The Scottish Environmental Protection Agency (SEPA) liaises with operators and owners of legacy assets to improve the ecological continuity of the river network in Scotland. In particular, weirs act as barriers to natural fish migration in the river. Current guidelines define a weir as being passable if, roughly speaking, a certain proportion, such as 20%, of the total fish population can traverse the barrier. On longer rivers, such as the Scottish Tyne, the cumulative effect of the barriers is such that only a small proportion of fish are able to successfully migrate upstream to breed. It is therefore desirable to mitigate the impact of these barriers – from modification to full removal.
The financial cost associated with weir modification and/or removal is substantial and it is necessary to target resources to the barriers where intervention is expected to deliver the most gains in fish population. The aim of this project is to develop the statistical tools required for a data-centric management strategy for the river network. If successful, these tools could be used as a decision support tool by river management stakeholders, such as SEPA, in order to better target resources to the benefit of fish ecology in Scotland. To achieve this, a statistical model (a continuous time, discrete state Markov chain) will be developed to understand the movement of fish between different sections of the Scottish Tyne. Data from fish catches, geomorphology and engineering assessment of the barriers on the Tyne will be used to fit the parameters of the statistical model, reflecting the specific difficulty of each barrier at a species-specific level. The passability of the barrier combines with the flow conditions in the river to determine the proportion of fish that are able to successfully pass it. Once fitted, the model will be used in a predictive capacity to determine which of a selection of hypothetical interventions would produce the most benefit, as measured by fish migration, per unit cost for the engineering work undertaken. Predictions would depend on future climate projections, allowing for both larger floods and prolonged droughts, and appropriately taking their associated uncertainties into account.
The PhD candidate can expect to receive a comprehensive statistical training, including participation in the Academy for PhD Training in Statistics (apts.ac.uk). Specific skills that will be developed include statistical modelling, scientific computing, data analysis, project management and stakeholder mapping and engagement.
This project is part of the ONE Planet DTP. Find out more here: View Website