Adaptive water resources planning to manage future risks to London's water supply

Water resource management in the 21st Century is challenged by climate change impacts, increases in consumer demand, and environmental degradation. In the UK, water planners must secure long-term public water supply whilst facing increased exposure to climate-related hazards from increased aridity and uncertainties in future precipitation. It is widely accepted that to overcome such risks, water agencies must manage water demand, reduce the amount of water that is wasted through leakage, modify operational and regulatory policies, and build new supply infrastructure to increase water system capacity. However, significant uncertainties complicate adaptation decisions. This thesis presents three novel risk-based frameworks that address some of the challenges surrounding adaptability and uncertainty in 21st Century water resource planning. The aims of the thesis are: (i) to develop an adaptive risk-based planning methodology which is designed to cope with uncertain climate and demand projections by identifying decision rules that inform adaptive water resources planning decisions; (ii) to investigate the vulnerability of water management planning options to heterogeneous patterns of water use and spatially connected extreme climate events with characteristics not represented in the historical record, and examine how management decisions can be informed by extensive stress testing; and, (iii) to better incorporate river ecosystem health in water resource planning decisions, so regulatory decisions do a better job at preserving the aquatic environment in changing climatic conditions. The research presented in this thesis uses state of the art hydro-climatic data and water demand projections alongside simulation and optimization-based methods to achieve the aims described above. The proposed planning frameworks target uncertainties inherent to the London water supply system in the Thames Basin, England, but have been designed so that they can also be applied to regions facing similar planning uncertainties. This thesis demonstrates the value of systematically planning for hydro-climatic, environmental, and socio-demographic uncertainties. The findings suggest that water resources planning under uncertainty can be enhanced through monitoring of observable features of the water system, and by learning from simulation-based vulnerability analyses of proposed water management options and regulatory policies. Results from the case study investigations indicate that significant climate- and demand-related risks threaten water security in London and the wider Thames region. Without adaptation, water restrictions to domestic water users are projected to increase in frequency and severity throughout the 21st Century. For example, by 2099 the probability of severe water restrictions occurring on any given day in the Thames Basin are calculated to increase by 266% compared to the baseline period (1975-2000). Meteorological and hydrological droughts are also predicted to become more spatially extensive, affecting multiple water supply systems in Southern England and threatening the yield of existing infrastructure. Lastly, aquatic health is shown to deteriorate as water supply pressures increase. Overall, the results demonstrate how future risks in the Thames Basin can be navigated by adopting the quantitative methods presented in this thesis. By following a rule-based planning approach, water managers are able to make better informed investment decisions and could reduce the costs of water use restrictions by up to 24% over the next 60 years. Implementation of a new water transfer from the Severn Basin to the Thames is likely to increase the resilience of Thames Water’s water supply system to severe drought events, but only if adequate water storage is made available in the receiving region. Finally, modification of existing water withdrawal policies is necessary to maintain historically observed macroinvertebrate health (LIFE Scores) into the future. This thesis shows that increases to minimum required flow constraints along the River Lee, a catchment in the Thames Basin, could result in 21.3% improvements in LIFE Scores by the end of the century compared to the current policy.