EROWI - energy return of water invested

Energy Return of Water Invested (EROWI). From an article by Robert Service in Science Magazine. The data in the table originate from "Energy demands on water resources",report to the congress, 2006 link.

The readers of "The Oil Drum" are familiar with the concept of "Energy Return of Energy Invested" (EROI or EROEI). It is the ratio of the energy produced by an energy plant during its life cycle to the amount of energy needed to build, operate and dismantle the plant.

EROEI remains one of the most useful parameters that can be used for evaluating an energy technology, but it is not the only one. Another element is the need of water. Water is needed for irrigation of plants to be used as fuel and all large plants using thermal engines need water cooling. We can speak, then, of Energy return of Water Invested (EROWI). It is a concept much more recent than that of EROEI, but which is rapidly gaining attention and may be not less important.

Recently, Robert F. Service reported the comparative table that you can see reproduced at the beginning of this post. The data are taken from an article by Dominguez-faus et al. published in "Environmental Science and Technology" in 2009. Service's paper, as most of the studies published so far in this field, is dedicated to showing how water thirsty biofuels are. It is another drawback for a technology which has also a low EROEI, needs large areas, and competes for land with food production.

But the problem is more general and doesn't just involve biofuels. Nuclear plants, for instance, seem to be especially vulnerable to water scarcity. During the past few years, several plants had to be shut or slowed down, or allowed to drain water into rivers at higher temperatures than considered safe. A set of references on the troubles of nuclear plants during heat waves can be found here .

The problem may affect all thermal plants which are large and inefficient enough; coal plants for instance. According to Service's data, the problem can be eased moving from "once through" to "closed loop" cooling. But, if it were easy, there would be no "once through" plants. Evidently, closed loop cooling is more expensive and, in practice, the result of increasing EROWI may be to reduce EROEI.

Water is, of course, a renewable resource but a lot of the water used today is "fossil" water. It comes from deep aquifers which can be drained empty as it has happened, for instance in Saudi Arabia . In addition, climate change may further reduce the water supply in many areas of the world. How much these factors will affect energy generation worldwide in the near future is difficult to say at present, but surely the problem shouldn't be underestimated. The EROWI problem, in the end, is just an indication that we are hitting yet another limit of our finite environment.

The EROWI concept is examined in depth, especially for biofuels, in an article titled "Burning Water: A Comparative Analysis of the Energy Return on Water Invested" by Kenneth Mulder, Nathan Hagens and Brendan Fisher, in press on AMBIO (The Journal of Human Environment) .

Thnx for information!

This article is a little misleading. There's distinction between salt water and fresh water usage. Nuclear power plants built next to the ocean uses salt water for cooler. The resulting steam is cooled down and turned into fresh water. That's why it's called nuclear desalination.

Russia has a prototype nuclear reactor that's built into a ship. It also uses the seawater around it for cooling.

Biofuel is FRESHWATER intensive. No doubt about that. But that chart is deliberately trying to make nuclear power look bad.

George Orwell

Irrespective of any other arguments regarding nuclear, seawater intakes result in killing 100% of all the biomass they suck in (larvae, plankton, phytoplankton, etc). In the case of San Onofre in California, that's 1.2 billion gallons per day. If you are one of the two billion people who depend upon fish for the protein in their diet, then that is indeed a bad thing. There are huge dead zones around nuclear power plants.

Thanks for pointing that out, George. The disctinction probably won't matter to the reflexive nuke-haters though.

And how about molten salt reactors? I bet they don't use nearly as much H20.

What has the reactor type to do with it? This is waste heat remaining, from steam after turbines.

The problem with nuclear is the waste. Solve that one first, then get back to us.

Your link to that Science Magazine piece leads nowhere. Here are Search Results for Service's articles, none of which are available free.

According to Service's data, the problem can be eased moving from "once through" to "closed loop" cooling. But, if it were easy, there would be no "once through" plants.

This is exceedingly tendentious logic. Many power plants aren't fitted with CHG, either, from this do we conclude that CHG is prohibitively expensive? I see that the Calvert Cliffs Nuclear plant, in operation for 34 years now, uses closed loop cooling, and also desalinates via r/o. Perhaps others have more data about the cost of implementing closed loop cooling.

The numbers for water use have a lot of problems in terms of comparing apples to oranges. For example, in "once through", the water is NOT consumed, but ends up back in the river, ocean, etc. Thermal power plants need to cool and condense the low pressure steam after the turbine. Efficiency is actually a function of the difference between the high and low temperatures of the cycle. Thermal power plants actually have even a "dry cooling" option where no water is needed, but it is more costly and less efficient.

The most important conclusion from the numbers is that using irrigated crops for biofuels is extremely water inefficient.

I'm a member of AAAS and have access to what is behind the sciencemag privacy wall ... nothing worthwhile on this topic, I assure you. The quotation given by Ugo is a piece of blatant policy wonkery that is utterly without scientific merit. Such writing has become much to common in sciencemag in recent years. But ... it was published in a 'reputable' journal, and is, therefore, remarkable as an example of 'main-stream' thinking. It is an interesting basis for heated discussion here. It kind of sets a very low bar for what is an acceptable contribution in this forum.

Thanks for the comment. I will be watching my BAS for a retort...

Isn't cogeneration now a part of the nuke equation? A lot of that waste heat doesn't have to be wasted.

Perhaps the problem is the reliance on steam turbines at power plants. Open loop Brayton cycle turbines get around the water issue since the turbine exhaust is directly vented to the air. Thermal regenerators would improve fuel efficiency if that is an issue. The high temperature of a Molten Salt thorium floride reactor would pair up nicely with a Brayton cycle turbine.

You have no idea how happy I am to have this sensible idea start to get traction.

Come now pondlife, whats 500,000 years between species?

Not all nuclear power plants use salt water. In 2003 and I think at least once since that nuclear power plants in France shut down because it was so hot that the hot water they were releasing in RIVERS would make the already warm water too hot. Rivers are generally not salt water----

From this map it looks like the bulk of US nuclear power plants are inland. I doubt they pipe salt water to them to use for cooling.

Because of fresh water constraints, future nuclear plants are more likely to be utilizing saltwater.

So there will be no more building of nuclear plants in the vast interior of the US. Kind of hard to see how this fits with some people's grand plans for nuclear to solve the bulk of our future energy needs.

And are we supposed to feel safe about nuclear plants being built on coasts that will be disappearing under multiple meters of water as the oceans rise? And the idea of building them along the west coast which is famously overdue for a major earthquake is supposed to make us feel happy and secure?

There are some proposals to build new reactors on sites already approved for nuclear power generation. The huge cooling towers recycle water used in cooling.

Even coal generated power plants next to rivers discharged heat into the water. One near me is a favorite spot of people fishing from the river bank. They bottom fished near the discharge pipe where steam rose in the winter and fish sought the warm water and increased plankton living in that zone.

So you do accept that this practice does affect the water and alter the eco-system. You choose to focus on what you see as a benign affect, but what are the effects of increased temps there in the heat of the summer?

At Norh Anna in Virginia there is a large impiundment known as the cooling lagoon which is coonsistently warmer than the main lake-it has an ecology similar to any similar impoundment not used for cooling located a few hundred miles father south-locals enjoy Ga style fishing.

This is not necessarily a bad tradeoff, given the fact that the nukes produce almost no other pollution, except for spent fuel.

And the spent fuel can be either recycled or stored safely once we get our heads out of our butts politically.

How safely?

I don't know -and niether does anyone else -precisely.But safer than burning coal, certtainly.

Safer than civil war too.

Are you offering your property Old Farmer for storing the spent nuclear fuel???

I typed up a long answer and it wafted off into hyperspace somewhere-but in short -if I had a suitable property and got a fat price I would, and I would not mind living next door either.

Otoh I also live recklessly-Iused to work in that plant , sometimes inside the containment, during outages.I drive a car, I eat meat, I eat food contaminated with nasty chemicals,and have both air conditioning and a refrigerator.

Safety is a relative affair-we are storing other things-withouit ANY safegaurds-that are demonstrably far more dangerous-such as zillions of tons of co2 in the atmosphere.

We are engaged in a "war" over oil that might just escalate into the real thing-how safe is that?

I would like to see nuclear power phased out someday-but as a practical matter I believe it is a superb stop gap that just might help us get thru the crunch that is coming without a long extended visit by the Four Horsemen.

Sometimes solutions that are very reasonable on one set of grounds-such as abolishing nukes on environmental grounds-are totally out of the question on other grounds.

If the lights go off, law and order and any semblence of personal security will cease to exist within a week.

Which is why I'm also such a hidebound recalcitrant ignorant conservative in respect to my guns-I am truly aware of the possibility that the lights might just go out, and that there is only one cop for every thousand people around here, and only one for about every two hundred or so even in well patrolled cities.

