Water System Resilience (ARCC-Water)

Lead Research Organisation: H R Wallingford Ltd
Department Name: Water Management


Reliable water supply is fundamental to human health and wellbeing, and in the UK is underpinned by inter-linked infrastructure for abstraction, storage, treatment and conveyance of potable and wastewater. Climate change has the potential to affect the UK water system in a number of ways: through changes in the water available for abstraction and storage, especially through altered drought frequency and intensity, changes in demand and changing risk of infrastructure failure.This project aims to develop a set of analysis tools and data on climate change and future demand that will enable users to identify packages of options that results in heightened reslience of the UK water system to these uncertain future drivers. The multi-criteria approach to be developed will also allow alternative adaptation options to be assessed against other criteria, such as environmental sustatinability, energy costs and public acceptability.The focus of the study is South and East England, an area that is already experiencing water system stress, and likely to be subject to additional stresses in the future due to climate change and demographic changes.The methods and results of this research will enable the UK to better plan for adaptation of the water system to climate change, and will help identify the polciy and regulatory changes that would be needed for adaptation to take place.The project has been designed in collaboration with stakeholders from government (DEFRA, EA, OFWAT, GLA), the water industry (UKWIR, Water UK and a number of water companies) and NGOs with an interest in water. These same stakeholders will be involved throughout the project as project partners.
Description Regional water resources and infrastructure Database (WP1):

Through publically available reports and meetings with the different water companies and other stakeholders involved in the project, HR Wallingford has set-up a database of water resources information for the South East region. This database of information includes but is not limited to:

- Surface water and Ground water abstraction locations, licenses, and Deployable Outputs

- Transfer agreements information: locations, volumes and restrictions

- Reservoirs locations and capacities

- Water loses

- Water demands: sectoral demands and returned flows

- Supply demand Balances

- Population & Geological maps

Regional water system model (WP4):

A substantial part of HR Wallingford involvement has been to conceptualised and set up the Regional water system model (RWSM) using an free and open source rule-based water resource management simulation program called IRAS-2010 which has a particular focus on computational efficiency to support large ensemble modelling studies. The philosophy of approach of the model developed has been to:

- Build a model at the water resources zone level with all surface water sources modelled as independent nodes, groundwater sources typically represented as a node per licence and key infrastructure such as reservoirs and inter and intra-company transfers included;

- With a unique network skeleton replicable across the different water resource zones

- And flexible enough to enable future transfers and storage options to be "switched on" for options testing without having to rebuild a model

The model set up includes demands reductions, monthly demand profiles, abstractions and transfers limits, and rivers connectivity. It is made of over 1000 nodes and 1000 links with nearly 50 surface water abstraction points and 360 Ground water license abstractions.

It requires only few seconds to run several decades and outputs a lot of information such as:

- Water demand met of each water resource zone: it is the main indicator of a bad water management as it indicates any water demand not met (frequency and quantity).

- Reservoirs water levels at each time steps: it helps identifying any near shortage of water, or any under-used reservoir

- Transfers quantities and frequencies: We can assess for each transfer, how often it is used, and if there would be an interest to modify the quantity agreed

- 'Hands-off' flow and compensation releases: The frequencies and the durations of environmental flows and compensation released can be obtained

Projections of climate and hydrological changes (WP2):

A high proportion of water sources in the South East region are from groundwater, which is inherently difficult to simulate through physically based simulation models. Furthermore the computational expense of considering groundwater modelling was too high given the large number of sites and length of future time series being considered in the ARCC water project. To be able to assess spatially coherent current and future groundwater yields for input into the regional water resources model a practicable approach to groundwater analysis was needed.

This study adopted a multiple linear regression (MLR) based analysis to create statistical models between groundwater and precipitation between groundwater signature sites representative of different aquifers within the region. A total of 44 signature sites were used in the analysis. Recharge modelling was also undertaken for all sites to provide an alternative to the use of precipitation as recharge also accounts for variations in temperature. The MLR procedure was applied to the recharge data in the same way as precipitation to see whether an improved model could be realised. The statistical model was then used to simulate groundwater levels for an extended baseline time series (extended through resampling) and an extended future time series from UKCP09. The spatially coherent UKCP09 projections (SCPs), which are used to maintain consistency between sites, are available as 11 realisations for each of the low, medium and high emissions scenarios. The time horizon considered were the 2030s, 2050s and 2080s. The final goal of the groundwater modelling was to provide time series of deployable output (DO) for groundwater sources considered to be drought vulnerable by the water companies for input into the regional water resources model. The DO time series were created using a drought curve 'shifting' method with the scaling of groundwater levels at signature borehole sites to source DO based on information made available by the water companies.

An overview of the analysis for each groundwater site is provided in the following steps:

1. Quality check of observed groundwater data and removal of any years which had poor coverage of the groundwater minimum.

2. Derivation of MLR model between annual groundwater minimum and precipitation (and in some cases a lagged groundwater variable).

3. Application of MLR model with the resampled baseline climatology to produce an extended time series of groundwater minima.

4. Application of MLR model with all resampled future climatology (33 - 3 emissions x 11 SCPs) to produce extended future time series of groundwater minima.

5. Derivation of DO time series for each source through analysis of signature borehole with drought curve 'shifting'.
Exploitation Route There is now a regional scale model of the SE water system that could be exploited by the industry to test further water supply planning options.

ARCC-Water work has required close interactions with water companies and could lead to future investigations for them, such as building a more localised\detailed model of their area and challenge their water management strategy. This could be done outside or inside a full regional model.

ARCC-Water approach\findings could also be used at a broader scale (national ?) for an even more global water management strategy.
ARCC-Water is of first relevance to water companies and regulators as it demonstrates that a spatially coherent, simulation based model that supports scenario and ensemble modelling are possible at a regional model. It opens the door to all sort of decision making analysis studies such as RDM or real options analysis at that level.

It should be noted that the RWSM and the database contain confidential information. Agreement from water companies and Environment Agency would be required to make use of these outside the Arcc Water project.
Sectors Environment

Description Yes, The concepts developed in this project have been used in subsequent UK Water Industry research on Risk Based Decision Making and now become standard practice as part of the industry's resilience agenda.
First Year Of Impact 2015
Sector Environment
Impact Types Policy & public services

Title New approach to regional modelling 
Description this work has paved the way for other regional modelling projects such as the WREA project as the industry looks to promote increased sharing of water, regional scale solutions and risk-based modelling techniques. Project has tested the viability and provides a 'blueprint' for how such models should be developed 
Type Of Material Computer model/algorithm 
Provided To Others? No  
Description ADB climate change knowledge partnership 
Organisation Asian Development Bank
Department Environment, Climate Change and Disaster Risk Management
Country Philippines 
Sector Private 
PI Contribution Providing training in climate change and adaptation
Collaborator Contribution Sharing knowledge
Impact Training Research Dissemination
Start Year 2014
Description World Bank Disaster Risk Assessment 
Organisation World Bank Group
Department Global Facility for Disaster Reduction and Recovery
Country United States 
Sector Multiple 
PI Contribution Developing training in disaster risk assessment including drought risk assessment
Collaborator Contribution Sharing knowledge from around the world
Impact Training of around 70 economists in developed and developing countries
Start Year 2014
Title water resource management simulation program: IRAS-2010 
Description the project lead to significant development of the freely available and open source IRAS 2010 software. 
Type Of Technology Software