Water: Green Energy’s Ace, the Achilles Heel of Conventional Power

October 30, 2013

I’ve discussed water as the limiting factor on conventional energy before.  This is one more reason why it is wrong to suppose the next 30 years of energy production will be the same as last – and one more way that climate change is sneaking up to bite fossil fools on the ass.

New study out of Argonne, profiles Texas as a microcosm – but the situation is global.

Christian Science Monitor:

..up to now, studies that have tried to glean the most economically viable mix of technologies for generating electricity have focused on the mix needed to meet some sort of cap on CO2 emissions.

“But that never considered the water,” he says, even as other studies looked at the water consumption that coal, nuclear, or gas-fired plants require for cooling and for steam to spin the turbines running the generators.

Researchers call this confluence of water and energy production in a world facing an ever-warming climate the water-energy nexus.

The issue made headlines in August 2008, when the Tennessee Valley Authority had to temporarily shut down three reactors at its Browns Ferry nuclear power plant in Athens, Ala., after a drought reduced water levels in the Tennessee River and a heat wave boosted the water temperatures. In principle, the plant still could have used the water for cooling its reactors, but the temperature of the effluent pumped back into the river would have exceeded limits set to protect aquatic life. Two years later, low water levels forced the utility to throttle back the reactors there to 50 percent capacity.

Similar concerns over water temperature and availability have affected nuclear plants from Kansas and the Connecticut coast to Europe.

Nuclear plants are not the only types of generating facilities affected.

In Texas, the state’s power plants should be able to tap existing surface-water supplies  through 2030, according to a study published in January that looked at the impact of weather variability on the state’s electric utilities and their future access to water.

But population growth and the need for more power plants are expected to force utilities to slake their thirst from other sources. These range from aquifers containing drinkable or brackish groundwater to some limited additional consumption of water currently being used to irrigate “low value” crops, the report suggests.

All of these are likely to be more expensive than currently available surface water supplies, according to the report prepared by energy and water specialists at Argonne National Laboratoryin Argonne, Ill.; Sandia National Laboratory in Albuquerque, N.M.; and the University of Texas at Austin.

Argonne National Lab:

This report summarizes a study to determine the medium‐term (through the year 2030) impacts of future climate and drought scenarios on electricity generation by the Electric Reliability Council of Texas (ERCOT). Because water in reservoirs is used to cool many steam cycle‐based power plants, significantly low water levels can reduce the ability to cool power plants. This reduced cooling ability can come from physical supply limitations or environmental constraints (power plant effluent temperatures exceeding permitted limits).

Water Availability

Water is projected to be available for ERCOT thermoelectric power plant operations until 2030. However, water for new development will likely need to come from sources other than unappropriated surface water. This conclusion largely means that future water supplies for thermoelectric power will be more expensive than historical supplies. Specifics are as follows:

  •   In general, very little unappropriated surface water is available for any use, including thermoelectric power.
  •   Water availability from appropriated surface water supplies, assumed as “low‐value” agriculture, is limited. This appropriated water is present in quantities > 5,000 ac‐ft/yr in only a few HUC8 basins.
  •   Several HUC8 basins have wastewater, potable groundwater, and brackish groundwater availability at greater than 10,000 ac‐ft/yr (enough for a large power plant).
  •   A number of basins (14) with severely limited water supplies are targeted for siting of new electric power production.


David Lochbaum, a former engineer at Browns Ferry who now works with the Union of Concerned Scientists, said last week that nuclear power plants are about 33 percent efficient, so two-thirds of the waste heat they generate has to be cooled using water. Lochbaum said increasingly hot water in the Tennessee River places another potential burden on Browns Ferry operations.

Lochbaum was part of a conference call last week with area environmental advocates who said electricity generation poses an ongoing strain on water resources, especially in times of drought or the reduced rainfall levels being experienced in North Alabama this year.

