Abstract
As fishery production and habitat quality continues to experience declines, there is a growing need to improve the scientific methodologies used to assess and sustain economically and ecologically important fisheries. This requires a sound understanding of the life histories and population dynamics of each species, and development of a robust framework for population modeling. Realization of the multi-species nature of fisheries has prompted a shift towards ecosystem-based approaches to modeling. To evaluate alternative methods for modeling predator-prey interactions within a physically variable coastal-estuarine ecosystem, a food web of national economic and ecological significance comprised of Atlantic menhaden (Brevoortia tyrannus), striped bass (Morone saxatilis), and bluefish (Pomatomus saltatrix) is considered in this study. Understanding the dynamics of biological communities such as this is challenging and requires the formulation of realistic mathematical models. This should be a stepwise process in which the underlying assumptions, parameter sensitivities, and fundamental behaviors of interacting species dynamics described by relatively "simple" to more "complex" models are delineated and quantified. In this study two alternative multispecies modeling frameworks were utilized to evaluate the dynamic interactions between predator and prey populations, and to understand the influence of fisheries and environmental temperature change on predator-prey and food chain communities. First, relatively "simple," age-independent, predator-prey and food chain models representing generalized, ecological-scale interactions between different trophic groups were developed and analyzed. Sensitivity analyses revealed the relative importance of model parameters and the effect of varying levels of fishing mortality on model dynamics. Overall, the predator-prey and food chain models were shown to be a valuable tool for understanding general patterns in the dynamic behavior of interacting populations. Next, an environment-dependent, age-structured, Atlantic coast spatial dynamic multispecies model was investigated. This more complex model links individual-scale bioenergetic processes controlling growth to ecological-scale rates of natural and predation mortality. Simulations were performed using recent stock assessment estimates of fishery condition and stock sizes to evaluate the nature and magnitude of linkages among menhaden and their key predators, specifically bluefish. This ecosystem model was demonstrated to provide valuable insights into the dynamics of menhaden and bluefish given the underlying dynamics and forcing in the Atlantic States fishery coastal ocean ecosystem. Additionally, the influence of environmental temperature on both modeling frameworks was investigated. This iterative process of model development and analysis advances the current understanding of the species and ecosystem of interest, and ultimately provides an improved basis for multispecies fisheries assessments.