I expect if the lights do go out that at least half of them will be busy with thier own families, and that a tenth of the other half will be dead within two weeks.Another tenth may turn into Mexican style cops-the kind that collect a bribe simply because you cross thier path.

oldfarmermac -

A lot of people don't understand that the concept of 'safe' cannot be viewed in absolute terms. Rather, it always has to be considered in terms of 'safe as compared to what?'

While it is always possible to screw up anything made by man regardless of how many safety features have been designed in, I think the important question is: what is the least hazardous pathway to take. And in that regard, nuclear power has over the last five decades proved to be quite safe. During that time period there was but one really catastrophic nuclear power event, Chernobyl.

But the Chernobyl plant was poorly designed, poorly constructed, poorly maintained, and poorly operated ..... a perfect example of Soviet-era indifference and incompetence on the eve of that regime's demise. If Chernobyl were a car, it would probably be something like a badly abused 1986 Yugo used as a taxi in Honduras and with over 150,000 miles on it. Let the anti-nuke people answer this question; how many serious nuclear accidents have there been in either Japan or France, both of which have been very heavily invested in nuclear power.

Given the current increasingly unstable geopolitical situation, if one insists upon worrying about dying from radiation poisoning, then I say use your worry 'capital' wisely and start worrying about a deliberate or accidently nuclear exchange resulting from the increasingly nasty competition over fossil fuel resources. Then throw in the situation with US/Israeli versus Iran and/or versus Pakistan, and you might have something substantial to worry about. As we all know, things can turn to shite real fast.

As for me, if I HAD to live right next door to a power plant, I would much rather live next to a nuke than a coal-fired one. (A highly remote chance of being fried by a massive radiation release versus being slowly impaired by all sorts of continuous harmful air emissions.)

Hey, nobody gets out of this joint alive anyway.

"Given the current increasingly unstable geopolitical situation"

Indeed, and into that increasingly unstable situation you think it is safe to build hundreds of new nuclear plants? ? ?

To the extent that nukes are safe, they are so because they are run by competent people and institutions in stable conditions. There have been no world wars since nuclear power plants started being built. Can we guarantee that no world war (or other major military disruption in countries with many nukes) will ever happen again? When it does, what will happen to the safety of these nuclear plants?

I always find it interesting that people who otherwise understand that the world is a very uncertain place suddenly assume perfect geopolitical stability forever when they are advocating for more nukes.

Personally I don't assume stability forever, or even from year to year.

I do estimate that the risk of war and mayhem is reduced by a factor sufficient to more or less cancel out the increased risk of the plants existence in a time of war.

I readily admit that I could be wrong in this respect-plants would be attacked as strategic targets of course but a conquering army would have a very strong incentive to preserve the plants if possible.

The primary thing is to prevent the next world war, which will almost certainly be a resource war in the last analysis.

"The primary thing is to prevent the next world war, which will almost certainly be a resource war in the last analysis."

Good point.

One thing that has struck me recently is that one of the most important (to our economy) things we do with energy is to extract more of other non-renewable resources. On the one hand, this means that more energy means more resources extracted and less reason for war over those resources. On the other hand, more energy means faster reductions in those resources. This is a variation on the slow crash vs fast crash issue.

The main point to me is that, more than finding new sources of energy or other resources, we have to focus on ways of avoiding catastrophic conflict in the limited world in which we live, over the short and long terms.

This has technical aspects, but it is not primarily a technical problem. As technologically sophisticated people, we tend to see every problem as a nail our technological hammer can potentially solve. This may blind us to the essential natures of the major challenges we face.

The Chernobyl reactor had no containment building and ejected a third of its core. There were less than 40 prompt fatalities and there may be a few thousand lives shortened over 40 years in the unlikely event that the linear no threshold theory of radiation effects is valid.

In war or peace, if nuclear plants can lure our enemies away from more vulnerable targets they will save thousands, perhaps millions of lives.

Ah yes, we trusted nothing the Commies ever said except their numbers on deaths at Chernobyl???? Amazing how trusting we can be of our enemies when it is to our advantage.

"... nuclear power has over the last five decades proved to be quite safe. During that time period there was but one really catastrophic nuclear power event, Chernobyl."

Woa, you're going to have to put a lot of qualifiers after that statement! Such as "only one really catastrophic nuclear power event, in the Soviet Union, that killed people, who were civilians, that we know of."

Do a little research with an open mind. Google for Daniel Ford, Harvey Wasserman, John Gofman, Ernest Sternglass... there's lots out there that you don't find in the New York Times or even Scientific American. I seem to recall that at least a hundred Americans have been killed in nuclear accidents -- and that's not counting the statistical evidence, such as Sternglass's astounding, well-researched study that showed that there were at least 400 "excess" infant deaths after Three Mile Island than there would have been normally. See also the sorry tale of "downwinders" living (and dying mysterious deaths) around Saint George, Utah.

In 1966, I, and everyone I had ever known in my eleven years of life, almost died. The liquid metal fast breeder reactor (LMFBR) that was under five miles from my house suffered a catastrophic meltdown -- one could call that an "event," no? I was in a little band, and the fiddle player's father was an engineer at the plant. During the emergency, they were tired, haggard, working around the clock, forgetting to wear dosimeters, cutting holes in pipes and putting mirrors on sticks into them, trying to keep a puddle of liquid plutonium from forming a critical mass that would "fizzle yield" into a multi-kiloton spewing of plutonium oxide into the atmosphere -- enough to kill every person on earth, if evenly dispersed.

Within a decade of this "event," Phil Harrigan was dead in his early 50's of leukaemia, leaving a grieving family and an un-provable trail between nuclear accident and death. How many more might there be?

I am not making this up! But the spin on this particular "event" is incredible. No one mentions Phil's untimely death. No one seems to know if other plant scientists and engineers have succumbed.

The problem is that cause and effect is not easily proven with this particular hazard.

Infant mortality was steadily declining in the Amish country around Lancaster Pennsylvania in the late '70's, then suddenly stopped declining when Three Mile Island melted, until the radioactive isotopes could be washed out of the air and water and soil -- if you subtract out the expected infant mortality from the actual, the area under the curve is some 400 babies' lives. But that doesn't make headlines, and a grieving miscarried mom can't really connect it with all the trucks and sirens that have been going on around the plant. A nuclear engineer dies an early death from cancer, a decade after working through an accident that probably 99 out of 100 Americans don't even know happened, and no one can conclusively say that a "nuclear power event" killed him.

Five decades of safety indeed. If ever there were an argument for the "precautionary principle," nuclear energy is it.

Now undoubtedly, someone is going to have some problem with the authors I've listed, or some premise I've made. They aren't going to bring Tony Harrigan's father back. This is personal, and I do take it personally when people make ignorant, outright false claims about nuclear safety.

Perhaps I owe my life -- and the nuke apologists owe their entire "five decades" argument -- to Phil Harrigan, who may have given his life in preventing a fizzle yield explosion that could have killed millions in the vicinity of Lake Erie, and downwind.

Another round fired from the blunderbuss of fear. Pack every scary statement you can find into a comment and fire it off. Pick one issue, provide the supporting evidence, facts not hearsay, and we can talk.

"nukes produce almost no other pollution, except for spent fuel"

This is simply propaganda. All nukes regularly have releases of various sources. And of course occasionally something really bad happens. And the plants themselves are toxic waste for ever (ok, only for tens of thousands of years--I feel much better now.)

But as Leannan likes to say, not every thread needs to devolve into a fight between pro- and anti-nukers. You like 'em, I'm suspicious, at best. Not much more needs to be said.

I worked and listened and learned while in the plants-each one sends a few truckloads of low level waste(safe enough to live on top of the dump, literally)to Barnwell every year-stuff like used disposable coveralls,dust mops, and various pieces of worn out machinery.

I expect that once the fuel is out that you could weld up the access doors and pour a few more loads of reinforced concrete and any one of the nukes I have been in would last as long as an Egyptian pyramid.

Again, I am always amazed that people who are otherwise wise and properly cautious are so ready to throw this to the wind when it comes to nukes.

So no one is ever going to be curious about what is inside of this encasement? No one will ever have the will and means to excavate the site? What language will you use to warn people not to do so that will be legible for thousands of years? Do you know about language change?

Read for me the following:

Hwaet we gardena on geardagum
theodcyninga thrym gefrunnon.

Having trouble? It's English of just about one thousand years ago (first two lines of Beowulf--from memory so I might be off on spelling). Such a relatively short time span yields a language incomprehensible to all but those persistent (some may say foolish) enough to spend years studying it.

And are future curious and powerful men likely to listen to the warnings of future foolish philologists?

