Modelling coral survival in a warming world

Overview

The ecosystem services provided by coral reefs are worth over $100 billion annually and include coast line protection, tourism, food and medical derivatives. However, the health of the constituent corals can be significantly impacted by coral bleaching. Coral bleaching is the loss of symbiotic zooxanthellae (Symbiodiniaceae) from tropical corals and can be caused by stressors such as thermal perturbations, disease and freshwater runoff (Fig. 1.). Thermal perturbations are thought to be the most significant bleaching trigger, and have been well documented in conjunction with major global bleaching events in 1998, 2002 & 2016/2017. These mass bleaching events caused widespread coral death with catastrophic ecosystem and service provision impacts. However, sub-lethal bleaching can also occur where the coral bleaches but recovers, and this may act as a ‘safety valve’ allowing coral hosts to survive periods of thermal stress in warmer waters.

Despite the devastation caused by severe coral bleaching, it is still difficult to accurately assess if corals will survive in the warmer oceans projected for the end of the century. Specifically, we do not fully understand if, and how, corals will spatially disperse by the end of the century. This is important as there is evidence that certain areas may act as coral refugia, harbouring colonies resilient to climate change.

Aim: To better understand future coral dispersal and any roles of coral refugia this project will integrate biological timeseries with spatiotemporal modelling to determine future coral reef extent.

Click on an image to expand

Image Captions

Figure 1.jpg Before and after widespread bleaching of a Samoan coral reef (Photo: The Ocean Agency / XL Catlin Seaview Survey)

Figure 2.jpg The supervisory team surveying a coral reef

Methodology

This project will include spatiotemporal analysis of coral distribution and resilience patterns from long-term ecological datasets, ensuring applicability to real-world conservation and resource management problems. Statistical inferential frameworks will be used to project future coral distributions under various climate change scenarios. There will also be the opportunity to explore how changing land-use might interact with climate change drivers on coral distributions.

Modelling: Spatiotemporal modelling of environmental data is often conducted using Bayesian hierarchical models, with the hierarchies representing the variation in temporal and spatial scales of the data or environmental process (Ledo et al (2016); Brown et al (2017); Jones-Todd et al (2018)). It also provides a natural probabilistic framework for predicting future coral distributions and their associated uncertainties. Estimation of relevant quantities of interest in spatiotemporal models can be conducted using a variety of statistical approaches, including Markov chain Monte Carlo (MCMC) and INLA. The project will determine the relevant choice of methods in the context of coral distribution modelling.

Field work: If desired, the scholar will have the opportunity to conduct model validation in relevant regions, such as the Caribbean, Red Sea and Mediterranean (Fig. 2; diving is optional), adopting state-of-the-art high-resolution 3D survey techniques.

Project Timeline

Year 1

Data base compiling, literature review, Bayesian hierarchical modelling, field work / ground truthing

Year 2

Data base compiling, Bayesian hierarchical modelling, field work / ground truthing, dissemination

Year 3

Bayesian hierarchical modelling, field work / ground truthing, dissemination, conference, publications

Year 3.5

Thesis completion, publications

Training
& Skills

Project support: The facilities and instrumentation available within the supervisors’ institutions provide a combination of leading laboratory, field and analytical capability and technical support that will be ideal for this proposed research, maximising PhD training from recognised experts in the field.

Scholar support: The Schools of Geographical and Earth Sciences & Mathematics and Statistics at the University of Glasgow have a large research student cohort that will provide peer-support throughout the research program. The scholar will participate in annual post-graduate research conferences, providing an opportunity to present their research to postgraduates and staff within the Schools, and to also learn about the research conducted by their fellow postgraduate peers. All project supervisors are highly research-active; the scholar will interact with all members of their research groups, providing an opportunity to learn about other techniques and research areas which may be applicable to their research. Additionally, the supervisors are all based in research-active departments that span a broad range of ecological and environmental research, exposing the scholar to a range of other research areas. To facilitate this, the scholar will actively participate in the ‘Marine Global Change’ group in GES, the Environmental Statistics group in Mathematics and Statistics as well as the Coastal Biogeochemistry group at the Lyell Centre. These group meetings provide opportunities to discuss cutting-edge topics in the field, review recent papers and to present current research plans to academics with a common research interest in an informal and supportive atmosphere.

The scholar will be encouraged to attend specialist courses that will directly contribute to the proposed project:
• The project involves a large component of modelling experiments and associated research and the scholar will be encouraged to attend relevant course throughout the PhD.
• This project may involve some fieldwork thus the scholar may attend a field first aid course.
• Relevant modelling and fieldwork training will be provided by the supervisors and / or specialist technicians for each piece of instrumentation required for analyses.
• The project supervisors’ will also support and encourage the scholar’s attendance on transferable skills training such as data management, scientific writing and science communication. The College of Science and Engineering at the University of Glasgow provides, for free, a large number of such courses, which are available throughout the PhD program.

References & further reading

C.J. Brown, S.D. Jupiter, S. Albert, C.J. Klein, S. Mangubhai, J.M. Maina, P.Mumby, J. Olley, B. Stewart-Koster, V. Tulloch, A. Wenger (2017), Tracing the influence of land-use change on water quality and coral reefs using a Bayesian model. Sci. Rep., 7 (1) p. 4740

Jonesâ€Todd, C. M., Swallow, B. , Illian, J. B. and Toms, M. (2018), A spatiotemporal multispecies model of a semicontinuous response. J. R. Stat. Soc. C, 67: 705-722. doi:10.1111/rssc.12250

Ledo, A., Illian, J. B., Schnitzer, S. A., Wright, S. J., Dalling, J. W. and Burslem, D. F. (2016), Lianas and soil nutrients predict fineâ€scale distribution of aboveâ€ground biomass in a tropical moist forest. J Ecol, 104: 1819-1828. doi:10.1111/1365-2745.12635

Further Information

Please contact Ben (ben.swallow@glasgow.ac.uk) before applying

IAPETUS2 applications: to apply for this PhD please use the url: https://www.gla.ac.uk/study/applyonline/?CAREER=PGR&PLAN_CODES=CF18-7316

Apply Now