IAP-24-115

Assessing the Hydro ecological Impacts of Climate-Driven Extreme Events on UK Rivers

Predicting the impact of extreme events on river hydro ecology and water quality under uncertain climate conditions is a complex but critical area of research [1]. This approach necessitates the integration of extensive datasets (e.g. climate projections, river flow, ecological data), advanced models (e.g. hydrological, ecological, computational), and sophisticated data science techniques, such as probabilistic methods for uncertainty quantification and mathematical extreme value theory [2, 3]. By combining these elements, the project aims to assess and predict how extreme events—like floods, droughts, and temperature spikes—will affect river ecosystems [4].

The research methodology will involve the integration of multiple datasets and models using a hybrid modelling structure. Key steps of research methodology will involve the following:
Stage 1 – Data Acquisition, organisation and pre-processing – Conduct a comprehensive literature review to identify suitable case study rivers across the UK. Acquire historical and future climate, hydrological, and ecological data from open-source databases and end-user communities. Thoroughly check data for quality and consistency. Organise and format data for subsequent modelling stages.
Stage 2 – River flow simulation: Calibrate and validate a suitable hydrological/statistical model (e.g., STL_HHM_GP [5, 6]) using historical data and simulate river flow under different climate scenarios (historic and future climate projections). The STL_HMM_GP is an ensemble model and could generate multiple realistically possible synthetics streamflow series. The project will explore opportunities of using ensemble models for simulating multiple realisations of river flows to allow a systematics uncertainty quantification in model-based uncertainty.
Stage 3 – Extreme Value Analysis (EVA): Analyse historical flow data and simulated flow sequences to identify extreme flow events and estimate their return periods. Create a sample of hydrographs with different return period events [e.g. 2, 5, 10, 50, 100, 200 years]. Explore the potential of mathematical Extreme Value theory to fit suitable extreme value distribution to climatic and flow series [7].
Stage 4 – Hydro-ecological models – Simulate hydro-ecological model such as those developed by UKCEH/EA, now advocated by the EA (as the HET package) to simulate the response of aquatic organism (e.g. invertebrates) to a range of flow hydrographs with different return periods and extreme climatic condition [8, 9].
Stage 5 – Training Statistical/ML approaches – Apply suitable Statistical/ML models for associating outputs of hydro-ecological modelling with simulated streamflow sequence and portfolio of climatic extreme events. The surrogate models will be thoroughly validated to assess their suitability for projecting ecological responses of rivers for multiple realisation of extreme flow and climatic conditions.
Stage 6-Uncertainty quantification- Apply suitable statistical methods (e.g. Bayesian) for incorporate prior knowledge and uncertainty into model predictions [10]. Further, apply QE-ANOVA/Monte-Carlo simulation to assess uncertainty quantification and propagation in predictions and entire modelling chain [11]. Apply sensitivity analysis to create a hierarchy of most influencing model parameters [12].
Stage 7 – Model application – Apply validated models to a range of climate projections (e.g. CMIP6, UKCP18) to conduct a extreme events impact analysis on ecological health of rivers.

Year 1

Year 1: Data Acquisition, Preparation, and Initial Modelling
• Literature Review (0 – 6 months)
• Data Acquisition, Quality Assurance and pre-processing (7-9 month)
• Training of Statistical, Hydrological, Hydro-ecological modelling (10 – 12 months).

Year 2

Year 2: Hydrological Modelling and Extreme Value Analysis
• Hydrological Model Calibration, Validation and flow simulation (13 – 15 months)
• Extreme Event analysis, return period and hydrographs (16 -18 months)
• Apply mathematical extreme value theory (19-21 month)
• One Conference and/or one journal Paper (22 – 24 months)

Year 3

Year 3: Hydro-Ecological Model simulation, Statistical Analysis, Uncertainty Quantification and Model Application
• Hydro-ecological model simulation (25 – 27 months)
• Statistical/Machine Learning ensemble model (28 – 30 months)
• Uncertainty Quantification and sensitivity analysis (31 – 33 months)
• Model application and papers writing (One conference and two journal papers (34 – 36 months).

Year 3.5

Year 4: Writing up and viva
Writing up thesis and viva examination (37 – 42 months).

By undertaking this research project, the student will develop interdisciplinary research skills (Data acquisition, analyses and modelling) in hydrology, ecology, computational approaches (statistical and machine learning), extreme value, uncertainty and scenario analysis. They will champion data interpretation, model development and calibration.

In addition to technical skills, the student will develop crucial soft skills:
• Problem-solving: Identifying and addressing research challenges.
• Critical Thinking: Evaluating information and drawing conclusions.
• Project Management: Planning, organising, and executing research projects.
• Communication: Effective communication of complex technical information to project partners and stakeholders.
• Collaboration: Working effectively with diverse teams.
• Time Management: Efficiently managing time and resources.
• Scientific Writing: Writing clear and concise research papers and reports.

The project will equip the student with the necessary skills and knowledge to pursue a successful career in academia, industry, or government. They will be well-prepared to contribute to addressing pressing environmental challenges, such as climate change adaptation, resilience and hydro-ecological health management.