I wouldn't waste (excuse the pun) my time with written words of warning.
I assume that images such as this would still be recognizable as having to do with death. That coupled with the eventual actual deaths of the those that ventured into the areas that might still be contaminated with radioactive wastes would probably become part of the mythology even of an unscientific tribe of surviving humans and cause such sites to be avoided. Granted that's not much consolation.

Good points, but of course we associate such images with medieval superstition, and future societies may look back at us with similar disdain.

OR.. if the future has a lot of Hippie-Deadhead Intentional Commune-istas, they might see these cool, gloomy signs (I'm guessing they'd be huge and carved from the living rock, or set in brass.. ) and decide to have their parties there.

Between the lead, asbestos and various other toxic and radioactive compounds that we've dredged to the surface and left in refined and purified piles here and there, it'll be a wild but barely perceivable minefield to travel our old byways, once the signs and memories have faded..

Any body capable of breaking into a sealed containment would be technically very savvy-and equipped with some serious heavy machinery.You might remove the stone and rubble used to seal a pyramid by hand but I can assure you that you will not remove a couple of meter of first class concrete welllembedded with reinforcement steel by hand unless you bring some lunch-a decade or twos worth.

Once inside they would have to cut up concrete and very heavy steel and machinery-to what purpose?There is very little of anything else except wire inside a containment building.Not many would succeed and if they did whatever they brought out would be harmless as long as you stayed away from it.
Of course if you did succeed in removing a small piece of really "hot" steel from (say) the reactor piping you could maybe leave it in your worst enemies house disguised as a gift and it might eventually kill him and his family.
Safety is relative-people go into caves for the fun of it, lion hunting for the fun of it, into piracy on the open seas for material gain.

Any body smart enough to get in will know what he is up against, in general if not exactly.

Something dangerous.

You clearly have a deep and abiding faith. I will try to dissuade you no further.

I have lost my faith in many things over the years.God.Conservative politics.Liberal politics.The gold standard. Fiat currency.The agricultural science I took my degree in.So called innate goodness of mankind.Public schools.Obama.

I do still believe in the friendship and affection of my hound dog-but on the other hand I wouldn't want to count on it if I quit feeding him.

I still believe in four dimensional space time but that may be only because I am too dumb or ignorant to get my head around modern physics.

I don't BELIEVE in nuclear power, in any fundamental way-I just believe that the risks of using it are within reason-given the possible consequences of not using it- and that people are ALWAYS GOING TO DIE-from snakebite, from arrow and club wounds,from the revenge of jilted lovers,accidents, old age, starvation, being eaten by predators,disease,asteroid strikes, nuclear war, global warming and rising sea levels, contagious disease, lack of exercise,war, and a zillion other causes.

Given the gravity of our current state of affairs, I see your objections concerning the well being of hypothetical future human beings getting radiation poisoning as overblown.

Again-I simply estimate that nuclear power, ALL things considered, is a net plus for our species at this time.Life is uncetain, risk is a given,the eventual results any given course of action cannot be known in advance.

We can only look at the hand we hold and play it as seems most prudent in our estimation.

Maybe I'm wrong.

But maybe our working nukes will keep enough lights on , enough refrigerators running , and enough toilets flushing to save tens of millions of lives in the next few decades.

As opposed to an uncertain but small number a few hundred or thousands of years from now that might be lost as the result of nuclear pollution.

" As opposed to an uncertain but small number a few hundred or thousands of years from now that might be lost as the result of nuclear pollution. "

Mac, well said. What are the odds that 500 or 1000 years from now a large fraction of people will still be dying from cancer? If technology goes forward or backwards that fraction will be small, and we know technology is not going to stay the same.

Love it when you get on a roll mac. No beautiful tidbits like 'singin to the cows' (I'm keepin that one) this time but that was short and sweet. We can make it without faith but don't lose hope. Here's to hoping we are one of the best shows on cable ?-) Far as I can tell modern physics is a piece of cake, except there really is no piece or cake

Having trouble? It's English of just about one thousand years ago (first two lines of Beowulf--from memory so I might be off on spelling). Such a relatively short time span yields a language incomprehensible to all but those persistent (some may say foolish) enough to spend years studying it.

And the language has changed far, far less since the publication of the King James Bible fixed so much of it in written form in 1611.

The language of science is going to change even less over time, because it is so useful and so easily confirmed to be accurate.  The ability to read this language is going to be worthwhile no matter what, and if there are warnings written in it, they'll be taken seriously.

Not that I expect this to be a problem.  If we quit acting like paranoid idiots we'll make fast-neutron reactors to burn the actinides and other long-lived crap.  The vast majority of it will become fission products with half-lives less than 33 years.  A thousand years is 30 or more half-lives, reducing the original amount by a factor of at least a billion.  Those fission products will most likely be buried somewhere, not stored at plants.  All we have to do is tell the anti-nukes to stuff their objections and do what we already know (from the examples of others, like France) to work.

Did you see the map I posted? Are they going to pipe salt water to Indiana, Kansas, etc. Looks to me like the vast majority of US nuclear power plants are NOT on the coast. I post the map link again

Building more nuclear power plants on the coast is not wise. Some coastal areas are prone to Hurricanes, others such as in CA to tsunamis and earthquakes.

Saul Griffith suggests they'll be on the east coast:

East Coast gets hurricanes as well. And with ocean level rise from global warming no coast will be a good place to have a nuclear power plant. East Coast can get tsunamis as well as in 2005
"One of the more immediate issues scientists are dealing with is the probability of a mega tsunami hitting the eastern coast of the United States. According to the History Channel and Wikipedia, there is a possible landslide that could occur off of the northern tip of the African continent that could send over 100 million tons of rock slamming into the Atlantic Ocean. This place is known as the Canary Islands. This type of wave could send a massive tidal wave reaching up to 1 half kilometer into the sky. That is higher than any skyscraper in the world. The massive tidal wave would destroy anything in its path with sheer force. Those on the eastern coast of the United States would helplessly see a massive wall of water coming their way."

And there are earthquake fault lines in the east as well

Earthquakes, Faults, and Nuclear Power Plants in Southern New York and Northern New Jersey

1 Lamont-Doherty Geological Observatory of Columbia University, Palisades, New York 10964
2 Lamont-Doherty Geological Observatory and Department of Geological Sciences, Columbia University, New York 10027

Seismic activity in the greater New York City area is concentrated along several northeast-trending faults of which the Ramapo fault appears to be the most active. Three nuclear power plants at Indian Point, New York, are situated close to the Ramapo fault. For a reactor site in use for 40 years, the probability that the site will experience an intensity equal to or in excess of the design (safe shutdown) earthquake is estimated to be about 5 to 11 percent.

Indiana has no shortage of fresh water for cooling.  Neither do Illinois, Wisconsin, Minnesota, Michigan, Ohio, Pennsylvania or New York.  We can stand some evaporative losses, because they are down in the noise compared to natural exchanges.

As for tsunamis originating in the Canaries, I doubt that there's enough energy there to have a 500-meter high wall of water hit a thousand miles away.  If it's ten meters at the US coast, that would ruin the day for millions of people but the reactors could SCRAM with hours of warning and the containment buildings are more than strong enough to come through.  Turbine and control buildings would probably not, but we'd have worse problems to worry about.

This is just another, major side of the Limits to Growth theme. It is particularly applicable to tar sand / shale oil mining, i.e. why they cannot displace a significant percentage of conventional oil extraction.

Energy Return on Water Invested is a totally bogus concept and the data in the chart is obviously biased against biofuels. The data chart compares different forms of energy which have differing utilities, prices, and availability among other things. This is a logical fallacy. Basing a choice of which form of energy to produce on water usage is nonsense.

Most ethanol and biodiesel does not come from irrigated land. Why put up this data and not data for non irrigated? It is a straw man argument. The rain that falls is the same whether it falls on corn, soybeans or weeds. Using corn and soybeans for biofuels has no effect on water used. Weeds or any other crop will consume the same amount.

Why wasn't hydro power from dams included in the data set? The water "requirements" per megawatt hour must be phenomenal for dams producing electricity.

Water unlike oil is not consumed when it is "used" to produce energy. It merely changes form (evaporates), runs off or sinks into the ground and is used by plants which recycle it to fall again as rain.

My region of Iowa which is one of the largest producers of ethanol/biofuels in the world is suffering from about 9 inches of rain in the last 5 weeks. The crops are standing in water in some places and the harvest is proceeding at a snail's pace.

This article is total nonsense and so is the EROWI concept.

This article is total nonsense and so is the EROWI concept.

Obviously one must first take into consideration the WROEI of obtaining the water that is then used in the EROWI calculation. As an example diverting 10,000,000 gallons of water for some purpose out of the Amazon river will probably not have the same WROEI as extracting the same amount form deep underground in some ancient desert aquifer. Of course one must also know the EROEI of the E in the WROEI to make any sense of EROWI... is it all starting to make sense now? ;-)

As I write this it is pouring cats and dogs outside my window and this is the dry season here.

