Monitoring on a shoestring: the use of commercial fishing vessels as inexpensive sampling platforms for long-term biological and oceanographic monitoring (Steven J. Barbeaux)
Long-term monitoring is a key component of an ecosystem-based approach to fishery management. Data time series enable the examination of changes in oceanographic and community metrics. Government funding for long-term monitoring of biological and oceanographic processes has dwindled in recent years, while the mandate for this type of information has increased.
If data-driven ecosystem-based management continues to be the goal, then methods for reducing the costs of data collection must be found while data quality is maintained. An example of this type of innovative approach can be found in the Alaska walleye pollock fishery where researchers have teamed with commercial fishers to deploy inexpensive temperature and depth data storage tags on trawl nets. At the same time, data on fish density and distribution are being collected using the fishing vessels' own acoustic systems.
These data are being used to validate oceanographic models, to assess the effects of oceanographic conditions on bycatch in the walleye pollock fishery, monitor the impacts of the fishery on the stock across a wide range of temporal and spatial scales, and evaluate the effects of oceanographic conditions on walleye pollock density and distribution. This project demonstrates a cooperative monitoring program in which researchers work with other sea-going stakeholders to inexpensively collect biological and oceanographic data that can be integrated into a long-term ocean observing system
The hidden story: what are the potential genetic effects of different types of fisheries management?
(Ingrid Spies and André Punt)
Under a precautionary approach to fisheries management, management units should correspond with a single genetic population or stock. Such an approach is intended to preserve individual populations, biodiversity, and the overall resilience of the stock complex. Although genetic population structure has been documented in many marine fish species, no clear method exists to translate this information into a meaningful management strategy.
Here, we simulate marine fish populations with two types of genetic population structure: panmixia and discrete populations. Panmixia occurs when individuals in a population move about freely within their habitat, possibly over a range of hundreds to thousands of miles, and thus breed with other members of the population, as opposed to discrete populations between which interbreeding does not occur.
Fish are partitioned into a linear array of spatial units, and fishing pressure is modeled with effort proportional to the distance from the fishing port. A population study is simulated by randomly selecting 100 individuals from each unit, and standard genetic methods are used to determine where a boundary (or boundaries) should be drawn to form management areas. Two types of management are simulated: combined management (all subpopulations are managed as one unit) and separated management (subpopulations are managed individually with boundaries selected using population genetic methods).
Population dynamics and genetic population structure are then projected under both management plans with annual stock assessments and fishing pressure for 100 years. Performance measures such as total catch and population size are compared under both management scenarios, as well as any genetic population structure changes.
This project is loosely based on the dynamics of Pacific cod in the Bering Sea and Aleutian Islands area of Alaska. Initial results show that fishing can result in loss of genetic diversity and depletion in population size of the more proximate population, and that smaller populations are more susceptible to loss of genetic diversity.
Role of habitat in moderating species distributions and interactions
(Matt Baker, Anne Hollowed, M. Elizabeth Clarke, and Ray Hilborn)
Several mechanisms drive ecosystem structure and stability in marine systems, including competition between species, climate, and fisheries extraction. Many systems subject to perturbation show stability in total biomass and structure, despite shifts in relative species abundance. This suggests sequential replacement related to compensatory dynamics.
We examine species dynamics within and between functional guilds in the eastern Bering Sea to weigh evidence for compensation, resource partitioning, and common forcing via external drivers and examine how habitat governs species interactions. We applied random forests to determine the importance of environmental variables on individual species distributions. We then extended these methods to species assemblages, synthesizing cross-validated coefficient of determination and accuracy importance measures from univariate analyses to quantify compositional turnover along environmental gradients.
These outputs were applied to define distinct regions within large marine ecosystems based on unique aggregations of community composition and physical habitat. We also applied centroid-based clustering methods to time series trends of species abundance to explore evidence for sub-structure in exploited stocks within this system. We are currently integrating these methods to inform approaches to examining relative effects of various drivers of species abundance and community composition through multivariate autoregressive state-space models.
By Anne Hollowed, Steve Barbeaux, Ingrid Spies, and Matt Baker.
