May 29, 2006
Affordable Energy Can Solve Water Shortages

Some folks like to fantasize and prophesize about how this or that trend is going to eventually bring about a collapse of civilization. FuturePundit has a few worries about the continued existence of humanity. But most of the scenarios about civilizational collapse are extremely unlikely. One scenario promoted in some circles is a worldwide shortage of fresh water. The exhaustion of fresh water resources poses a problem for the poorer countries in the world. But industrialized economies which can afford to produce large amounts of energy can not collapse due to shortages of fresh water. A Technology Review survey of the water desalinization industry in Spain provides some useful data about water desal costs.

One of the main challenges that remains with the desalination process is the cost of the energy required to produce freshwater. Though different processes demand varying amounts of energy (desalting seawater with membranes requires the most, as it takes tremendous pressure to push the water through the membrane), it remains an issue in terms of cost and environmental issues, as nations around the world battle rising greenhouse gas emissions, such as those emitted by power stations.

In the last 30 years, the amount of energy required for desalination has fallen precipitously, and along with it the price. Decades ago it took approximately 12 kilowatt-hours of energy to produce one cubic meter of freshwater using RO technology; today it takes on average between 3 and 4 kilowatt-hours of energy. Even today, however, the cost of that energy makes up about 40 percent of the total cost to produce each cubic meter of water.

“We are very close to the minimum energy for desalination,” says Juan Maria Galtés, director of special projects for Inima. “There’s a point where it’s impossible to go any further,” because of the high pressure needed to separate salt from water.

Developments in new kinds of membranes or other tweaks in plant efficiency could help engineers continue to shave off small amounts of energy, reducing both the cost and the environmental impact.

A cubic meter is 264.173 gallons. Electric costs vary within and between countries. But 10 cents a kilowatt-hour for retail sales is close enough to the average to be useful for rough calculations. The 4 kilowatt-hours needed to produce 1 cubic meter of water therefore cost about 40 cents. Since the article claims energy is only 40% of the cost of desalinization that suggests desal costs about $1 per cubic meter (these are high side estimates btw). Maybe the Spaniards assumed higher or lower cost per kwh of electricity. So that's a rough guess. But a useful one. See below where I connect that number to information about how much water the United States uses.

New nuclear plants probably could produce electricity for less than 5 cents per kwh. But what about solar and wind? They cost more. But in theory their lack of continuous availability shouldn't pose a problem since we could store up purified water when the sun shines and the wind blows. But the problem with this idea is that current designs of desal plants require continuous availability of power. However, industry and governments are looking for ways to make desal from non-continuuos energy sources more feasible.

The engineering involved in using renewable energy to power a desalination plant can be relatively simple: solar or wind generators can be hooked up to an existing utility grid, which then offsets the power demands of the desalination plant.

The challenge, however, in coupling desalination directly with renewable energy such as solar or wind power lies in the variability of renewable energy. The membranes used in reverse osmosis need to be kept wet, and the systems that make up a desalination plant have been developed to handle a steady stream of water. Solar energy is plentiful when the sun shines and wind power only when the wind blows.

Researchers in the Canary Islands have spent the past decade developing stand-alone small plants that could provide water for approximately 100 to 300 families, about the size of a small village in a developing country. ITC projects are also carried out in conjunction with other international research institutes or companies.

On one Canary Island test site, photo-voltaic panels are hooked up to a battery, which feeds a steady supply of electricity to a small desalination plant. “But batteries aren’t great because you have to replace them after, say, five or 10 years, and then you have to dispose of them as well,” says Piernavieja. “It’s better to develop a system that needs no batteries in the first place.”

Some day really cheap photovoltaics combined with cheaper equipment for doing desalinization will make desalinization much cheaper. But for now nuclear power would be tbe best way to scale up desalinization - unless you are willing to either let coal burning electric plants pollute or to pay much more for cleaner coal burning plants.

I like to work out costs of switching from current methods of getting resources to alternative methods because the costs of alternatives represent worst cases for what happens if we really do run out of this or that resource. Suppose the United States had to totally switch over to using only desalinated water. The US uses 40 billion gallons of water per day.

A report by the U. S. Geological Survey (USGS), "Estimated use of water in the United States in 2000" (USGS Circular 1268), shows that about 408 billion gallons of water per day were withdrawn for use in the United States during 2000. Withdrawals in 1990 averaged nearly 1,620 gallons per day per person; in 2000, the per capita average had declined to about 1,430 gallons per day. During the same decade, the United States experienced a population increase of about 33 million. Total withdrawals increased steadily from 1950 to 1980 but have varied less than 3 percent since 1985.