I don't think the EROWI idea is total nonsense. It is a component in decision making, it is NOT a "rule of thumb" or "figure of merit". I own a farm in Pennsylvania and was trained as an engineer. Where you might use such data is in areas where the re-charge of the water table was in question for making good policy, vs say the ridiculous policy on ethanol based fuel who's energy return is negative. A similiar table for plantings for food use would be useful for say the Chinese in their northern wheat region which hasn't recharged its aquafer probably since the Communists took over. Food and Fuel and a dozen other things are difficult to determine the "true costs" on, due to the intervention of government and the ignoring of impact to others outside of your boundaries. If I had this data for soybeans for food, It would influence me in certain areas to make long term decisions, including what to do with another farm I own in Africa.

Well said.

You make a valid complaint, but it doesn't invalidate the Water Use issue at all.. it just reminds us that there are much more complicated relationships to be considered.

As you rightly suggest, Biofuels use water almost entirely within the existing hydrologic systems, and as is often said with Trees and their Carbon content, that water is still within the cycles that it should be.

One vital thing the chart does not address (and does not claim to) is contamination, and ways that these power sources interrupt or defeat natural processes that also used that water, such as Heating of Rivers/Estuaries when water is used for coolant, whatever chemicals are leaching into waterways from Shale production, Tar Sands, Mountaintop removal, from Farming pesticides or excessive fertilizers, or in fact from Cleaning Solar Mirrors, etc. It's an issue where quality is easily as important as quantity.

I'm sorry you remain so inflexible about the concept of comparisons. They are absolutely central to figuring out what we're doing.


Agreed. The water use problem does have many facets that need to be considered. IMO, however, using corn or soy for fuel is off base, not because of water used but because in that case you are burning food. Some celulose sources are 'free,' because they grow where crops cannot unless irrigated. In the coming years the availability of food will become more problematic, not because of water [irrigation will at that time be too energy intensive to consider, and fertilizers will be mostly natural], and the stuipidity of rendering corn into alcohol will be obvious. Meanwhile, if people desist because of the water issue, so much the better.

One other thing, using corn to feed cattle is not really very bright either. Poultry is a better meat source, using far less corn. Best - rabbits! Watch of commercial rabbit farms - which use very little land, and far less grain per pound of meat produced. Also, IMO rabbit is delicious!

Ah, another knee-jerk defense of biofuels from x. I can't describe how not surprised I am.

When I hear rural folks concerns about biofuels, it's not about the poor EROEI, or the deleterious effect on GW, or any other of the many problems people have pointed out about it here and elsewhere.

It's about water.

As the chart accurately shows, biofuels use massive amounts of water, water that cannot then be used for other purposes. This is a real and vital problem. We can live with out fuel, but not without water. Already, aquifers are being depleted at an alarming rate. Bio-fuels are truly dooming those who are not directly benefiting financially from it to a water- depleted existence.

But hey, at least ethanol gives SUV drivers a smug feeling that they are being environmentally benign, so it's probably worth it.

Also, we had an unusually cool summer in Iowa, so global warming is obviously a hoax.

This is a truly bizarre report that does more to misinform the reader than it informs the reader. Following the links back, the true source of the data is a Sandia Lab report to congress.

This is an excellent report that explains clearly the difference between water withdrawal and water consumption.

If a farmer withdraws a million gallons of water from a river to irrigate a field, none of it returns (hopefully, since it would be contaminated with pesticide, fertilizer and would carry away valuable top soil). All of that water is consumed.

When a power plant withdraws a million gallons of water it is all returned to the river, actually a little more than a million gallons due to thermal expansion, but the same mass.

If the plant is well upstream from the ocean some of that water will evaporate over the next few days due to the higher temperature. That extra evaporation is the consumption.

From table B1 on page 65 of the report we see that a nuclear plant with cooling tower consumes about 590 gallons per MWh, and with open loop cooling only about 430.

The U.S. generates about 13.1 MWh per person per year. If all that was generated by one type of system our annual water consumption for electricity per person would be;

Coastal or offshore plants………………… 0.0 gal
Nuclear with dry cooling………………… 394 gal
Coal……………………………………… 5,300 gal
Nuclear with once through cooling……… 5.600 gal
Nuclear with cooling tower……………… 7,800 gal
Solar thermal trough……………………... 11,000 gal
Solar tower……………………………….. 10,000 gal
Geothermal……………………………….. 18,400 gal
Hydro (evaporation)……………………… 59,000 gal

Total water consumption/person/yr………. 121,000 gal

" Nuclear plants, for instance, seem to be especially vulnerable to water scarcity. During the past few years, several plants had to be shut or slowed down, or allowed to drain water into rivers at higher temperatures than considered safe. A set of references on the troubles of nuclear plants during heat waves can be found here . "

Most of the references are failed or old. The problem is not a characteristic inherent to nuclear power.

" According to Service's data, the problem can be eased moving from "once through" to "closed loop" cooling. But, if it were easy, it would have been done already. "

Actually it has been done. Travel down the Ohio River and you will see many power plants, nuclear and fossil, with cooling towers, so as to avoid overheating the river. Apparently France did not build enough cooling towers to handle extreme conditions. They can provide more cooling towers or wave the temperature limits if emergency conditions develop.

I assume water that evaporates into the atmosphere will eventually return as rain. Is there a "half-life" number that quantifies this or it more complex than I've described?

Sleepless, this depends on what kind of water one is using. If it is "fossil water", such as the water coming from deep aquifers, well, the time to reform may go from a few hundred years to hundreds of thousands of years, perhaps more. Depends on which aquifers. Saudi Arabian aquifers, for instance, are gone forever. Using river water, then it is rainwater; it will return every year. But it may not be there when you need it.

To support your point - water as a limiting factor.
Libya uses a river of fossil water from the giant aquifer.
The MENA region as a whole is looking at major water constraints on future farming related to population growth, not just further industrialization.
Melt water or rainfall drainage mostly evaporates or reaches the ocean, but humanity depends on doing something with the water as it passes, as those relying on the Himalaya are finding out as their situation changes.
The use of old aquifers is not confined to Saudi Arabia or Libya.
I posted the following in support of comments by memmel in May this year.
We can note that fossil energy is in turn a limiting factor in obtaining and using water.

There is an interesting chapter on Oglalla irrigation in Geoff Cunfer's 'On the Great Plains'that adds detail to your comment. His expert review as of 2005 concluded that Texas and the southern Great Plains had pretty much drained their end of the Oglalla aquifer, but Nebraska and the north might have a longer trajectory than Texas before irrigation declines, because it has a larger resource to draw from. But Cunfer writes

" Only with the development of underground water were farmers able to surpass the natural limits imposed by low rainfall. Irrigation worked only in conjunction with the development of large amounts of fossil-fuel energy to lift deep water. ... That solution built on technology, a cash economy, and imported supplies of energy, was only temporary. Farmers will eventually use up their underground water supply and will then be pulled back within natural limits imposed by climate. Likewise, interruption in the supply of gasoline, diesel fuel, natural gas, and electricity would doom irrigation regardless of water supplies. ..."

You are right, Phil, there is much more to the problem than just Saudi Arabia and Lybia. By the way, Lybians seem to have embarked in this big project called "the great man made river", which also exploits fossil water. They may have plenty of water for - how many years? Maybe a century, maybe a decade, maybe not even that. And then?

Then of course, there are the countries of the MENA regions and the Oglalla aquifer in the US. It is all part of a gigantic problem of irrigation and water consumption. I am involved, here, with a research proposal about irrigation in the MENA regions - hugely interesting, but also a disaster; especially in view of climate change, erosion, oil depletion, etcetera...

This is one of many reasons that the Third World must not be allowed to export its population to the First World.  Using fossil resources to build a population which must flee or starve when the resource is exhausted is a recipe for disaster, both for the society which does it and the societies where the refugees attempt to flee.  Closing off the door of emigration forces countries such as Libya to look at their own long-term problems with an eye toward solving them rather than kicking them down the road.  At worst, they become an example for others rather than a problem for others.

This is one of many reasons that the Third World must not be allowed to export its population to the First World. Using fossil resources to build a population which must flee or starve when the resource is exhausted is a recipe for disaster, both for the society which does it and the societies where the refugees attempt to flee.

I'm sure they would be fine with that if the the small number of residents of the first world didn't hog the majority of the world's fossil and all other resources in the first place.

And perhaps if the first world payed back the enormous quantities of human and non-human resources they have pirated from the third world for centuries.

