Details

Related name

• Princeton University. Department of Geosciences [Browse]

Summary note

• Over the past decade, fisheries management efforts have placed increased emphasis on ecosystem-based management, where the interactions between a target stock species and its physical and biological environment are considered in addition to sustainability of the stock itself. At the same time, global-scale climate models that historically focused only on physical and biogeochemical variables are increasingly incorporating biological variables. With these shifts, the historical separation between climate modeling and fisheries modeling is closing, with increased interest in the concept of end-to-end models, i.e. models that incorporate dynamics from physics to top predators.
• In this dissertation, I develop a modeling framework that fully couples a one-dimensional physical mixed layer model, a biogeochemical model, and an upper trophic level fisheries food web model. I present a thorough description of the model itself, as well as an ensemble-based parameterization process that allows the model to incorporate the high level of uncertainty associated with many upper trophic level predator-prey processes. Through a series of model architecture experiments, I demonstrate that the use of a consistent functional response for all predator-prey interactions, as well as the use of density-dependent mortality rates for planktonic functional groups, are important factors in reproducing annual and seasonal observations.
• Following the development and validation of the end-to-end ecosystem model, I use the model to simulate the response of an ecosystem to a bottom-up perturbation, namely an increase in net primary production due to alleviation of micronutrient limitation. I also look at the impact of non-predatory mortality, one of the least-constrained model processes, on the energy flow through the system. We find that the relative changes in production at higher trophic levels are amplified under density-independent non-predatory mortality assumptions but damped under density-dependent assumptions. However, the high parameter uncertainty masks this effect in predicted values for most functional groups.
• Overall, the model developed in this dissertation addresses a growing need to thoroughly diagnose the behavior of, and quantify the uncertainty associated with, complex ecosystem models that bridge physical, biological, and socio-economic boundaries. Such detailed dynamical characterization of model behavior is essential before ecosystem models can be applied to management applications.

Notes

• Source: Dissertation Abstracts International, Volume: 74-04, Section: B.
• Includes supplementary digital materials.
• Sarmiento, Jorge L., advisor
• Stock, Charles A., committee member
• Ward, Bess B., committee member
• Medvigy, David M., committee member

Dissertation note

Ph.D. Princeton University 2012

In

Dissertation Abstracts International 74-04B

ISBN

9781267784070

OCLC

841237791

Statement on responsible collection description

Princeton University Library aims to describe library materials in a manner that is respectful to the individuals and communities who create, use, and are represented in the collections we manage. Read more...