Use this URL to cite or link to this record in EThOS:
Title: Predicting the effects of climate change and disturbance on marine sessile communities
Author: Knowland, Cheryl Anne
ISNI:       0000 0004 2736 9625
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
This study aims to evaluate and advance our ability to predict the effects of disturbance on communities using statistical models. This is important because climate change is likely to alter the patterns of disturbance experienced by ecosystems. In the first chapter, I review the evidence for effects of climate change on the disturbance regimes experienced by marine ecosystems, particularly coral reefs. I also outline theories about how disturbance affects communities, and give background information on two field sites: Heron Island in the Great Barrier Reef, and Salthouse Dock in Liverpool. Chapter 2 is a comparison of four models for coral community dynamics at Heron Island; a discrete-time Markov model and a nonlinear continuous-time model, each with and without the effects of cyclones. The models are fitted using Bayesian methods. The addition of cyclones improved the fit of the model, with cyclones causing some coral mortality, although recovery was rapid. Recruitment appeared less important than the growth and mortality of existing corals. In Chapter 3, I use the best model from Chapter 2 to examine how different the dynamics at Heron Island for the years 1962-1992 might have been under a range of cyclone frequency scenarios. Under current or slightly increased cyclone frequency, coral cover could be either high or low, depending on the timing of cyclones. With increasing cyclone frequency, there was a transition from the outcome being contingent on chance events to being almost certain about the outcome. In Chapter 4, we take the two models with cyclone effects from Chapter 2, and run them forwards in time until the year 2008. We then compare the predictions with newly collected field data using Bayesian methods. The continuous-time model predicted significantly less coral than was observed in 2008. Conversely, the discrete-time model predicted significantly more coral than was observed. Overall, the predictions made by the continuous-time model are closer to the observations. Chapter 5 is a two-year experimental study on the effects of disturbance on marine sessile communities. A two-year study was carried out at the Salthouse Dock site to determine whether a disturbance regime would cause persistent changes in community composition. Recovery was rapid. A simple mathematical model suggests that recovery is almost inevitable in systems of this kind, especially for species with planktonic larval stages. Comparing the Heron Island and Salthouse Dock results highlights the importance of external recruitment in recovery from disturbance. I discuss the advantages and disadvantages of alternative modelling strategies, and argue that simple models based on extensive data may be useful.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available