Blog posts

How do droughts and hosepipe bans work in the UK – by Dr Barnaby Dobson

Water companies are required to publish drought plans every 5 years. In this they set out what they will do under a set of different drought scenarios to ensure reliability of water supply. These drought plans are then iterated with the Environment Agency to ensure they are defensible, cost effective and sufficiently protective of the environment. This whole process is explained in quite a friendly way here: MaRIUS_Drought_Primer_2017.pdf (

In the drought plans, there are defined ‘trigger levels’ that say how severe the current water supply situation is. These are usually based on reservoir levels or groundwater levels. The level that the water supply situation is in determines what emergency actions the water company may take.

Here is a screenshot from the South East Water (SEW) drought plan. We can see that the different groundwater level triggers are changing cyclically. We can also show a black line with the historic groundwater level over time, starting off in the green area (increased communication around water use), at some point entering the orange area (hosepipe bans) and even at one point close to the red area (non essential use bans). From a water company perspective, this is what a drought looks like, and they will be monitoring these levels carefully to see what state their supplies are in.

Here is another example from the SEW that shows what actions they may take depending on the drought level. We see that when we are in level 1/2/3 the water company can do things like hosepipe bans (temporary use bans) and activate some of their water supplies that might have some environmental impacts (e.g., drawing down a river more than they usually would).

Exactly what are the actions associated with the different levels, and what are the triggers depend on the water company. But all water companies will associate the different levels with different expected return periods (i.e., over a long period of time, how frequently would we expect these things to happen). For example, Thames Water for example expect Level 1 with a return period of 5 years, Level 2 with 10 years, Level 3 with 20 years and Level 4 never.

There isn’t anything particularly controversial about the implementation of a hosepipe ban or droughts in general provided a few things are true:

  1. The water company has implemented the previous actions that they said they would implement in their drought plan before getting to the point of hosepipe ban (e.g., communication campaign)
  2. The water supply situation is at the level when a hosepipe ban is warranted (e.g., the groundwater level is within the Level 2 triggers) and that they reached this place through lack of rain and not through operational failures.
  3. The actions are being implemented as specified (e.g., if the plan says that an abstraction location can increase abstractions from 0.5m3/s->0.75m3/s then the water company should still not be abstracting at 1m3/s), and that these specifications are not being changed at the last minute (e.g., the water company hasn’t petitioned the EA to raise the abstraction level to 1m3/s).

(I should note that I have no data and have done no research on the current drought so cannot say anything about whether these are true or not. Reaching out to water companies is probably the best bet!)

An interesting point is that SEW says in their drought plan that the expected return period of a hosepipe ban is 10 years. This of course depends on our hugely variable rainfall patterns, however it is to interesting note that their last hosepipe ban was in fact 10 years ago.

Whether this drought planning is the most effective way to plan for droughts is a very complicated question, but I would say that in the middle of a drought is not the best time to decide! Whether climate change is impacting a specific drought is also very difficult to answer, although we can say with some confidence that if carbon emissions continue to follow their business as usual trend that they have been following since we started doing carbon emission projections (RCP8.5 tracks cumulative CO2 emissions | PNAS) then the probability of a hosepipe ban occurring is likely to increase. My research shows that, averaged across the UK, a business as usual carbon emissions scenario would likely double the probability of a hosepipe ban by 2050 (The Spatial Dynamics of Droughts and Water Scarcity in England and Wales – Dobson – 2020 – Water Resources Research – Wiley Online Library).

Systems Engineering for Prototyping Modular Construction Systems Checklist

This checklist is part of a more comprehensive Toolkit developed by the Centre for Systems Engineering and Innovation (CSEI) at Imperial College London as part of the Impact award associated with the Royal Academy of Engineering and Laing O’Rourke Chair in Systems Integration. The Toolkit is work in progress and aims to provide Systems Engineering principles, tools and pointers for developing configurable product platforms for Design for Manufacturing and Assembly (DfMA) in infrastructure.

