Water is essential for the production of energy. Energy facilities both consume water and have impacts on the aquatic ecosystems they interact with. These interactions are complex however and it is a mistake to over-simplify – one we must avoid if we are to meet our future energy needs in the most sustainable manner possible.

Let me start off by asking a question, which of the following electricity sources has, generally speaking, the “worst” water impacts: nuclear, coal, natural gas, Hydro power, wind, solar, biomass/biofuel?

If you answered nuclear you would be wrong. If you answered gas you would be wrong.  If you answered hydro you would be wrong and if you answered wind, well you are also wrong. In fact, if you answered at all I’m afraid you have just illustrated the point I’m about to make – namely that the question of energy-water impacts has, as with so much of the energy debate, been turned into a stick with which to beat up certain energy technologies. Generalisations are at best false, and at worst they pander to our innate biases and lead us towards some potentially very bad decisions as far as the environment is concerned.

The problem is that knowing only the generating technology is not enough to come to a meaningful judgment. Missing, first and foremost, is the location. Does a coal plant on the coast have a greater or lesser water impact than a solar thermal plant in the desert? (No, I’m afraid that one doesn’t have a ready-made answer either). To understand the water system impacts we need a detailed picture of local water availability and the effects the plant has on its local aquatic environment and other potential users.

This much should be obvious to water and environment experts, and yet it didn’t stop the UNFCCC retweeting this image (figure 1) produced by the International Renewable Energy Agency.


Figure 1: or do they?

It is honestly disappointing to see two prominent intergovernmental organisations publicly express such a tech tribal message in relation to the energy and water issue. I wish the problem were confined to these groups but evidence suggests it is not. Beyond being merely disappointing, the statement is overly simplistic, since the largest sources of renewable energy are currently biofuels and hydropower which have complex aquatic interactions that come with no assurance of low-water consumption. An additional concern, however, is that it seems to reduce the energy- water issue down to that solely of consumption. This is really not on!

An all-consuming matter

Let’s get one thing clear. There is no shortage of water on the planet. Over 70% of the Earth’s surface is covered with the stuff. Water is not destroyed upon consumption, rather it simply changes state – either transformed into steam or has materials added so that it is deemed no longer fit to be returned to the original source. Most water bodies constantly get replenished and the question of how much you should take depends fundamentally on how quickly this happens.

Concerns over water availability really relate only to those energy facilities that extract from fresh water systems. Such facilities compete with a range of other human water-intensive activities such as agriculture, mining, domestic and (non-energy) industrial activities. Agriculture alone counts for the vast majority of fresh water consumption. Energy accounts for most of the industrial consumption as shown to scale in that tiny pink bar to the right in Figure 2. It frankly makes no sense to prioritise concerns about today’s levels of energy-related water consumption over that of other human activities.


Figure 2 estimated global fresh water use by sector. Source: UNEP:  VITAL WATER GRAPHICS An Overview of the State of the World’s Fresh and Marine Waters – 2nd Edition – 2008

I can’t speak for fossil, but most nuclear plants in the world are not sited on rivers and none draw from aquifers. They are located on the coast, or on large inland lakes with no obvious water availability issues, as Table 1 indicates. Of the nuclear plants which are sited on rivers, it is important to realise that these locations were selected based on what at the time would have been the rather ample availability of water. Even now these plants do not contribute to water shortages most of the time. In periods of severe drought or high temperatures they may contribute to water stress, but the question has to be asked – would that water stress be there minus the energy activity?

Table 1: Nuclear power plant siting and cooling sources – IAEA Efficient Water Management in Water Cooled Reactors

Nuclear plants using once-through cooling Nuclear plants using closed cycle cooling
Sea Lake River Cooling Towers
45% 15% 14% 26%


Figure 2 highlights one of the more interesting facts of energy related water-use. Most of what gets extracted is in fact returned to where it came from. In the case of thermal plants (which make up the majority of the world’s electricity production) the primary use of water is for cooling. Steam which has passed through the turbines needs to be cooled back to water before being fed back to the boiler. There are two main types of thermal plant cooling systems in use as illustrated in Figures 3 and 4. Both require water, but use it in different ways. Once-through cooling requires a comparatively large volume of water but returns the vast majority of this to the source albeit a few degrees warmer. Closed cycle wet cooling uses the atmosphere as a heat sink and employs cooling towers as a heat exchanger, but this still requires some make-up water input. A little-known fact is that evaporative cooling towers actually consume more water than once-through cooling does. There is also dry cooling technology used at some thermal plants. However while this process requires no water it comes at an efficiency penalty, meaning more fuel is required to produce the same amount of energy.