And if we payed back the enormous ecological debt we have incurred by spoiling everyone's collective nest with our own spew, the hundreds of billions of tons of CO2 we've dumped into the atmosphere being just one example (though a pretty good sized one!).

(See Andrew Simms' book "Ecological Debt" for more information on this injustice.)

Fmaygar, Dohboi,

I am sympathetic to your arguments-but EP HAS a point-one that has been raised before by some well credentialed environmentalists such a Edward Abbey.

We have a choice -preservation of a part of the world and civilization, or the destruction of all of it.

If we allow the refugees to over run us, we will postpone thier day of reckoning at the expense of moving up our own.

I and several others here have been criticized for advocating a soft landing and a gradual power down as being insufficiently sensitive to the plight of the Earth itself.But a fast crash is actually far more likely to result in Armegeddon nuclear style than a nice clean planet cleansed of men.

As far as the robbery of the third world is concerned-Darwin and ther rest of the biologists laid it out for us.Right now we here in the states are being robbed by our banking industry.

It's unjust.But it's reality.I see little likelihood that this will ever change.After all, the chimps behave the same way.It looks kinda like this is built in behavior.

Anyone who thinks differently should read just about any book on anthropology written within the last decade or so.

The best recent one imo is BEFORE THE DAWN.

As a resident of the first world myself, with all that that entails, I at least am cognizant that there is a flip side to the coin that EP seems to think is one sided only.

Sometimes it is helpful to put oneself in the shoes (assuming they have any) of those on the other side of the divide between the haves and the have nots. I was just suggesting that some humility might be in order.

Then again I doubt that EP has had the opportunity to share a meal, drink and dance samba around a campfire with people in places like the shanty towns of Rio de Janeiro. Since I have had the opportunity to do so I can say from first hand experience that it has changed my perspective somewhat.

Its quite amazing how remarkably human those third worlders can be.

"I was just suggesting that some humility might be in order."

As was I.

It is of course always convenient for those who have benefited from a long history of injustice to convince themselves that we must resign ourselves to living in an unjust world.

For the record--

I generally don't think there should be massive migration;

And I do think there should be massive reparations for historical harms done--to third world countries, to native populations...most importantly to the land itself (that is, to the future).

I also think it is inevitable:

that there will in fact be massive migrations;

and that there is essentially no chance that there will be any such massive reparations.

Speaking of migration:
The largest and earliest mass migrations are likely to be from Oceania and South Asia, but I'm not sure where they will find to go. Islands are already being inundated yet no one is willing to take these early climate refugees (even though the island inhabitants contributed almost nothing to the problem).

In So. Asia, much of Bangladesh is simply going to be uninhabitable in coming decades, and the North Indian aquifer that supported much of the population explosion there is collapsing as we speak. Where are they going to go? Pakistan?? Burma/Myanmar?? Tibet?? Will they all hop on boats? To where? Africa? Australia? Every possibility seems bizarre and improbable.

Fmaygar, I do hear you.Never had the opportunity to travel widely, but like Thoreau I have seen a lot locally.

The black guys I worked with over the years are just as good as anybody else and mabe on the average better than most- rather than being the slightly less than humans I was raised to believe in.

The dirt poor people around here are more willing to give you a hand in a pinch than the well to do.

If I ever am destitute, I expect no help from my well to do liberal friends but the farm hands who have next to nothing will drop me off a load of firewood and a bushel of potatos.

My own family fled the oppression of the English in the nineteenth century.We would have played a serious role in depriving the "Indians" of thier lands except for the fact that we got here a little late for that.

I am all for doing whatever can be done to help out the people in the third world but I can't see allowing our society to be overrun to no useful end.
Maybe what we should do is divert half of our military research and development money into producing some sort of long acting birth control technology that can be used like a food supplement and put in in free food-well and truly labeled.

That way we could help in both the short term and the long term.

As a resident of the first world myself, with all that that entails, I at least am cognizant that there is a flip side to the coin that EP seems to think is one sided only.

When did I say it was one-sided?

Yes, individuals can improve their situation by migrating.  Unfortunately, it makes the overall problem harder to solve.  We're seeing this right now with the health-care debate; the eligbility verification requirements in the current bills are much looser than for other Federal assistance programs, but the House Hispanic Caucus has let it be known that they will torpedo the bill if they are raised to the standards of the SAVE Act (let alone E-Verify).  The demands of a group favoring non-citizens may cause the whole effort to fail.

The USA needs to put its efforts and capital into slashing consumption of fossil fuels (and carbon emissions).  Every new immigrant demands roads, schools, housing and medical care; all of these consume energy and capital.  The Mexican peasants and Somali pastoralists we're taking in are not the skilled tradesmen and engineers we need to design and build wind farms and nuclear powerplants; if we have to devote too many resources to the former, we will have no capital to devote to the latter.  This is ultimately suicidal for everyone.  Would-be migrants need to be kept home, in no small part for their own good.

Transfer of knowledge is a good thing.  Transfer of people too easily becomes a disaster.

One problem with migration is that it provides a safety valve for problems that need to be fixed.

The US needs to be much more aggressive about helping poor Mexicans get good education and job & entrepreneurial opportunities at home.

There have been quite a few 'refugees' from the UK/Eire at times, especially when the going got rough. We even sent some of our prisoners overseas for quite a while.
I sometimes wonder where any future refugees from here might be welcomed if and when push comes to shove again? We certainly have used up most of our resource base and multiplied our population ten fold over the 250 years.

This is one of many reasons that the Third World must not be allowed to export its population to the First World.

Too late (have you been to California lately?) But we're doing it to the third world in return. Take a look at the Southwest US and the toll that irrigation of the California Central and Imperial valleys, and consumption by cities like LA, Las Vegas, and Phoenix has had on the Owens Vally and the Colorado.

Actually I'm understating the complexity of the CA water system - see Aquafornia for a lot more information.

Twenty years ago when I lived in the SF area, I recall that agriculture consumed 90 percent of the state's water supply. Meanwhile, residents had to share showers and "let it mellow if it's yellow." Clearly, Ag had more lobbyists, and even with 36M inhabitants today I daresay Californians will be drinking from the arroyos before the fields run dry in Bakersfield.

We need to consider Peak Water and Peak Food with the same verve that TOD members have for Peak Oil. Why are those of us in the NE trying so hard to create local food economies? Because one day, with the Colorado and the Ogallala dry, we might just have to make it on our own. We might drive less or live with less electricity, but without food or water, well, life wouldn't be much fun.

"We need to consider Peak Water and Peak Food with the same verve that TOD members have for Peak Oil".

..........and Peak Population.

I think sleepness wants to know how much of the consumed water we will get back as enhanced rainfall. A general rule of thumb is that a molecule of H2O lasts a week or two in the atmosphere. Of course if it gets into the jet stream and moves at a hundred miles per hour it could get half way around the world before being rained out. So I think your answer is almost none comes back locally. If you are on the upwind side of a large continent, then maybe half will become someones elses rain. If you are on the east coast of the US -as your handle suggests, most will end up in the Atlantic ocean.

I'm not a promoter of Nuclear, but the numbers quoted in the chart for Nuclear are grossly misleading.

Hannahan, about your last point; yes, there are many plants using closed loop cooling. But by no means all of them. I made a small modification to the text to clarify this point.

Bardi, regarding your change;

" According to Service's data, the problem can be eased moving from "once through" to "closed loop" cooling. But, if it were easy, there would be no "once through" plants. "

The data shows that water consumption with a cooling tower is 37% higher than using once through cooling. So if both were equally easy once through would be preferable. That is one of the reasons plant sites near large heat sinks are desirable.

Ah, the joys of reductionism.

Open loop causes other problems; the small increment in water consumption for closed loop is negligible.

What other problems, and how big are they? Find out for yourself - it'll be a tiny first step in weaning yourself off reductionist thinking.

" What other problems, and how big are they? Find out for yourself - it'll be a tiny first step in weaning yourself off reductionist thinking. "

A transparent ploy to have me do your homework imbedded in an insult, no thanks.

If rigorous research shows that damage is being done I support putting a price on it and adding it to the cost of the electricity, just as we should price the effect of mercury, sulfur, particulates NOx, CO2 etc. and add those costs to the cost of electricity from coal and gas.

I support a level playing field.

On a level playing field cooling towers might be better than once through cooling, fine with me.

The term 'closed loop cooling' could mean both
1) towers with evaporative losses but most water being recycled
2) air cooling akin to an auto radiator.

If I understand it right supercritical coal plants (eg Kogan Ck, Queensland) use water which stays in the supercritical phase but changes density as it passes through fan cooled tubes. Note that in heat waves river water is also sprayed on the outside so it's not strictly closed loop. I've heard the amazing figure of 55 ML (say 15 million US gallons) daily evaporative losses from cooling towers from a 1 GW thermal plant. That's about 2000L per Mwh.

