Publication Date:
2022-05-25
Description:
Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and Woods Hole Oceanographic Institution September 2017
Description:
Population connectivity is a fundamental process that governs the spatial and temporal
dynamics of marine ecosystems. For many marine species, population connectivity
is driven by dispersal during a planktonic larval phase. The ability to obtain accurate,
affordable, and meaningful estimates of larval dispersal patterns is therefore a
key aspect of understanding marine ecosystems. Although field observations provide
insight into dispersal processes, they do not provide a comprehensive assessment.
Individual-based models (IBMs) that couple ocean circulation and particle-tracking
models provide a unique ability to examine larval dispersal patterns with high spatial
and temporal resolution. Obtaining accurate results with IBMs requires simulating
a sufficient number of particles, and the sequential Bayesian procedure presented in
chapter 2 identifies when the number of particles is adequate to address predefined
research objectives. In addition, this method optimizes the particle release locations
to minimize the requisite number of particles. Even after applying this method, the
computational expense of IBM studies is still large. The model in chapter 3 seeks
to increase the affordability of IBM studies by transferring some of the calculations
to graphics processing units. Chapter 4 describes three algorithms that assist in interpreting
IBM output by identifying coherent geographic clusters from population
connectivity data. The first two algorithms have existed for nearly a decade and recently
been applied separately to marine ecology, and we provide a direct comparison
of the results from each. Additionally, we develop and present a new algorithm that
simultaneously considers multiple species. Finally, in chapter 5, we apply these tools
and a trait-based modeling framework to assess which species traits are most likely
to impact dispersal success and patterns in the Gulf of Maine. We conclude that the
traits influencing spawning distributions and habitat requirements for settlement are
most likely to influence dispersal.
Description:
Financial support was provided by the Department of Defense (DoD) through the
National Defense Science & Engineering Graduate Fellowship (NDSEG) Program,
Woods Hole Oceanographic Institution (WHOI) via the Ocean Ventures Fund (OVF),
and the National Science Foundation through grant numbers OCE-1459133, 0928442,
and 1031256.
Keywords:
Marine ecology
;
Plankton
;
Larval phase
;
Dispersal
Repository Name:
Woods Hole Open Access Server
Type:
Thesis
Permalink