Figure 3: Once through cooling in thermal power plants. Source: BP – Water in the Energy Industry. For a typical nuclear plant withdrawal for cooling may be as high as 200m3/MWh, but consumption due to downstream evaporation will be between 1 – 2m3/MWh. The actual consumption depends on a range of factors – primarily plant thermal efficiency.









Figure 4: Thermal plant with wet cooling tower (closed cycle). Source: BP – Water in the Energy Industry. For a typical nuclear plant make-up water withdrawal rates are only slightly higher than ‘consumption’ – which corresponds to evaporation loss of about 3 -4m3/MWh. The actual consumption depends on a range of factors – primarily plant thermal efficiency.

During periods of high temperature it is often not water availability that is strictly speaking the problem. In France, nuclear plants have occasionally had to curtail operations because of regulatory/technical limits set upon river temperatures. This bears reflecting upon, since climate change means the natural range of river temperatures may well be up for re-definition. Idling river-based nuclear plants for thermal impacts is rather counter-productive if your aim is to prevent higher water temperatures in the first place!

On the other hand thermal discharges (as the warmed water is called) can create a genuine environmental problem. Increasing water temperatures reduce oxygen levels which many organisms are sensitive to. Somewhat ironically, it can also contribute to rather spectacular blooms of jellyfish, blue-green algae and other species. Outlet water temperatures most certainly need to be effectively monitored and controlled. The process of extracting water for once-through cooling also causes impacts to aquatic eco-systems by impinging fish and other organisms on inlet screens and dragging them through the system. It is unfortunately the case that thermal power plants kill what appear to be large number of very small aquatic organisms this way – although on the local ecosystem level these numbers are not particularly significant.

As a nuclear energy supporter and someone who cares about the natural environment I tolerate these impacts, just as being a supporter of wind energy I tolerate a level of avian kills. I don’t like it in either case but accept, grimly, that all human activities have their environmental toll. The aim is to keep this low, reduce it if feasible, and certainly to avoid any biodiversity loss or large ecosystem changes.

Of course, there are a range of potential aquatic ecosystem impacts associated with non-thermal energy technologies too. The most obvious target here is large hydro, but also clearly in the firing line are wave and tidal, as well as offshore oil gas and wind facilities. Even the mining and manufacturing requirements for solar PV have on occasion led to fish kills. So why then is there not more international focus on these and how to minimise them? It is this question that underscores my sense of frustration at the current state of the energy and water debate.


Figure 5: Three Gorges Dam. Source: Wikimedia Commons. (Author: Allen Watkin)

Against the tide

I have become sensitive to the shoddy state of the global energy-water dialogue because of my participation in the Water For Energy Framework (W4EF) initiative. This project is working towards creating a common methodology that any energy operator can employ to better understand the water risks their facility faces. On one hand, the framework considers the risk of potential impact on ecosystems and other human needs related to water consumption and discharge quality factors. On the other hand, it considers the risks to the operator related to the energy site’s water supply. The output is a set of simple indicators which illustrates these risks and informs management whether action is needed as well as allowing for better communication between energy operators and stakeholders at local and corporate level.

The W4EF methodology puts site specific water risks into a local context.  It is designed to be applied to any energy technology and doesn’t get distracted with seeking to vilify any of them in particular. Instead, it focuses on whether problems exist and what, if anything, needs to be done about them. I eagerly look forward to the next phase of the work and hope that it receives strong support from energy operators – but also the support of international associations and climate bodies that claim to act in the best interests of the environment.

Case studies from a range of energy facilities have impressed on me the fact that energy facility water interactions are not at all straight-forward. Complicating matters straight off the bat is that responsible operators seek to manage and reduce their impacts, especially where they are based in water-stressed environments. Some have created their own canal systems. Others purchase municipal waste water.  Many have invested in screens (or more advanced screens) to reduce aquatic organism kills. There are options which can be introduced to mitigate impacts for most energy technologies, not to mention innovations which may be available in the future. For nuclear plants the upgrades I would most like to see are those which will lead to improvements in plant thermal efficiency. This would reduce the need for water withdrawals and associated thermal discharges, but should also improve plant economics and performance – a win-win. It would also be rather nice to see uses explored for waste heat.

It is important that whatever steps are insisted on, they help to solve real environmental problems and not theoretical ones. Even more important is that steps don’t lead to even bigger environmental problems in the long-run – for example climate change, worsened by insisting unnecessarily upon the closure or curtailment of low-carbon generation. We can’t afford to keep getting distracted by energy ideology. The world deserves a more mature response to the water and energy challenges of the 21st century, and requires a handy tool, such as W4EF, that is up to the job.

energy collective