09 October 2009

Inconvenient Truth: Some "Green" Energy Can Guzzle Water

...that was the title that the Davis Enterprise used for Todd Woody's recent NY Times article exploring a recent surge in solar projects and a recognition of the need for sometimes significant amounts of water being consumed by large-scale commercial solar power plants. The title caught my eye. Being a numbers guy by nature, I was wondering what this article may be calling a "guzzler". The lead example for the article (and associated blog) were two planned solar farms in the Amargosa Valley (right next to Death Valley). Annual consumption would be 1.3 billion gallons of water per year. Sounds like a lot of water, doesn't it? I prefer to use cubic meters or acre-feet and compare it to equivalent farm-water use, which is where most of our water use is besides environmental flows - I think that would provide a much better perspective than the one-gallon water bottle we buy in the supermarket:


The 1.3 billion gallons are equivalent to 4,000 acre-feet per year. Compare that with USGS estimates of total groundwater use in the Amargosa Valley: as low as 14,000 acre-feet and as high as 21,000 acre-feet per year recently. Indeed a big chunk of that basin's water. From an Amargosa Valley farmer's point of view about enough to water alfalfa crops on 800 acres (1+ square mile) area (also see the company's construction plans, p. 25). Not much perhaps by California standards, but a big chunk in the Amargosa Desert.

From a water consumption point of view, the dry-cooling alternative is much more attractive than the above cited wet-cooling technology: the total water use would be 400 acre-feet per year instead of 4,000, the equivalent of an 80 acre farm. Turns out, the two solar farms themselves take up about 4,000 acre-feet of public land (not previously farmed), generating approximately 500 MW of electricity (about half that of a nuclear power plant).

Another example cited in the article is the BrightSource Energy Ivanpah project in Southern California: for a 440 MW dry-cooled plant, the water use is 80 acre-feet (25 million gallons). This is another 80% reduction in water use over the similarly sized Amargosa plants under their better dry-cooling option! To put it in perspective relative to farm water use: 80 acre-feet would be enough water to irrigate a 10-15 acre alfalfa farm or a 30 acre vineyard in Southern California. The real important point here is not the "25 million gallons of water" but that there is already technology that provides huge improvements in water use for these alternative "green" energy sources.

For a yet better perspective, let's convert these numbers to gallons of water withdrawal per kWh generated . I will rely on my own experience here. My home's 2.5kW solar panels are designed to generate about 3,500 kWh of electricity a year (this is probably a low estimate - after four months, I am already on course to generating near 4,000 kWh per year). The 440 MW BrightSource Energy solar farm should therefore produce about 700+ million kWh per year. The water use would be 1 gallon for every 28 kWh produced.

Compare this to water withdrawals for electricity production from coal power plants (which produce 52% of the U.S. electricity supply):  25 gallons for every 1 kWh produced (check Sandia National Lab's "The Energy-Water Nexus" primer). In the NY Times article,Todd Woody quotes UC Berkeley's Daniel M. Kammen: "When we start getting 20, 30, or 40 percent of our power from renewables, water will be a key issue".  I would counter: How could a 700-fold (!!!) decrease in water withdrawal per kWh make water a bigger issue than what it already is?

Lets put this in another perspective: according to "The Energy-Water Nexus" primer, the U.S. demand for new energy will be about 400,000 MW over the next 20 years. That is 1,000 new BrightSource Energy solar farm plants (80,000 acre-feet of water use). If we were to build these with power plants that use as much water as the BrightSource Energy solar farm and trade agricultural water in to meet the additional demand - that would be 30,000 acres of California farmland - or approximately two-thirds of the vineyard acreage in Napa Valley, California.

Sounds like a lot? Were we to meet this with coal power plants, and retired a water-equivalent amount of farmland, we would need to retire 21 million acres of irrigated farmland to save 56 million acre-feet of water - that is twice the total irrigated acreage in all of California or all of the irrigated lands in Arkansas, Texas, and California combined - representing one third of the total national irrigation water usage.

Despite the water use of solar power plants - with sound technology alternatives they seem far superior in their water footprint to standard power plants. Clearly, location and water source are central issues here, as the Amargosa and California water-energy discussions show. For a good discussion, see Mike Hightower and Suzanne Pierce's Nature article: "Some specific suggestions for reducing water usage by electric power generation include: using waste water, sea water or brackish groundwater for cooling and processing instead of freshwater; using cooling technologies that require less water or no water; switching to renewable energy technologies that do not need water for cooling — such as wind and solar electric; introducing technologies to condense evaporation from cooling towers and capture and reuse the water." And there is always my favorite: conservation. But let's be sure: to brand solar as a water guzzler ignores the real water use in much of our conventional power generation.

For a complete picture, we need to consider that not all water use in coal power plants is consumptive. Much of the water use in conventional power generation is simply for cooling while the water stays in the river. Here is a table courtesy of Mike Hightower from Sandia National Lab [from: Sandia National Lab Report SAND 2007-1349C):



Note that 1 gal for every 28 kWh in the case of the BrightSource solar plant is equivalent to 35 gal per MWh (the unit used in this table). A similar table can be found on p. 38 of a Report to Congress by this same group. Solar still comes out ahead! Check out Sandia National Lab's Energy-Water Nexus Website. And if you want to see, where solar is being build in the Southwest, check out SmartEnergyShow's solar map.

No comments:

Post a Comment