A study by the River Network, released last week, found that it takes on average 40,000 gallons of fresh water to produce a megawatt of electricity. The water is used, polluted or consumed in making the electricity, said Wendy Wilson, national director of the River Network’s energy and climate programs.

Wilson said a megawatt of electricity is generally what it takes to power a household for a month.

Browns Ferry has the capacity to generate 3,300 megawatts in the summer, according to TVA.

While the River Network and the Southern Alliance for Clean Energy both called for less water-intensive electricity production, such as wind and solar, the current system is stable, if somewhat challenged.


25 Responses to “Water: Green Energy’s Ace, the Achilles Heel of Conventional Power”

  1. jimbills Says:

    Water issues will go critical in the next 50-60 years. We’re using entirely too much water on not just energy, but agriculture, too:

    On energy, one of the hidden facts behind fracking is that we’re substituting one resource for another:

    The technological response, of course, is massive desalination plants. But these require energy, which require water. We’ll work harder and harder to stay in the same place, until we can’t even do that.

    • Desalination plants start with saltwater.  Are you seriously proposing that the OCEANS will run out of water to cool nuclear plants?

      • jimbills Says:

        Are you suggesting that we can only build nuclear plants near the ocean, or should we add building and powering massive seawater pipelines onto the list of complete infrastructure rebuilding projects?

        I rather think we’ll likely build desalination plants regardless of the cost – but it will eventually reach EROEI limits regardless of the resource swap. On this particular post, I’m really addressing fossil fuel energy as opposed to nuclear, though:

        • Are you suggesting that we can only build nuclear plants near the ocean

          If desalination plants included.  There is no point building one anywhere you lack essential feed streams.

          There are already nuclear plants using wastewater for cooling; this is another way of addressing the water issue.  I would like to see plants performing supercritical water gasification of organic effluents to fully clean, sterilize and remove the BOD from aqueous waste streams, but that requires temperatures upwards of 700°C and current water-cooled reactors operate around 300°.  At a high enough temperature, open-cycle air turbines are feasible and cooling water can be eliminated.

  2. Here are some tar sands facts:
    The water requirements for oil sands projects range from 2.5 to 4.0 barrels of water for each barrel of bitumen produced.
    It takes about 28 cubic metres (1000 cubic feet) of natural gas to produce one barrel of bitumen from in situ projects and about 14 cubic metres (500 cubic feet) for integrated projects. Currently, the oil sands industry uses about 17 million cubic metres (0.6 billion cubic feet) per day of purchased gas, or about four percent of the Western Canada Sedimentary Basin production. By 2015, this increases to about 40 to 45 million cubic metres (1.4 to 1.6 billion cubic feet) per day, or nearly 10 percent, assuming gas production stay level at 467 million cubic metres (16.5 billion cubic feet) per day.
    Producing a barrel of oil from the oil sands produces three times more greenhouse gas emissions than a barrel of conventional oil.

    • jimbills Says:

      Right – the tar sands exchange water and natural gas for oil. We’re hitting the peak in oil, so we’re using other resources as virtual proxies for oil. They’re also building a nuclear plant specifically for the tar sands:

      Canada has a LOT of fresh water, but even they will feel the pinch eventually:

      Click to access Water%20Depletion.pdf

      • The LEADIR TRISO-fuelled, lead-cooled reactor concept has also been suggested for the tar sands as a carbon-free supply of steam for SAGD.

        LEADIR operates at atmospheric pressure, using a “thermal fuse” of a lead jacket around the reactor to provide passive temperature limiting.  The promoter suggested it for other purposes as well, such as district heating.  LEADIR would generate steam at around 300°C, which is far above the temperature suitable for space heat.  This leaves plenty of temperature/pressure drop to operate a turbine and generate electricity as well.  As a passively-safe reactor with nothing under pressure except the steam itself, it could be located in places like beneath parking garages.  An air chimney would provide backup cooling by thermosiphon.

      • Powering tar sands production with thermoelectric PP adds even more burden on the water system. Damage comes first from the fracking, second from the thermal PP.