I notice in the TV footage of the Tour de France cycle race that nuclear plant next to the Loire River uses towers, not river level submerged heat exchangers. The conclusion would seem to be that new thermal plant should be on the coast with good inshore currents to dissipate heat. That is also where people like to live.

Dear Ugo,

Thanks for providing this interesting and useful data that many people ignore.

I have in mind that we are already consuming about 4,000 km3/year from the about 9,000 km3/year world cycle of fresh water accesible to man.

And for the solar thermal (solar thermoelectric) plants, we have already some experience and figures:

For a typical 50 MW plant, producing some 75,000 MWh/year, the plant is consuming, between refrigeration and washing of the mirrors, between 100,000 (the most efficient) and 300,000 m3 (the most common) of fresh water per year. for washing, it needs to be deionized.

This implies a consumption of between 1,333 and 4,000 liters per MWh. In the middle of your table.

Thanks for that addition. Can you offer any information on the quality of the water leaving the system? Are there detergents, surfactants or other chemicals that have to be removed or would otherwise reduce the water quality coming out of a Solar Facility?

I'm not anti CSP, but I do want to know the bad news if I can get it.


I am neither against CSP. Just checking out figures in real world.

Solar thermal plants, to the best of my knowledge, do not use many chemicals. Mostly many filters and membranes or chemical products to decalcify in their own internal water plants. Basically, most of the water usage is refrigeration (lost in the form of steam). The water for washing the mirrors has to be very clean, demineralized, desalinated, decalcified and deionized, but it is a tiny part of it.

The question is that these solar plants were intended to be located in the sunniest places of Spain, basically in the Southern regions. The promoters have slightly changed the recommended regions since their original, primitive prospects. They moved from Southeast regions (Almeria, Murcia), which are very scarce in fresh water, to the Southwest regions, with some more fresh water. The reason behind has been the negative of several administrations to grant permissions to exploit for the solar thermal (CSP) the needed flow of 6 liters/second for each 50 MW plant. These flows come from either subterranean fast depleting fossil water tables or from desalination plants, and are in frontal competition with the big users: the biggest greenhouse winter vegetable production compounds of all Europe (see, for instance in Google maps the plastic ocean at El Ejido at 36º 44' 41,15" N and 2º 44' 25,40" W from 60 km heights and lower)

Interestingly enough, the area is strongly demandingg desalination plants, to meet the ever growing local water needs, which, in turn, consume a lot of electric energy.

A first approach. Considering that 1 m3 of desalinated water costs 4 kWh of electric energy, a 50 MW solar thermal plant needs, at 6 liters/second = 21.6 m3/hour = 86.4 kWh/hour = 757 MWh/year or about 1 percent of the generated electgricity (other start up, build up or financial expenses excluded). This has to be added to the burden of the selfconsumption of the plant, between a 2 and 4 percent of the total generation (transmission losses to the grid, excluded)

Spain already uses 600 million cubic meters of desalinated fresh water/year, a 2 percent of the total fresh water.

Two approaches from this figure:

1. Water to electric energy. Present desalination volume of water in Spain would need some 1.2 GW solar thermal plants.

2. Electric energy to water. Present electricity consumption in Spain is about 300 TWh/year. If made through this type of renewables, it will demand some 4,000 typical plants of 50 MW and a refrigerating flow of 24 m3/second, a little bit more than the whole Tajo river streeam of today in its middle course, the longest river of Spain. In fact, thjis is more or less in line with the 20% of all river streams, used today to refirgerate generating plants in Spain.

Pedro Prieto,
I assume that the typical CSP (tower type) plants in Spain rely on cooling towers in order to condense the steam.
Are there any CSP plants that do the condensation process by air cooling ?
Using air to condense the steam would require a lot of energy (for the fans), but saves water.

The question is that most of the CSP plants in Spain are parabolic trough mirror plants, rather than central tower type with flat mirror around and to the best of my knowledge, they do not use cooling towers. There are plans to use air cooling, but I have not seen any project online as yet. Again, a compromise: water versus energy.

Very dry places like Afghanistan find water to be a limiting resource in putting up electrical power plants. Heading Out wrote about the situation on his blog.

I am a big fan of Ugo's article's and realize the table provided is not presented as a best estimate, but as one rough order of magnitude to raise awareness and stimulate discussion about the role of water in energy conversion. A more balanced table would have addressed the following;

- Only oil shale surface retort is mentioned; in situ is by far the most viable and the most likely technique requires the pumping and dumping of all of the groundwater in the area being mined.

- Standard coal fired power plant water usage is completely missing

- The coal IGCC does not take into account the pollution of surface and ground water caused by coal mountaintop removal or standard mining practices

A coal fired power plant with cooling towers with 3% tower makeup water requires about 10-20 barrels of water per MWh.

A cooling tower uses about 1/100th of the water at a 10 degree rise(thermal pollution) that a once-thru cooling system requires with negligible fan power.

This makes seawater cooling claims ridiculous.

A dry cooler uses perhaps 20 times the fan power that a cooling tower does and 1-2% of power plant output is needed to run those fans.

" A cooling tower uses about 1/100th of the water at a 10 degree rise(thermal pollution) that a once-thru cooling system…"

A nuclear plant with a once through condenser flows about 45,000 gallons of water per MWh. A hydro plant with a 100 foot drop flows about 3,500,000 gallons per MWh. But in each case the water is available for other uses after flowing through the plant.

Comparing flow is misleading. Water consumption is what counts.

As usual everyone gets involved in arguing about the data presented rather than using it to further the sense that this dilemma we are in is huge and multifaceted. As I pointed out in a previous comment, global warming will affect nuclear power and has already caused nuclear power plants to shut down as in 2003 in France. As another person pointed out water requires energy to be available and with what is likely peak ground water upon us it takes more and more energy to get that water. Global warming will increase our need for energy to hold even at this level of life, but using energy increases global warming. Other resources are taking more energy to find and process (uranium included). Everything that is is connected but we persist in trying to deal with problems piecemeal, probably because EVERYTHING is too big for us to deal with, yet everything that is is the world that we are adapted to live in. We meddle with the very source of our life.

EROWI is why I mostly look at crop byproducts rather than dedicated fuel crops†.  The water invested was used to produce the primary crop, and little or no additional expenditure is required to harvest the inedible parts of the plant.

† If land is planted in perennials for e.g. erosion control, the selection of a high-yielding biomass cultivar isn't such a big deal.

The harvesting of stalks, leaves, and stems that would normally be plowed back underground as green fertilizer raises the cost of nitrogen-phosphate-potash fertilizer required to grow the plants. The NPK fertilizer has non-renewable components of phosphate and potassium elements. The mining of phophate rock and sylvite is rapidly depleting global reserves.

Great! So now we have to worry about peak phosphate rock and sylvite!

Seriously, though. Our farm soil is being depleted very rapidly through factory farming, using high N2 fertilizer made from natural gas [and, lest I be chided for neglecting it, gas produced from coal is also available for that purpose, though it does reduce total coal as a resource even faster than it is already being depleted].

The point is that, when the soil is gone, it takes a few thousand years to rebuild it to what we once had in the central US. Some of our near coastal and plains farms are already about shot.

Using our biomass, through composting and use of composting toilets can help to slow or reverse this process. In what I see as our necessary future, with no commercial fertilizers and little fuel for tractors, our farm animals will, once again, contribute their perfume to the hinterlands, as we find our new limits.

If we can preserve enough of our fossil fuels and water, it could be a fairly good, and comfortable life. Harder physical labor, and a different sort of life, but not bad if we can make it through the transition safely.


We could do away with about half of our soil depletion and fertilizer problems by slapping a ten dollar a pound tax on beef, pork, and dog food.

Just to keep things in perspective politically do I mention this;>)

I think the whole purpose of human society has to shift from maximizing us to giving back to the land--not only halting erosion, but replanting native grasses and other plants that increase soil levels and soil fertility over time.

Essentially, we have to stop robbing from the future and start investing in it.

(And I do like farmer's idea of a ten dollar tax on meat and meat products.)

The dependency of electrical power on water supplies is a relationship often unseen.

When I worked with the NYC-DEP to document part of their Third Water Tunnel project in Brooklyn (Brooklyn was just the part I was involved with), the engineers mentioned that if the Tunnels were to stop flowing (cave-ins/valve failures, etc), electricity production in the city would have to shut down, and without Water or Electricity, that would necessitate the evacuation of the affected parts of the city.

I don't know how much NYC would be able to lean on the grid from surrounding boroughs/counties if it lost a significant chunk of its generation, but that's the conclusion DEP was working with.