That 408 billion gallons of water per day amounts to about 1.544 billion cubic meters. Well, if desalinization of water costs about $1 per cubic meter then it would cost about $1.54 billion per day or $563.7 billion per year to get all water in the United States from desalinization. That would be rather expensive and the price of water would get so high that people would find many ways to drastically cut their water usage. So it is unlikely we'd ever spend a half trillion dollars a year on water even if the United States suffered a massive continent-wide drought. But since the United States has a $12.4 trillion a year (and growing) economy it could afford to spend a half trillion a year on water.

A massive shift to desalinization plants could be done in concert with a nuclear power plant building program to lower the cost of electricity. Plus, huge demand for desal would drive the development of cheaper desal technology. So the cost of desalinated water would drop for a couple of reasons. At the same time, the higher price for desalinated water would drive demand for a shift to technologies that used water more efficiently and therefore lowered the demand for water. As a consequence of all this I'd be surprised if a shift to desalinated water would cost more than a couple hundred billion dollars per year. To put that in perspective, our current approximately 21 million barrels of oil consumed per day in the United States works out to about 7.6 billion barrels per year. Well, an oil price rise from $20 to $60 per barrel cost about $300 billion per year. Worse yet, most of that money was exported.

A continent-wide drought is probably not possible. But if it happened our need for crop irrigation would rise to compensate for lack of rains. However, we could switch to the methods of agriculture used in Israel and reduce evaporation with ground covering, no-till farming, and plant more crops with lower water needs.

Mind you, I do not expect a massive nationwide drought that dries up all the rivers. IRather, the lesson here is that shortages of other resources can be solved given energy which is cheap enough. Given sufficiently energy we can solve other resource limitation problems by performing filtration, purification, and many other processes on readily available forms of matter to produce anything we might need for food, shelter, transportation, and other needs.

Some might argue we couldn't run massive desal plants because we are headed for an economic collapse due to a world wide peak and decline in oil production. But we can convert coal into liquid fuels for about $40 to $45 per barrel. Plus, we can run the desal plants on nuclear power. So sorry doomsters. There's no bone dry Mad Max Peak Oil Apocalypse awaiting us in the future.

Update: About half of the water used in the United States is used for electric plant cooling.

Thermoelectric-power plants accounted for 48 percent of total withdrawals (195,000 million gallons per day [Mgal/d]) in 2000. Surface water was the source for more than 99 percent of total thermoelectric-power withdrawals, one third of which were saline. Historically, large supplies of water, mainly for cooling, had to be available to operate thermoelectric-power plants. For this reason, large power plants have been sited near the oceans, the Great Lakes, the Gulf of Mexico, and large rivers.

Withdrawals for irrigation were about 137,000 Mgal/d, second only to thermoelectric power nationwide. Irrigation represented 34 percent of total withdrawals and 40 percent of total freshwater withdrawals. Eighty-six percent of irrigation withdrawals, and 75 percent of the total land irrigated in 2000 were in the 17 conterminous Western States. Withdrawals for irrigation have remained nearly stable since 1985 despite an 8-percent increase in total acres irrigated.

Since salt water is usable for power plant cooling a large number of nuclear power plants sited near coastlines to generate electricity for desalinization would not themselves increase the demand for fresh water. Also, if fresh water ever become extremely scarce then salt water could be piped inland for use in cooling inland electric power plants.

If fresh water was made into a market then rising prices would provide incentives for much more efficient water usage. In the event of severe water shortages the most rational response would be to create water markets.

Update II: Rates of water usage vary greatly within the United States.

For instance, per capita use of public water is about 50 percent higher in the West than the East mostly due to the amount of landscape irrigation in the West (see map below). However, per capita use can also vary greatly within a single state. For example, in 1985 the demand for municipal water in Ancho, New Mexico, totaled 54 gallons per capita per day (gal/cap/day) while in Tyrone, New Mexico, municipal demand topped off at 423 gal/cap/day (Grisham and Fleming, 1989). Rural areas typically consume less water for domestic purposes than larger towns.

The areas that have high usage rates could cut way back to rates closer to what the lower usage areas have. Granted, there'd be a lot less lawn grass. But civilization would not teeter, let alone collapse.