  3. it takes on average 40,000 gallons of fresh water to produce a megawatt of electricity.

    Scientifically illiterate sentence.  Gallons is quantity, watts is rate.

    The water is used, polluted or consumed in making the electricity, said Wendy Wilson, national director of the River Network’s energy and climate programs.

    Wilson makes no distinction between consumption and use.  Once-through cooling consumes 1-2% of the water, “using” but returning the rest.

    The alternatives to nuclear are fossil fuels, and “renewables” which are inevitably backed by… fossil fuels.  Heating the climate makes some areas wetter, but dries many more through greater evaporation without compensating precipitation.  You may have less waste heat per kWh from a gas-fired combined-cycle plant, but the net effect on the planet is much greater than nuclear.

    Wilson said a megawatt of electricity is generally what it takes to power a household for a month.

    The typical household consumes about 1 kilowatt.  That’s average power.  ENERGY consumption is measured in units like cubic feet of gas, gallons of oil or propane, and kilowatt/megawatt/gigawatt HOURS of electricity.  Anyone who throws around terms like “megawatt” as energy is a scientific illiterate and should be laughed at derisively.

  4. http://www.ucsusa.org/clean_energy/our-energy-choices/energy-and-water-use/water-energy-electricity-overview.html#sources.
    Increased temperatures may result in more thermoelectric PP curtailment and decreased efficiency. Increased cooling output temperatures also increase evaporation, thus reducing the available water supply.
    For a look at thermoelectric water use impacts:

    Click to access 50900.pdf

  5. ” “renewables” which are inevitably backed by… fossil fuels.”

    Or, eventually, more renewables.

    • The backup has to be something with a stockpile of energy to draw from, not a flow (like PV or wind).  The list of options for this isn’t very long.

      • Not a flow like rainwater? Hydro? How about geothermal? Wave? No chance to overlap or oversize intermittent sources?

        • Hydro reservoirs are stockpiles, like stacks of firewood are stockpiles.  They’re replenished by flows but they are not flows; they are available on demand.

          Geothermal is a stockpile.  Sadly, most geothermal reservoirs are non-renewable on a scale of centuries.  Once those hot rocks are chilled, it’ll take a very long time for the milliwatts-per-square-meter of heat flow from the mantle to heat them up again.  Active volcanic zones are an exception.

          There’s been talk about fraccing hot deep rocks outside of conventional geothermal hot spring zones, but AFAIK none of these efforts has been remotely economic.  There’s also the issue of bringing up dissolved minerals, some of which are (ironically) radioactive (NORM).

          You can oversize and overlap, but the more you do that the less return you get on your capital investment, both financial and physical.  Steel and concrete have their own environmental impacts.

      • Forgot to mention, any chance of controlling demand to match supply? What about processes like evaporating water for salt, water pumping. Cannot they be driven by a flow?

        • Demand-side management is always a possibility, depending how important the deferred consumption is.  I think we’d both agree that neither the computers nor the air conditioning in a Georgia office building can be load-managed too much without major negative effects on the work done and even human health.  Japan has had thousands of cases of heat stroke due to A/C cutbacks.

          What gets me is that there are energy sources such as osmotic power which are not being exploited yet.  The amount of energy lost as fresh water mixes with saltwater is staggering.  Osmotic membranes and fibers must be too expensive or too hard to keep clean.

  6. andrewfez Says:

    ‘We could start building 7,000MW of wind and 7,000MW of solar projects on the existing federally owned, hydro transmission systems, which have all seen catastrophic reductions in their power output. The hydro systems we built in the 1960’s on the upper Missouri are only delivering about 15% of their original design power output annually’.

    Minute 54 of the video:

    • Right after that, he makes the same point I’ve been making:  hydro is best suited to backing up other sources (and providing peaking power).

      Ontario uses nuclear for baseload and mostly hydro for peaking.  Ontario’s electricity supply is almost completely de-carbonized.  This is not a coincidence.

