This is an abridged and adapted version of an article written for Green Building Magazine. The full article is available here

As defined by the Water Footprint Network (WFN), the water footprint is the volume of water consumed over the lifecycle of a product and is made up of green, blue and grey components. Green water is the amount of precipitation that does not become ground or surface water. Blue water is the volume of surface and ground water sources consumed. Grey water is the volume of freshwater that is required to assimilate a pollutant load, and we can add these volumes together to calculate a water footprint1. Water footprints can also be calculated for regions, or individuals. For example, the average UK consumer has a water footprint of around 4600 litres per day2. This is a big number, but is it meaningful? Looking at the calculations and assumptions behind water footprints reveals several problems.

Size isn’t everything

A litre of water differs enormously in environmental impact depending on where and when it is used; using 1000 litres of water in the UK in a wet winter is going to have a completely different impact compared to 1000 litres of water in a Californian summer. An entire sub-discipline of water footprinting is therefore devoted to calculating weighting factors3 and we cannot simply assume that small water footprints are a good thing.

Renewable resources, stocks and flows

The second interpretation problem arises from the fact that that water is renewable over relevant timescales; it moves through the hydrological cycle over the course of days/weeks/years, so when we say that water has been ‘consumed’, all we actually mean is that at that moment in time it is not liquid freshwater. It may have evaporated (e.g. in a cooling tower when producing electricity), or transpired through the leaves of vegetation. Some may have been incorporated into the product. Water that we consume in the present be used again in the near future; it isn’t going to run out in any permanent sense, we are simply moving it about. This means that a high water footprint for one activity does not always lead to a lack of water for others, except at a very local and immediate scale. The hydrologist’s approach of calculating a water balance based on stock and flow methodology therefore allows a more complete analysis than a water footprint study.

The volume of interest – withdrawal or consumption?

When we say we have used 100 litres of water, are we referring to the volume of water withdrawn or the volume of water that has been consumed (the volume withdrawn minus the amount that is discharged and is therefore available again for other users)? Which number is the most useful depends on the application. For a hydrologist considering how large a water withdrawal might be allowed for irrigated agriculture, the consumed volume is vitally important; this is water intrinsically linked to crop growth via transpiration, plus that evaporated from bare ground, but this is water that will not immediately recharge groundwater or be available as surface water for municipal or industry use. In contrast, if an identical volume of water is withdrawn for municipal water use, virtually the entire volume might be discharged again into the same watershed following sewage treatment; so our domestic water footprint might be practically zero, and reflects neither the importance of the water to the individual, or the impact of our temporary removal of it from the environment. A consumptive water footprint is not always a useful number.

Water pollution and the grey water footprint

The other particularly problematic aspect of the water footprint is the idea of the ‘grey’ water footprint4. Whilst we can calculate a volume of environmental water that would be required to assimilate a pollutant load, this is a theoretical volume (it may exceed the total water available in a river basin). Expressing a quality parameter as a volume is also questionable; multiple pollutants may interact with each other or have unknown environmental effects, and we have to know what the safe environmental concentration of a pollutant is in the first place; new understanding of environmental impacts of a specific pollutant could easily change the grey water footprint of a product by a factor of 100.

Flawed calculations, useful lessons?

Despite the problems with the calculations, the discipline has made some useful contributions. Until the phrase ‘green water’ was coined by the work of Falkenmark & Rockstrom5, the immediate fate of rainfall and its importance in global food production was rarely discussed by policy makers, agronomists and farmers, despite the fact that 75% of the total water consumed by agriculture is rainfall as opposed to irrigation water. Understanding how to optimise green water productivity is vital given the context of rising global populations and climate change. However it does not necessarily follow that green water footprint calculations aid this understanding.

The fact that the water footprint emphasises that withdrawal and consumption are not the same thing is useful, particularly at the scale of a watershed. Multiple stakeholders (agriculture, municipal, industrial) will all have water needs, and an environmental regulator needs a means of allocating water to the various users, which should be underpinned by an understanding of how much water withdrawn from the environment is consumed compared to being re-available for downstream users. It might be that agricultural users should be incentivised to make better use of rainfall, and not be allowed ground or surface water withdrawals, but the knock on effects of agriculture’s ‘efficient’ use of rainfall in reducing surface water runoff generation and groundwater recharge must also be considered6. Again, water balance calculations offer a more rigorous accounting method than water footprints.

Some would argue that the idea of a water footprint has been useful as a tool for public engagement. However, the uncertainty of the underlying calculations and the difficulty in coming up with a sensible conclusion or message to follow the water footprint calculation carries with it a risk of subsequent confusion, misplaced objectives, disengagement and cynicism. In short, the water footprint illuminates some important concepts in hydrology, but the calculations themselves do not necessarily have much application.

Where next?

There are a plethora of tools and reporting initiatives relating to water, and understanding which is most appropriate in any given circumstance is not easy7. The excellent water risk filter by WWF8 allows a business to do a very quick analysis on how their activities affect the water environment and how they may be impacted by water availability. Water Stewardship initiatives (e.g. Alliance for Water Stewardship9) are also useful, and can develop alongside initiatives in other environmental realms (e.g. marine and forest stewardship standards). Reporting standards (e.g. the Carbon Disclosure Project has a water standard)10also exist, but it remains to be seen whether these drive environmental improvements. The ‘gold standard’ for detailed environmental accounting of products undoubtedly remains Life Cycle Assessment (LCA), and the forthcoming inclusion of a specific water footprint ISO standard within the LCA framework will hopefully ensure that water footprint studies in the future are more transparent and useful than many of those undertaken to date.


Judith spent 18 months pondering water footprints at the University of Leeds, but recently escaped back to Wales and is now based at the Institute of Biological, Environmental and Rural Sciences (IBERS) at the University of Aberystwyth. Prior to that she spent time at the Welsh School of Architecture in Cardiff, and the Centre for Alternative Technology, Machynlleth.



  1. The approach to calculating water footprints according to the WFN method is detailed in the (freely downloadable) Water Footprint Assessment Manual. Hoekstra (2011).
  2. Orr & Chapagain (2008). UK Water Footprint: the impact of the UK’s food and fibre consumption on global water resources. WWF.
  3. Reviewed by Kounina et al (2013). Review of methods addressing freshwater use in life cycle inventory and impact assessment. Int J LCA 18(3). 707-721.
  4. This bears no relationship at all to the concept of household ‘grey’ water (generally defined as the wastewater derived from a household, with the exception of WC flush water).
  5. Discussed in their excellent book – Balancing water for humans and nature: the new approach in ecohydrology. 2004: Earthscan.
  6. The separation of water into green and blue has led to some policy makers believing that green water is effectively free – the result is an increasing number of closed river basins, caused by increasing efficiency of small scale capture of green water and so a lack of groundwater recharge and hence streamflow downstream. An excellent example is Ramgarh dam in Jaipur – built to serve the needs of the city and with a capacity of 75 million m3 it has been completely dry owing in large part to initiatives to increase agricultural water efficiency upstream.
  7. WBCSD, Water for Business. Initiatives guiding sustainable water management in the private sector, 2012, WBCSD, SustainAbility, IUCN.
  8. WWF and DEG. The Water Risk Filter. Available from:
  10. Collective responses to rising water challenges, 2012.

Posted July 2014

Water footprints – what do they really tell us? – Judith Thornton, UK