Of course, firefighting and a number of other assumed-available services would be undermined as well.. so tunnel three went in to offer another parallel source, in order to facilitate maintenance on the older ones (built ca. 1937)

While I acknowledge that the availability of usable water resources is a growing problem, one closely linked to no only gross population growth but also to the distribution of that population, I feel that the concept of EROWI as something akin to EROEI is highly flawed for several reasons.

First, after energy is put to some useful purpose, such as in a power plant, running a refinery, or extracting oil from tar sands, it is essentially lost forever (at least as a usable entity).

On the other hand, as an example, the once-through flow of cooling water through a power plant essentially entails no increase in the entropy of the water itself. It takes energy to move that water, but the water itself is still in the same physical state as when it entered (albeit somewhat warmer). If that water is taken from a river, it merely represents a temporary diverting of a portion of the river's total flow in the vicinity of the plant. Hence the river's total flow rate upstream and downstream of the plant essentially stays the same.

The situation is a bit more complex if the power plant has cooling water recirculation system that includes a cooling tower. In that case, due to evaporative losses (the primary mechanism for cooling), the required cooling water make-up has to be larger than the cooling water blowdown stream discharged back to the river. However, and this is a highly location-specific factor, that evaporated water is part of the hydrological cycle and will eventually return to some body of water sometime and somewhere be it a river, lake, or groundwater regime. Thus, unlike the burning of a fuel, the evaporated water is not lost forever.

Despite water shortages in specific regions, the earth's hydrological cycle is self-sustaining over the long run, or as long as the sun shines and acts as the engine to drive the cycle. Man can screw things up on a local or regional basis, but the cycle still keeps rolling on.

For these reasons, I feel that EROEI and this so-called EROWI are not even close to being related concepts. They are hardly even analogous.

In this month IEEE spectrum, there is an article on the impact of Energy scenarios on land/water requirements:

The model is interactive and can be manipulated here:

They evaluated the water requirements in terms of fraction of available rainfall (does not take into account the presence of aquifers, etc.). It`s too bad they did not look at the impact of peak oil and they are evaluating the various mixture of energy sources and not their absolute levels. They also think that if we want higher standards of living, we will get it :):

Many futuristic energy models, including some much more complicated and sophisticated than ours, constrain future economic development in the name of sustainability. On both moral grounds and as a matter of practical politics, we reject such constraints. There is no justifying a regime that would limit a large fraction of the world's population to a much lower level of prosperity than North Americans, Europeans, and Japanese now enjoy, nor would Brazilians, Chinese, or Indians agree to such an arrangement. Therefore, in evaluating the demands that the consumption of energy in whatever form would place on the earth, the water supply, and the atmosphere, we assumed that nobody in the world would be denied the aspiration of reaching the current U.S. level of consumption and prosperity just because available resources have been consumed by the wealthier and more developed nations. Some may quibble that our growth projections are ambitious, and so they are; our key assertion is that sufficient energy must be available to enable economic growth if the average world citizen is to live at current U.S. levels by 2030 or 2050.

"we assumed that nobody in the world would be denied the aspiration of reaching the current U.S. level of consumption and prosperity"

Quite an assumption indeed.

A surprising amount of water usage for the pure solar option. Is that for manufacturing? Or are they using a lot of water in Concentrated Solar Thermal Power arrays?

My reaction: clearly the world is overpopulated.

The land and water numbers for solar appear way too high: world solar insolation is 100,000TW, and current world human energy consumption is only 12TW!

Plus, the optimal solution would have at least as much wind as solar, and probably much more.

I was waiting for our paper to be published (I was told Jan or Feb 2010 issue - it was written in early 2007..;-) before posting EROWI on TheOilDrum - but I'm glad Ugo brought up the importance of the concept. Our paper certainly won't be the final word in creating a methodology to compare costs in energy and water terms -as we know EROI itself is loosey-goosey and adding water -which can be measured via both consumption and withdrawals makes such a metric even more complicated. Here are a few relevant excerpts from 'The Energy Return on Water Invested':

Furthermore, the mixing of the energy quality of both inputs and outputs highlights an ongoing problem with net energy studies. Not only are all BTUs unequal in their value to society, but the markets pricing hierarchy of energy ‘types’ by cost, may not correlate with long term scarcity. The quality issue is further complicated by cost/benefit tradeoffs within different energy/water technologies that could increase EROWI while decreasing EROI just by altering how an input is procured. For example, nitrogen fertilizers (the dominant energy cost of fertilizers) are mostly produced using natural gas, but future electricity could be generated from a different subset of primary energy sources, lowering the energy input for biofuels. As energy is also an input for irrigation and water delivery systems, an interesting and relevant follow-up to this paper might be an analysis of the Water Return on Energy Invested.

Finally, demand side policy changes may have water implications just as will the supply side. The current move towards electric vehicles, without a major change in the sources of electricity would create major new water demand. If hybrid/electric cars would fully replace gasoline vehicles, approximately three times more water is consumed and 17 times more water is withdrawn, primarily due to increased water cooling of increased thermoelectric generation. (43) Furthermore, demand side moves away from meat consumption would allow more land to be used for bioenergy as the water/land intensity is much lower for vegetarian than meat intensive diets. (44) As such, future refinements to an energy and water framework will likely have to extend beyond those two vital commodities.

Our work here, looking only at water demand, predicts:
• the development of bioenergy in scale sufficient to be a significant source of energy will likely have a strong, negative impact upon the availability of fresh water;
• Assuming the water requirements for infrastructure development are minimal, technologies such as solar and wind which do not require on-going water inputs will be at an advantage in many contexts.

Above all, we believe our analysis demonstrates that energy technologies must be assessed in a multi-criteria framework and not just from the perspective of energy alone. Ultimately, we should strive to have a renewable portfolio aggregating the highest returns on our most limiting inputs.

So are these excerpts from your paper or the one Ugo posted? And is your paper finally going to see the light of day?

excerpts from my paper - yes, after 3 years it is finally going to see light of day. in retrospect I should have put it online somewhere so others could advance the methodology. It came within whisker of being accepted at SCIENCE, but back then I don't think these concepts were popular with their readership - still might not be but obviously (based on Ugos link) they are starting to be. Ultimately multi-criteria analysis should be better integrated to look at finite/critical resources like soil/water/oil that have no real substitutes. And energy needs to be treated outside the production function as we can't live off the drippings of Ghawar forever...

This may have been hinted at in earlier comments but there are major energy savings in multiflash distillation compared to reverse osmosis if you have a heat source such as a thermal power plant. RO requires seawater to be pumped through membrane tubes at 27 bar or 400 psi if I recall, then periodically back flushed. In multiflash 'free' hot seawater is sprayed into a chamber then given a momentary pressure drop. Fresh water then condenses on the walls of the chamber. That hot water has already helped cool the exterior heat exchanger on the thermal plant and is a near freebie.

I understand multiflash needs 20-50% of the electrical pumping effort of RO. Since coal is supposed to be phased out or something that new thermal plant would be nuclear. However people want it all their own way; fresh water, living on the coast and not living next to a nuke.

Remember that the primary loop of PWRs and the steam cycle loop of BWRs is contaminated and that tritium has a bad habit of migrating through plant systems. I wouldn't feel comfortable linking a source of fresh water to a nuclear turbine's extraction steam.

As usual, people tend to ignore the economic side of the debate. Where water is scarce, it has been more or less "commoditized." If the price is right, the local energy generators will be designed to optimally use water. An air cooled heat exchanger isn't exactly magic.

People knock the thermal output of power plants on the ocean, but that's usually where the fishermen head towards to get the best fish. If you compare the effect of adding heat from power plants to the ocean with people dumping trash and chemicals in the ocean, I don't think you should have a problem with the plants.

Why on Earth would someone make a table like this and list open loop Nuclear plant cooling without listing open loop coal plant cooling?

The only reason that nuclear, coal, and solar thermal use different amounts of water per MWh is because of the different boiling temperatures, which can be simply stated. The diversity of technologies for the cooling system of the plant are applicable to every kind of plant.

I agree with the other commentators that this table (and the whole article) is doing far more to mislead rather than anything else. A reader should first be educated on what a thermal cycle is, then start the comparisons.

May I slightly whoppingly digress to the concept of EROHI, Energy Return on Human Input, which I have briefly mentioned before. I suspect it will also be important.

For instance, if in the future to get to those vast coal reserves people would have to personally climb down hundreds of metres and manually chip out low quality coal then manually raise it up, then it may not be worth their bother relative to other life-choices.

Also the mental/cognitive/time factors as well as the muscle factor. Picking isolated peanuts off the ground might have great eroei and yet still involve too much time and mental effort, hence fail on EROHI.

"For instance, if in the future to get to those vast coal reserves people would have to personally climb down hundreds of metres and manually chip out low quality coal then manually raise it up, then it may not be worth their bother relative to other life-choices."

Ah, but if someone can get slaves to do it, it may well be worth their while.

"if someone can get slaves to do it"
Perhaps yes, but coal mining has to be one of the most unpleasant of tasks and there's a limit to how much the effort of controlling slaves repays itself. Haven't you ever thought "it would be easier to do this myself". (I suspect that a coalmine of slaves would be very vulnerable to suicidal sabotage explosions destroying the mine and slaves and hence seriously unprofitable.)
Anyway, my main point is I suspect there will be other ways in which EROHI unexpectedly complicates the future.


I wonder why no mention of wind & solar?

IIRC a recent new scientist article had figures for solar power

they were calculating the feasibility of large scale sahara solar hooked up to a euro-asian grid

issue 2731


Nick, I don't know why no mention of wind and solar. I suppose it is because such things as PV or standard wind don't need any water. But the table is incomplete in many things. Even for "Solar" water is needed for thermal solar, concentrating solar, and other variations of the concept

EROWI is an interesting concept - as is EROHI - but neither are comparable to EROEI. EROEI sets an absolute limit - you must always use <1MWh to produce 1MWh, otherwise the game is up. Doesn't matter where you are, or what value you place on 1MWh. (OK, I know EROEI goes below 1 when you're converting coal to electricity, but that's end-user conversion rather than production.)

In contrast, the significance of an EROWI number depends critically on whereabouts you are. In Libya, using a million litres of fresh water to produce 1MWh of energy might not make sense. But up here it would, because fresh water is something we have a superabundance of, so its value is negligible compared to the value of energy.

Similarly, the significance of EROHI depends on the relative value of the human individual versus the value of energy, and this too varies. A poor society with a rapidly rising population, keen to industrialise, may pay a high price in human toil and miners lives for the energy it needs - e.g. India/China today, or ourselves 200 years ago. In wartime, the Operation Pedestal convoy in August 1942 lost 12 ships, to get one aviation fuel tanker through to Malta. The mission was a success, but the EROHI must have been completely off the scale...

In our paper we illustrate a 'net EROWI' - which layers water use per MJ on top of EROI figures - so it combines both - not just the water needs.

To calculate a gross EROWI we attempted to estimate the total water requirements per unit of energy produced. Where data allowed, we estimated separate EROWI measures for both water withdrawals and water consumed. Water consumed is likely to be much smaller than that which is actually withdrawn, and for this reason the data available generally only indicate water withdrawals.

While the gross energy returned per unit of water invested is of interest, some technologies demand a relatively large energy investment as indicated by the EROEI. For this reason, following Giampietro et al., we used estimates of each technology’s EROEI to calculate a ‘net EROWI’. From both a policy and technology perspective it is the net EROWI that we are interested in because for the process to be sustainable, some of the energy yield must be reinvested as indicated by the EROEI. Thus:

where ω = EROEI / (EROEI – 1), which is the amount of energy production required to yield 1 unit of net energy. Note that ω increases non-linearly with declining EROEI, approaching infinity as EROEI approaches 1. Equivalently, Net EROWI approaches 0.

Here is a summary graphic (on log scale):

net energy = Eoutput-Einvest
So (EROI-1)/EROI = Net energy/ Eoutput
If EROI = 1 then net energy = 0
If EROI = infinity then net energy = Eoutput

Missing from the table is the kw/hr water usage of shale gas extraction. Is this known?

Of course in all this, it's not simply a matter of usage, but how much is polluted, made unusable.

My question also, I went through all the comments assuming someone would have supplied this value.

This gentleman is not right and to use his information on your web site is a dis service to the hard work you go though to post good information. I know of not one farmer in five states who's crop goes to ethanol or bio diesel that requires irrigating. This is pure bunk. Sweet corn is irrigated but that is not used to make ethanol. This man is too far from the farm.

Lonn, I wonder why you think rain water should not be counted. The water in rivers arrived as rain or snow.

If you were not growing corn for ethanol you could be growing food. If the price of water gets very high you could make more money collecting it and selling it than you make growing a crop.

Ugo - in the table up top, closed loop NGCC has a range from 230 to 30,300 litres / MW hr - actually overlapping with the open loop. Is this a typo?

Euan, give a look to the original paper from where the data are taken. There is a link in the text. It should be explained there.

Re the table of values for water use per MWh, can anyone tell me whether this represents water withdrawn for use or the water consumed?

I would much appreciate knowing where such information might be found, as well as sources of water consumption for solar thermal, PV, and wind devices

Solar Thermal at the Residential level might be a really useful number to find, considering the closed-loop systems probably use under 40 gallons for a stretch of a few years before having to mix up more glycol. Of course, the quality of the used, dirtied water must also be known and weighed in.

Howard, the data come from this link.

which contains more details but, in the end, for the full explanation of how the data are measured it refers to this document:

Energy Demands on Water Resources; Report to Congress on the
Interdependency of Energy and Water; U.S. Department of
Energy: Washington, DC, 2006; p 80.

which, unfortunately, doesn't seem to be available on line. So, I am afraid that we have to take the table only as a qualitative "order of magnitude" piece of information. If I find a link to the original source, I'll post it here.

Actually, it was easy to find. I had missed it at first, but it is here

In any case, the way I understand the point is that in the table they refer to water withdrawn. In an open loop cooling system, almost all of this water is returned to the source. The problem is one of availability of water and of temperature. If the water that goes back to the river is too hot, then it does damage. In the case of closed loop systems, water is evaporated. Much less of it is consumed, but more is lost.

I'm amazed that the author could write this article without even mentioning Canada's tar sands, which is the largest single supplier to the US of unconventional oil.

I also note that the author chose only to measure Liters per megawatt-hour. Omitted was the obvious barrels per barrel, since the prime use of water is recovery of liquids - shale oil, tar from tar sands, etc.

In the mind of a typical audience, how are they going to relate some abstract number like L/Mw-hr? Certainly the measure has a use - but the omission of the other set - the volumetric equality with recovered forms that are also volumetric - prevents the audience from truly grasping the horrific scope of the ecological and moral disaster which exists in the production of energy liquids from or by using water.

From: Water Use in Canada's Oil Sands (PDF)

"Net fresh water use in oil sands production today averages about 4 barrels of water per barrel of oil produced by mining operations..."

From: Canada's Oil Sands - Opportunities and Challenges to 2015: An Update Questions and Answers

"The water requirement ranges from 2 to 4.5 cubic metres of water to produce one cubic metre of synthetic crude oil in a mining operation."

And what matters just as much as how much water is used, is what is being done with the water AFTER it has been used? Today, as you read this, what is happening NOW. How TOXIC that water is, and WHERE it GOES.

In Canada all that TOXIC soup full of Lead, Mercury, Cadmium, Arsenic, carcinogenic Polynucleic Aromatic Hydrocarbons and more, that vile, lethal water is 'temporarily' impounded in vast toxic lakes that leach into the ground and whose walls break releasing billions of cubic meters of this deadly crap into the surrounding communities, into the surrounding environment.

The toxic soup continues to accumulate rising ever higher and making an ever-larger danger to everything and everyone nearby.

It is this growing nightmare that exists because of the fuckers driving their fucking world-killing cars. Those subhumans who by their actions cause risk to my life every day on the streets - it is these creatures whose money and passive complicity is funding and enabling such horrors.

The customers buying such product from the criminal corporations deserve to be drowned in the same lethal ponds of shit they are creating like those huge flocks of birds who are dying all the time to fund their filthy fucking driving habit.

And the executives and employees of such firms ought to be required to drink, wash, and otherwise use exclusively the effluent of their endeavours in their own magnificent homes.

They also ought to pay the cost of the custom piping and labor required to bring the freshest most saturated polluted water directly from its point of origin to their home's water hookup.

In comparing the table, water requirements for energy production given in this post with its putative source, I can't see any connection made between water withdrawals for energy production and consumption of the withdrawn water. The numbers in the table do not match those in the report cited, and both differ greatly from other published sources, e.g. A confusing (to me) matter is that withdrawals are much larger than consumption where both values are distinguished. If not all of the water withdrawn is consumed, what happens to it? The answer I am given is that it is returned to its source, but this is a very messy issue. Large fractions of cooling water in energy operations are released to the atmosphere through cooling towers or evaporation from retention ponds, but the route for that water to return to its source [via the hydrologic cycle] is both very long-term and uncertain for any particular source.

This is particularly evident for water use in desert power plants which are largely dependent on groundwater. Release of that water to the atmosphere is not likely to find its way back to the aquifer from which it was withdrawn.

The real issues, it seems to me, are depletion of the immediate sources of water used in energy operations, and contamination of water returned to its source (consumption of fresh water not water per se).

If I am off track, please set me straight.