Share |      Randall Parker, 2006 May 29 08:46 PM  Energy Policy

back40 said at May 29, 2006 11:12 PM:

"it would cost about $1.54 billion per day or $563.7 billion per year to get all water in the United States from desalinization. "

It seems there might be quite a lot of valuable coproducts. Sea water has many minerals such as uranium and so might provide the fuel for the power plants among many other things.

g510 said at May 30, 2006 2:23 AM:

In point of fact, wind can provide up to about 20% of grid capacity before intermittency becomes an issue with grid stability. For example 80% nuclear to 20% wind is a decent mix, and the percentages can vary somewhat if you factor in solar and other sources. (The idea that there's some kind of either/or competition between nuclear on one hand, and solar & wind on the other hand, is a false dichotomy. It's not an either/or, it's an and/both.)

However, let's not make the mistake of assuming that cheap energy and cheap desal will provide an excuse for reckless reproduction. Sooner or later a limit is reached in one resource or another, and at that point the choices are ZPG or dieoff. Launching the excess humans off to pioneer (or exile) on other planets is nonsense; the way I figure it, would require launching a space craft with capacity for 250 humans every 90 minutes, 24/7, indefinitely.

Jake said at May 30, 2006 10:06 AM:

Outlaw lawns as some new communities in Arizona have done and water usage drops by 40%.

Close the loop on sewage treatment and drinking water. This is technically possible but probably not politically possible. Again communities in Arizona are pumping purified water from sewage into the ground and then pumping it back out again for the water supply.

These two changes would make desalination unnecessary for the forseeable future.

hamerhokie said at May 30, 2006 12:05 PM:

Interesting issue regarding "affordable energy" was pointed out over on "The Cost of Energy" blog. Even though photovoltaics are going to get cheaper, they currently only make up about 50% of the cost of new installation. The other costs are for the inverters, data collection, metering, cabling, and labor, which are probably fixed costs (except for the inverters, which will probably drop as demand rises). There seems to be about $3-4 per watt of fixed costs built in. So even if, say, cheap plastic photovoltaics get down to $1/watt, we need to find a way to get the associated costs down as well.

Sean Lynch said at May 30, 2006 12:45 PM:

Step 1 (at least in California): end farm water subsidies. The amount of fresh water available in California will take the state well into the future assuming everyone pays the same market price. I don't know how well this carries over to the rest of the world, but in any case the first step before the government starts stealing more of people's money to apply to poorly managed "programs" should be to figure out other places where the government can *stop* stealing people's money in ways that create the problems they're trying to "solve" to begin with.

aa2 said at May 30, 2006 2:16 PM:

Great post Randall. A lot of doomers are saying water is the next resource that is about to run out. The reality is with nuclear power we can create fresh water to our hearts content.

It may be an issue in nations like third world nations or first world ones headed towards third world status.. That can't get their act together to build infrastructure.. But for an advanced nation it isn't really a big long term concern.

Having said that it will still be interesting to watch to see what paths are taken and what innovative technologies can make the process even cheaper.

Paul Dietz said at May 30, 2006 2:21 PM:

Note that 'withdrawals' includes things like water drawn to cool thermal powerplants, water that is returned to a lake or river. Only a small fraction of that is evaporated and actually consumed (although, as we noted in an early discussion thread, that amount can be a problem out west, where there isn't much water.)

Agriculture is the big user, and that could be made much more efficient if water were more expensive.

If one wants to desalinate brackish water (such as, spent irrigation water), then capacitive deionization might be preferable, although I don't know if it's proved itself on the market yet.

Robert Silvetz said at May 30, 2006 2:47 PM:

What about the old idea of evaporative transport? Long acqueduct with glass enclosure, the saline water evaporates and the fresh water condenses along the length of the acqueduct into parallel transport lanes.

Or better yet, most areas have great marine layer fog. Pump icy cold sea water from depth, drive it thru condensation towers, let water return to the depths where it was taken from. Once the water is moving, the pump up/restore cycle requires little energy. They wanted to do that in San Diego...

RueHaxo said at May 30, 2006 3:23 PM:

Amen Sean Lynch. American agriculture should be centered in the more water-rich Midwest and South. The government should stop subsidizing farming in the drier Southwest and Great Plains.
I hadn't heard of new Arizona communities prohibiting lawns, but Las Vegas was paying homeowners to tear up their yards and landscape it in a way more suitable to the desert. Perhaps something similiar could be done throughout the Southwest.

