University of Edinburgh Featured PhD Programmes
University of Sussex Featured PhD Programmes
University of Glasgow Featured PhD Programmes

Developing Risk-Based Approaches to Modelling of Microbial Contamination in Scottish Rivers


Postgraduate Training

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
Dr M Glendell , Dr L Avery , Dr D Oliver No more applications being accepted

About the Project

This is an exciting studentship opportunity providing a platform to build an interdisciplinary research career in applied microbiology and modelling in the context of diffuse water pollution with human health impact.

Addressing sources of microbial water pollution from faecal matter, including diffuse pollution from agriculture, private septic tanks and sewage treatment works, is imperative for the protection of water resources (private water supplies, abstraction for public supplies, recreational/bathing waters and shellfish waters). A number of modelling approaches have been developed to aid the understanding of the potential sources of microbial pollution and the effectiveness of land management mitigation measures to inform policy and land management practice at a range of scales. However, process-based models often require more data for calibration and validation than is generally available at both catchment- and national-scales, while the simpler export-coefficient based models only allow a limited understanding of the underlying processes governing microbial pollution, commonly measured using faecal indicator organisms (FIOs). Therefore, novel modelling approaches are required that will overcome these limitations and inform optimal land management mitigation strategies in rural catchments.

Bayesian Belief Networks (BBNs) are probability based graphical modelling tools that allow system-level thinking, revealing possible causal relationships between controlling factors that may not be apparent otherwise and in situations where controlled experiments are not possible, such as diverse river catchments; allowing the evaluation of probable resilience factors and land management strategies in cases where empirical data or understanding are sparse, whilst accounting for uncertainties associated with both the model and data. Therefore, this research will explore the application of BBNs at different scales (catchment and national) to address the following questions:

1. How can BBNs be applied to understand microbial water pollution risk at catchment and national (Scotland-wide) scales and inform effective land management interventions?
2. Which in-stream processes are most important for determining downstream FIO pollution risk as implied from the model sensitivity analysis?
3. How does the risk of FIO pollution relate to other indicators of water quality such as dissolved organic carbon, particulate organic carbon, suspended sediments and nutrients?
4. What are the advantages/disadvantages of risk-based approaches to modelling of FIOs at a range of scales as compared to other modelling strategies?

Existing high resolution data (both spatial and temporal at nested scales; including FIO, water quality, soil physical and chemical properties, soil drainage, topographic, climatic, hydrological, land use and land management) will be used to build initial conceptual framework on which to base BBN (directed acyclic graph - DAG) for two data rich catchments (Tarland, River Ayr) with diverse land use. The conditional dependencies and probabilities will be derived directly from available data, including other model outputs, as well as literature, expert opinion and stakeholder understanding. This will be followed by sensitivity and uncertainty analysis, to identify areas of high model uncertainty and inform further data collection. The network will be revised and tested on further catchments to represent different catchment typologies in Scotland (NE, SW, W) and then up-scaled to a national scale using national data sets to allow predictions and scenario modelling in data poor areas across different catchment typologies. Gaps in existing process understanding will be identified and will inform future empirical research. Both the inputs and outputs of BBNs will be presented as a clear visual spatial network, ideally suited for stakeholder involvement in both model development and application. Model structural uncertainty will be tested and comparison with other modelling approaches (e.g. SWAT and VIPER-SCIMAP) will be undertaken at catchment and national scales to examine the accuracy, reliability and utility of different model structures as well as different modelling approaches, to make predictions of FIO fate and transfer in poorly studied catchments.

The studentship will broaden the scope of the applicant’s skills base by providing specialist training in the safe handling of Hazard Group 2 microorganisms & microbiological methods, and by developing expertise in the use of a wide range of innovative modelling techniques.

Funding Notes

The studentship is funded under the James Hutton Institute/University Joint PhD programme, in this case with the University of Stirling. Applicants should have a first-class honours degree in a relevant subject or a 2.1 honours degree plus Masters (or equivalent).Shortlisted candidates will be interviewed in Jan/Feb 2018. A more detailed plan of the studentship is available to candidates upon application. Funding is available for European applications, but Worldwide applicants who possess suitable self-funding are also invited to apply.


FindAPhD. Copyright 2005-2021
All rights reserved.