This is the fifth post in my series Drought in California. In Part 1: California Climate and Drought I surveyed the current drought in California and climate projections for the future, finding that drought is projected to be the “new normal” climate in California. In Part 2: The Status of California’s Current Water Resources I found that California is already depleting both it’s groundwater and surface water resources. In Part 3: California’s Total Water Deficit I constructed an overall estimate of California’s future water deficit, concluding that it will be about 25.1 million acre-feet per year, about 39% of California’s current dedicated water supply. In Part 4: The Potential to Procure Additional Ground and Surface Water I found that a variety of obstacles and problems made it unlikely that California could cover the predicted future water deficit by tapping additional groundwater or surface water resources.
In this post, I will explore whether California could use desalination to cover some or all of the projected future water deficit. A report by the National Academies of Sciences concludes that desalination is likely to have a niche in the national water management portfolio of the future. The size of the niche, however, depends on a host of complicated and locally variable social, economic, environmental, and political factors. Thus, this post will be a bit lengthy and technical: my apologies in advance. I have noted with each post in this series that certain limitations and problems apply to the data and analyses used. They apply to this post, as well. You can read about them in the introduction to the series.
Three main problems hold the potential to limit California’s use of desalination: boron, the power infrastructure required, and the cost.
Boron is a naturally occurring chemical element that is present in ocean water. In normal doses, it is harmless. Desalination, however, concentrates it. Desalinated seawater typically exceeds EPA and WHO guidelines for exposure. Desalinators would need to reduce the concentration of boron to acceptable levels. In addition, little is known about long-term exposure to boron at low levels. Research to explore those effects is needed, but presumably it would not occur, and perhaps could not occur, until desalination plants were already constructed and the population exposed. (Committee on Advancing Desalination Technology, 2008)
Whether California would need to address the problem of boron, and how it might do so, is unknowable at this time.
The Amount of Power Required
Desalinating salt water uses a process called reverse osmosis. The salt water is pushed through a series of filters that are so fine that they filter out the salt molecules, but allow the water molecules to pass. It can be accomplished for as little as 3.4 – 4.5 kWh/cubic meter. (Committee on Advancing Desalination Technology, 2008) In Part 3 of this series, I projected that California’s future water deficit would be 25.1 million acre-feet. At 3.5 – 4.5 kWh/cubic meter, the total amount of energy needed would be 105-139 billion kWh per year.
(Click on graphics for larger view.)
California would need to construct the additional generating capacity to provide the power. I will look at three technologies for generating the necessary power: nuclear, natural gas, and solar.
.The rated output of the nuclear power stations currently operating in the United States is 502 – 3,937 MW. (Energy Information Administration, “How Much Electricity Does a Nuclear Power Station Generate?”) If they operated at full capacity 100% of the time, their annual output would be between 4.4 and 34.5 billion kWh per year. Thus, the energy required to operate the desalination plants could be obtained from 3 large nuclear generating plants, or more smaller ones.
It seems unlikely that new nuclear energy facilities would be constructed in California, however. The problems associated with them are notorious, ranging from storage of nuclear waste to interminable delays and cost overruns. But in addition to these universal problems, California is seismically unstable. Utilities in California have decommissioned nuclear plants because making them earthquake-proof would have been cost prohibitive. Further, the contemplated nuclear plants would consume 2.5 – 3.4 million acre-feet of water per year. (Jones, 2008) But this whole exercise is about the fact that there already is a water deficit: the needed water would not be available.
The most common type of generating plant in California is natural gas. The largest natural gas power plant in California is rated at 2,484 MW. (California Energy Commission. 2014) If it operated at full capacity 100% of the time, its annual output would be 21.8 billion kWh. Thus, to provide the needed energy would require 4 – 6 natural gas generating plants of this size.
Natural gas power plants don’t create nuclear waste, and they don’t require a great deal of water. They do, however, emit carbon dioxide, which has been identified as the primary cause of climate change. Burning natural gas to create the power for desalination would emit 44 – 59 million metric tons of CO2, an increase of 47 – 87% in California’s current emissions. (California Environmental Protection Agency, 2013) California has been decreasing CO2 emissions. The California Global Warming Solutions Act of 2006 requires a reduction to 1990 emission levels by 2020. A large increase in GHG emissions would be incompatible with this legal requirement, and it would be socially unacceptable given California’s commitment to reduce its GHG emissions.
Here let me comment that to construct coal-fired generating stations would create even more new emissions than would natural gas stations, roughly twice as much. For this reason, it will be assumed to be completely objectionable, and will not be considered further.
The amount of power produced by a solar farm varies according to weather and location. Topaz, one of the largest solar PV stations in the world, occupies 9.5 sq. mi. about 50 miles west of Bakersfield CA, and provided 1.05 billion kWh of electricity in 2014. (Wikipedia, “Topaz Solar Farm”) Solar panels deteriorate over time. The exact rate is uncertain, as solar power is a relatively new technology, and the quality of solar panels has improved over time. (Jordan and Kurtz, 2012) If the modules were to lose 20% of their capacity over 25 years, then the initial capacity of the solar farms would have to be increased by 20% to allow for the loss. Thus, the initial solar power required would actually be 126 – 167 trillion kWh per year, which would require a solar farm 1,139 – 1,507 sq. mi. in size, more than twice the size of the City of Los Angeles.