      • andrewfez Says:


        Ontario publishes an hourly pie chart of what’s being used to satisfy the load. They’ve got a little wind in there also that sometimes gets up in the 1,500 MW region. But lots of wind farms are set to come online in 2014:


        I didn’t count ’em all up, but at least another 2,000MW nominal capacity.

        • Ontario publishes an hourly pie chart of what’s being used to satisfy the load.

          You’ll find an hourly tally in the sidebar at Canadian Energy Issues.  Right now, it’s showing 11,181 MW of nuclear on the Ontario grid; nuclear and hydro are producing about 90% of the total.  Wind is producing 310 MW, which is about 17% of the 1727 MW nameplate.

          Ontario has problems with wind already; it pays a feed-in tariff for wind, but has to export a lot of that power during off-peak hours.  That off-peak power sells for much less than the FIT, so a lot of Ontario’s RE is generated at a loss.  (I wonder if any of that power finds its way across the St. Clair river and west to Ludington?)

          lots of wind farms are set to come online in 2014

          Which will magnify the existing problems, as well as requiring a lot more gas-fired (and carbon-emitting) capacity to compensate for the periods when the farms are producing 17% of nameplate.

    • Thanks for the reference. This is the kind of thing we need to be studying. If nothing else, those transmission systems are idle absent the flow from the dams. They are available for direct transmission from other sources. There are many places with hydro and wind. If other sources are used to pump the water back into the dam, the transmission systems could be kept running at a higher level. This guy has his thinking cap on.
      Here are some more interesting combinations of wind and hydro.
      That is not the only way wind could be coupled to water.
      In California, a massive amount of electricity is used to pump fresh water from the Delta, over the Grapevine and into LA. It is a massive system of reservoirs, canals, and pumps. This is an example of a high electric demand process that couples naturally with wind. There is a great potential for wind coupled with processes like this that need it.
      There needs to be a lot more thought on the appropriate uses of energy by harnessing the sources to the load more effectively.

      • andrewfez Says:

        No problem –

        What’s interesting about the BPA graph is that when the wind gets going it can displace the thermal contribution credibly (whether that actually translates into gas/coal powering down or the operator just switching it off, I have no idea, but it’s a start). Then it occurs to me that they need a lot more renewables, as the wind is displacing the hydro with little net increase in overall renewable use to chase the load (I guess some of it is getting exported though, so it’s still possible to be offsetting something that’s eating carbon). It’s late and i just got off work, so i could just be tired and interpreting that wrong, but look at the peaks and troughs of the hydro first when the wind is peaking then when it’s absent. Wish they had a ‘renewable’ curve adding the two sources together to make it easier to see the net RE production.

        The Wyoming storage projects look awesome – hope they all get built. Pretty exciting times, watching all these new ideas actually start playing out.

        Yeah, every once in a while i get a flyer from the Los Angeles Dept of Water and Power about how conserving water helps conserve energy. I suppose they recover some of the pumping energy with the 5 hydro plants mentioned in the link, though wind driven electric pumps with the hydro plants downstream sounds like a win-win. Amory Lovins talks a lot about the idea of components of a system having more that one purpose to maximize efficiency. I’m sure he’d get a kick out of the duel water-delivery/power production aspect of such a system.

        • Andrew. Right. The BPA curves are hard yo read. They operate primarily as flood control and as an energy source for aluminum smelting and export. Recently, a spat has developed where BPA refused their contractual obligation to buy wind. So the wind numbers you see do not necessarily represent wind output. Also, currently turbines furl when the wind exceeds the cutoff speed. So power can be zero when wind is high. There is an opportunity there. A compromise could give some turbines designed for high wind and some for low with more on time. For a better look at wind output see ERCOT or Cal ISO.

  7. E Michael Says:

    One of the consequences of reflecting sunlight to cool the planet, is that the world will get darker. This seems like it will have a pretty significant impact on the life cycle of plants, and on the efficiency of solar power cells considering that we would need to reflect enough sunlight to drop global temperatures by several degrees.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: