escapement at MSY. We further show that the expected long-term average yield of a fishery managed for a spawner escapement target that produces MSY strongly depends on the average marine survival. Our results illustrate the usefulness of incorporating information from other sources and highlight the importance of accounting for variation in marine survival when making inferences about the management of Pacific salmon.

The Industrialization of Commercial Fishing, 1930–2016

Nations rapidly industrialized after World War II, sharply increasing the extraction of resources from the natural world. Colonial empires broke up on land after the war, but they were re-created in the oceans. The United States, Japan, and the Soviet Union, as well as the British, Germans, and Spanish, industrialized their fisheries, replacing fleets of small-scale, independent artisanal fishermen with fewer but much larger government-subsidized ships. Nations like South Korea and China, as well as the Eastern Bloc countries of Poland and Bulgaria, also began fishing on an almost unimaginable scale. Countries raced to find new stocks of fish to exploit. As the Cold War deepened, nations sought to negotiate fishery agreements with Third World nations. The conflict over territorial claims led to the development of the Law of the Sea process, starting in 1958, and to the adoption of 200-mile exclusive economic zones (EEZ) in the 1970s. Fishing expanded with the understanding that fish stocks were robust and could withstand high harvest rates. The adoption of maximum sustained yield (MSY) after 1954 as the goal of postwar fishery negotiations assumed that fish had surplus and that scientists could determine how many fish could safely be caught. As fish stocks faltered under the onslaught of industrial fisheries, scientists re-assessed their assumptions about how many fish could be caught, but MSY, although modified, continues to be at the heart of modern fisheries management.

Salmon consumption by Kodiak brown bears (Ursus arctos middendorffi) with ecosystem management implications

The ecological role of large predators in North America continues to spark heated public debate. Although brown bears (Ursus arctos L., 1758) and the salmon (genus Oncorhynchus Suckley, 1861) they feed on have declined in many areas, the Kodiak archipelago is famous for large brown bears and abundant salmon. Salmon have generally been managed for maximum sustained yield in a fisheries sense, but those levels may be well below what is necessary for maximum ecosystem productivity. Consequently, we used stable isotopes and mercury accumulated in hair to estimate intake of salmon by Kodiak brown bears (Ursus arctos middendorffi Merriam, 1896). Salmon intake increased from subadult males (592 ± 325 kg·bear−1·year−1) to adult males (2788 ± 1929 kg·bear−1·year−1) and from subadult females (566 ± 360 kg·bear−1·year−1) to adult females (1364 ± 1261 kg·bear−1·year−1). Intake within each group increased 62% ± 23% as salmon escapement increased from ∼1 500 to ∼14 000 kg·bear−1·year−1. The estimated population of 2300 subadult and adult bears consumed 3.77 ± 0.16 million kg of salmon annually, a mass equal to ∼6% of the combined escapement and commercial harvest (57.6 million kg). Although bears consume a small portion of the total mass of adult salmon, perpetuation of dense populations of large bears requires ecosystem-based management of the meat resources and environments that produce such bears.

Maximum sustained yield: a policy disguised as science

Finley, C. and Oreskes, N. 2013. Maximum sustained yield: a policy disguised as science. – ICES Journal of Marine Science, 70: 245–250. Overfishing is most commonly explained as an example of the tragedy of the commons, where individuals are unable to control their activities, leading to the destruction of the resource they are dependent on. The historical record suggests otherwise. Between1949 and 1958, the US State Department used fisheries science, and especially the concept of maximum sustained yield (MSY) as a political tool to achieve its foreign policy objectives. During the Cold War, the Department thought that if countries were allowed to restrict fishing in their waters, it might lead to restrictions on passage of military vessels. While there has been much criticism of MSY and its failure to conserve fish stocks, there has been little attention paid to the political context in which MSY was adopted.

Comparison Between Maximum Sustained Yield Proxies and Maximum Sustained Yield

Attaining maximum sustained yield (MSY) is a central goal in U.S. fisheries management. To attain MSY, fishing mortality is maintained at FMSY and biomass at BMSY. Replacing FMSY and BMSY by “proxies” for FMSY and BMSY is commonplace. However, these proxies are not equivalent to FMSY and BMSY. The lack of equivalency is an important issue with regard to whether MSY is attained or whether biomass production is wasted. In this paper we study the magnitude of the equivalency. We compare FMSY/BMSY (calculated using the ASPIC toolbox) with the proxy estimates, F40%/B40%, published in GARM III. Our calculations confirm that in general the FMSY/BMSY calculations differ from the GARM III proxy estimates. The proxy estimates generally indicate that the stocks are overfished and are at relatively low biomasses, while the ASPIC estimates generally reflect the opposite: the stocks are not overfished and are at relatively high levels of abundance. In comparing the two approaches, the ASPIC estimates appeared favorable over the proxy estimates because 1) the ASPIC estimates involve only a few parameters in contrast to the many parameters estimated in the proxy approach, 2) “real variance” estimates for the proxy are not available so that it is difficult to evaluate the statistical adequacy of the proxy approach relative to the ASPIC approach, and 3) the proxy approach is based on many components (e.g., growth, stock and recruitment, etc.) that are subject to considerable uncertainty.

Semidiscrete biomass dynamic modeling: an improved approach for assessing fish stock responses to pulsed harvest events

Continuous harvest over an annual period is a common assumption of continuous biomass dynamics models (CBDMs); however, fish are frequently harvested in a discrete manner. We developed semidiscrete biomass dynamics models (SDBDMs) that allow discrete harvest events and evaluated differences between CBDMs and SDBDMs using an equilibrium yield analysis with varying levels of fishing mortality (F). Equilibrium fishery yields for CBDMs and SDBDMS were similar at low fishing mortalities and diverged as F approached and exceeded maximum sustained yield (FMSY). Discrete harvest resulted in lower equilibrium yields at high levels of F relative to continuous harvest. The effect of applying harvest continuously when it was in fact discrete was evaluated by fitting CBDMs and SDBDMs to time series data generated from a hypothetical fish stock undergoing discrete harvest and evaluating parameter estimates bias. Violating the assumption of continuous harvest resulted in biased parameter estimates for CBDM while SDBDM parameter estimates were unbiased. Biased parameter estimates resulted in biased biological reference points derived from CBDMs. Semidiscrete BDMs outperformed continuous BDMs and should be used when harvest is discrete, when the time and magnitude of harvest are known, and when F is greater than FMSY.