Grantee:  Garrison Environmental Consulting

Contact:  Dr. Lance Garrison (drlpgarrison at yahoo.com)

Project Summary:

Traditional fisheries stock assessment models largely ignore the role of ecological processes in the population dynamics of exploited species and typically assume that natural mortality rates due to predation is constant across time.  For species that are both important targets for fisheries and prey items for multiple predators, these approaches often do not accurately capture important dynamics and ignore the connected nature of marine ecosystems.  Multispecies Virtual Population Analysis (MSVPA) is an extension of traditional fisheries stock assessment models that includes a simple feeding model to estimate mortality due to predator-prey interactions in exploited fish communities.  The MSVPA approach was initially developed by an ICES working group during the early 1980s, and it has been applied to the North Sea, the Georges Bank/Gulf of Maine, the Baltic Sea, and the eastern Bering Sea fish communities.  The model is used by fisheries stock assessment scientists and managers to quantify predator-prey interactions, estimate predation and fishery mortality rates, and more effectively model the dynamics of exploited fish stocks.

With support from the Atlantic States Marine Fisheries Commission (ASMFC), an expanded version of the MSVPA model, termed MSVPA-X, has been developed.  The MSVPA-X addresses several known limitations of the original method.  This new approach includes a “tuned” single-species VPA approach, modifies the input and implementation of feeding selectivity parameters, incorporates temperature dependent consumption model, incorporates a growth model to examine the effects of changing food availability, and implements an improved functional feeding response.  The model also includes a forecasting component to allow exploration of multispecies population trajectories under a range of management scenarios and assumptions about future recruitment.  The MSVPA-X model is a proven tool that has already been used to incorporate ecosystem considerations into fishery management on the U.S. east coast.

In this project, I will build on this foundation to both incorporate new capabilities and improve upon the existing implementation and user interface.  First, I will expand the capabilities of MSVPA-X by incorporating a broader range of population assessment models that will allow the inclusion of long-lived species such as sharks or unexploited species such as marine mammals or birds that cannot be directly modeled in the current formulation.  Second, I will improve the capability to explore uncertainty in model projections, derived reference points, and retrospective analyses.  The inclusion of uncertainty is a critical aspect of supporting decision-making in an ecosystem framework given the broad range of potential outcomes of management actions as they cascade through the food web.   Finally, I will improve the user interface for data input, incorporate diagnostic tools, improve error trapping and the specificity of error codes, and thoroughly document the current code.  The end result will be a distributable software package that will allow users across a broad suite of ecosystems to evaluate the relative importance of predator-prey interactions and fishing mortality on the population dynamics, yield, and sustainability of exploited fish species.

Links:

Atlantic States Marine Fisheries Commission, Multispecies Assessments Website
Internal Panel Review of MSVPA-X Model and Documentation (June 2003)

National Marine Fisheries Service, Northeast Fisheries Science Center
42nd Stock Assessment Workshop: Review of MSVPA-X Model