To access the checklist, click on the link below

Checklist – SE for Prototyping Modular Construction Systems Checklist

Comments and feedback are welcome.

Follow us on Twitter @CSEI_Imperial

Centre for Systems Engineering and Innovation – Introduction to the Systems Engineering Toolkit

Audio introduction to the toolkit

The Systems Engineering (SE) toolkit is work in progress. The first version brings together all the SE models and methods for the delivery of infrastructure. These span from functional and interconnectivity analysis to modelling languages and mathematics. This version aims to provide:

  • an overview of the fundamentals of systems engineering for delivering infrastructure,
  • a set of pointers to the existing resources,
  • a starting point for developing new knowledge about SE and Infrastructure.

Our next steps are:

  • Transition to version 2.0 organised by project stage (e.g. planning etc.).
  • Keep (do not replace with version 2.0) version 1.0 as an introduction and orientation to SE in delivering infrastructure.
  • Toolkit assessment through workshops and interviews with experts from academia and industry.
  • Early results from interviews suggest that there is a need to translate the output of the SE tools into an action plan.

Your feedback is important and encouraged to ensure that we cover all the tools and aspect of both SE and the delivery of infrastructure.

Version 1 of the full Systems Engineering in Infrastructure Toolkit is now also available

Slideshare Link




Small Project Proposal Call – 2017



The Centre for Systems Engineering and Innovation supports the application of systems engineering in the construction sector. The aim is to bring world class research on systems engineering and innovation into the sector to transform production of the built environment. The Centre acts as a hub for related inter-disciplinary research across campus on both systems engineering and innovation as applied to buildings and infrastructure. We are looking to fund feasibility studies through this call, we expect most proposals to be in the range £5k to £10k, although we will also consider smaller proposals.

Proposals are encouraged on (but not limited to) the following themes and topics:

1. Production systems, modularity, design and optimization of delivery

2. Visualization of interdependencies within and across infrastructures

3. Transitions to deliver value across the life-cycle

Deadline for submission of proposals: 17 March 2017. Submission by email to Decisions will be made by 7 April 2017 and applicants notified of outcome by email shortly thereafter. This internal call is open to any academics or students in Imperial College London, and proposals that connect with the interests of the Centre’s industry partners and members: Laing O’Rourke, Mott Macdonald, Project Production Institute and Bentley Systems are particularly encouraged. The proposal should fit on ONE page of A4 (Calibri 11 font), with the following table and check-boxes:

Title of project:
Principal Investigator:
Research background (brief context):
Fit with the vision and aims of the Centre:
Potential industry impact:
Added value:
Overall costs:
Follow-on plans:

Ο – I confirm that the above costs are FEC (please tick to confirm)
Ο If funded, I will be available to present the project and any follow-up research to the Centre for System Engineering and Innovation and its industry partners and members at the Research Showcase event on 7 September 2017. (please tick to confirm)
Ο Theme (please circle): 1 /2 / 3

CSEI_2017 call for proposals_final_template

EPSRC Systems Engineering workshop

Professor Whyte is looking forward to attending the EPSRC workshop on ‘Systems Engineering’ later this week. Our infrastructure systems are under stress, and systems approaches are appropriate to understanding their behaviour as complex systems and their interdependencies, within and across systems. There are new opportunities for data driven approaches, with the rise of new forms of cyber-physical systems, and new kinds production systems. Different research communities are taking different approaches to understanding their resilience, safety, security and socio-technical nature.  It will be interesting to see which topics the interdisciplinary engineering research community prioritise, and what lessons can be learnt across disciplines.   


Centre for Systems Engineering and Innovation

The Centre for Systems Engineering and Innovation is a hub for research across campus with a vision of bringing world class research on systems engineering and innovation into the infrastructure sector to transform production of the built environment. We work collaboratively with leading engineers and managers to achieve our vision through research, executive education and inspiring the next generation of engineers.