Nick said at May 31, 2006 8:37 AM:

"the first step before the government starts stealing more of people's money to apply to poorly managed "programs" should be to figure out other places where the government can *stop* stealing people's money in ways that create the problems they're trying to "solve" to begin with."

You're right, but you shouldn't blame government. What you're seeing is weak democracy, where a special interest gets a backdoor deal, not in the interest of most people (in effect, one group stealing from the rest). What you need is more democracy (meaning more public participation, more transparency in decision making, less secrecy, etc), and less purchasing of candidates. The best solution is probably publicly financed campaigns, which feels like "big government", but would actually get you better, cheaper government.

It's tempting to fantasize about a world without government, but actually, we used to have that. We had private individuals who raised private armies, conquered the other land owners and their private armies, and called themselves king. The history of the last 1 or 2,000 years has been the process of making things more democratic, with a shared, public, ruled by laws government.

Nick said at May 31, 2006 9:15 AM:

"So even if, say, cheap plastic photovoltaics get down to $1/watt, we need to find a way to get the associated costs down as well."

The answer to BOS costs is building integration and efficiencies of scale. Once conventional builders and roofers start including PV in their standard roofing packages, installation costs will mostly disappear. Once installation happens in large volumes, inverter costs will mostly go away: after all, the example system (at the "The Cost of Energy" blog) was just 1,500 watts - that's the power of a hair dryer!! With the proper volume to amortize the cost of the specialized DSP chips, an inverter that size should be $200 bucks.

gmoke said at May 31, 2006 2:36 PM:

Too bad none of you seem to have visited New Alchemy Institute back in the day or know the work on ecological waste treatment that John Todd has continued on with since then. You can go to to see what he's up to now but then there ain't no nuclear power plants involved or any "heroic" science for you Social Realist Engineers to drool over.

trollafrogg said at June 1, 2006 3:52 PM:

i would like to know how you intend to move all this desalinized water? i mean you must be looking at ocean water as your primary source, and presently their are limited systems, such a pumping and pipelines in place which would move this water all over the country. these missing pipe and pump systems would have to be built out and i should think that you need to add their cost and maintenance to your model as they will be significant costs. add to that pumps need drive, which means an increase of electricity cost and production to fulfill the dream.

Randall Parker said at June 1, 2006 4:32 PM:


The California aqueduct is 450 miles long.

Also, most people live near coasts and people are generally migrating to be near coasts. So most wouldn't need to have their water piped real far. Those who live far from large bodies of water and who live at higher altitudes will have to pay more for water. But even much of the Midwest of the United States is near the Great Lakes.

Randall Parker said at June 1, 2006 4:38 PM:

Oh, and get this: The aqueduct has to transport water up hill:

It moves water from Northern California to Southern California. It begins at the Sacramento River carries water south through the Central Valley where it often parallels Interstate 5 and then is pumped up 2,000 feet to cross the Tehachapi Mountains.

I drink and shower in this water btw.

Dezakin said at June 1, 2006 5:25 PM:

"Since salt water is usable for power plant cooling a large number of nuclear power plants sited near coastlines to generate electricity for desalinization would not themselves increase the demand for fresh water. Also, if fresh water ever become extremely scarce then salt water could be piped inland for use in cooling inland electric power plants."

I'm surprised cogeneration hasn't been mentioned. You generate the electricity for electricity, and the remaining waste heat goes for desalination.

You could even use the molten salt reactors discussed at:

"So even if, say, cheap plastic photovoltaics get down to $1/watt, we need to find a way to get the associated costs down as well."

If they get that cheap, an obvious application will be integrated into hybrid vehical hoods and roofs as battery chargers. No need for an inverter there.

Paul Dietz said at June 2, 2006 6:35 AM:

Much of those water withdrawals for thermal cooling could be eliminated with cooling towers. That increases water consumption, however, since there is evaporation. To eliminate that, one would go to higher temperature reactors and reject the waste heat into the air. High temperature gas cooled reactors could be appropriate.

Nick said at June 2, 2006 9:55 AM:

"If they get that cheap, an obvious application will be integrated into hybrid vehical hoods and roofs as battery chargers. No need for an inverter there."

Absolutely. This is already happening in Australia, where solar panels on RV's are now standard.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©