In addition, solar farms only generate power during the day, while desalination plants run 24 hours a day. Half of the power generated by the solar farm during the day (or more) would have to be stored for use at night.
A few types of grid scale storage have been proposed. Batteries may eventually become the most practical grid scale storage technology, but currently available systems are only able to provide power for up to 2 – 4 hours, and they degrade when put through long daily cycles of full-charge-to-deep-discharge, which is exactly what would be needed to power desalination plants. (Akhil et al, 2013)
Pumped hydro is the only proven technology available, and it is the most utilized grid-scale storage technology. Two reservoirs are constructed, one at a significantly higher altitude than the other. Combination generator–pumps are built between them. To generate electricity, water flows from the higher reservoir to the lower one, turning the generator–pumps to make electricity. To store electricity, it is used to power the generator–pumps to pump water from the lower reservoir to the higher one.
The largest pumped hydro station in the world is the Bath County Pumped Storage Station in Virginia. Its construction took 8 years and cost $1.6 billion ($6.2 billion in 2015 dollars). Its rated generating capacity is 3,003 MW. (Wikipedia, Bath County Pumped Storage Station) Operating at full capacity every night of the year, it would generate 19 – 25% of the power that would be needed. Thus, 4 – 5 pumped hydro generating plants of this size would be required to power the desalination plants.
California contains lots of high mountainous terrain, seeming to offer mountain valleys between which sufficient head could be obtained to build pumped hydro stations equivalent to the Bath County Station. Many prime reservoir location are already occupied, however. The California Department of Water Resources lists over 1,200 named dams within its jurisdiction, and 200 named reservoirs. (California Department of Water Resources 2015c, California Department of Water Resources 2015d). In addition, most of the mountainous territory lies within national parks or national forests. In the past, that has not prevented the construction of dams and reservoirs – many existing ones lie wholly or partially inside federal boundaries. However, national sentiment has changed somewhat, and it would introduce an unpredictable political variable. Because pumped hydro power stations require both an upper and a lower reservoir, 8-10 new reservoirs would be required, together with the dams to impound the water, and the pumping/generating equipment. Proposals to build several additional reservoirs have circulated in California for years, but have not been pursued. The Sites Reservoir, for instance, proposed in the 1980s, has not been pursued (Wikipedia, “Sites Reservoir”). It is not known whether 4-5 additional pumped hydro generating stations would be built, and even if they were, whether they could be built quickly enough to prevent widespread dislocations from water shortages.
Another limiting factor in using solar energy to power the desalination plants would be the availability of solar panels and their limited life expectancy. In 2013, about 38,352 MW (peak output) of new solar energy was installed worldwide. (Wikipedia, Growth of Photovoltaics) However, solar panels operate at less than peak output most of the time. To provide the 105-139 billion MWh needed to power the desalination plants would require an installed capacity of 58,167-76,986 peak MW. (National Renewable Energy Laboratory, PVWatts Calculator) That is 1.5-2.0 times the total world installed capacity in 2013. Now, PV manufacturing capacity is growing all the time, and the envisioned project for California would unfold over multiple years. Still, California would need to access a substantial fraction of the world’s PV production.
The need to access a huge supply of PV panels would be ongoing, not a one-time project, because PV panels degrade over time. The exact rate is a matter of controversy, as solar power is a relatively new technology, and the quality of solar panels has improved over time. (Jordan and Kurtz, 2012) A reasonable assumption, however, seems to be that after 20-25 years, the solar farms would either need to be enlarged to replace the lost capacity, or begin a program of module replacement. Were the solar farms to replace their solar panels, each replacement cycle would require a supply equal to the originally installed supply.
In summary, nuclear power has been judged to be too great a seismic risk in California. It also creates nuclear waste, and it consumes large a large quantity of water, which is not available. Natural gas would be seismically more safe. It would emit millions of metric tons of new carbon dioxide, however. This would be contrary to existing California law, and it would be highly problematic for climate change. It seems potentially possible to build solar farms to power the desalination plants. It would require the construction of solar farms on an almost unimaginable scale, more than 100 times larger than any constructed to date. Simultaneously, it would require the construction of a power storage infrastructure. A significant portion of the world’s production of solar panels would be consumed. Appropriate locations for pumped hydro storage would have to be secured, and the necessary permitting and planning issues resolved. Overall, it would be an immensely huge, complex infrastructure construction project, but not impossible on the face of it.
Because the construction of nuclear and gas-fired power stations were found to be problematic, they will not be considered further. Only solar power will be considered in estimating the costs of desalination.
It may be physically possible to desalinate sufficient water to cover the projected deficit, but are there economic barriers that make it impossible?
Available sources emphasize that various problems limit the ability to derive a reliable estimate of the cost of desalination. To date, most discussions of the cost of desalination have occurred in the context of other available water sources. Desalination may have been judged unfeasible because other water sources were less expensive. That may no longer be the case in California; the lack of other alternatives may make feasible a technology that would otherwise be thought too expensive. Worse, many of the estimates have been net of included subsidies that lowered the cost of the water, they have varied widely in which costs were included and which were not, and they have varied in assumptions about the operating life of the plant and its components. Further, many of the costs were site-specific, and were not be generalizable. (Committee on Advancing Desalination Technology, 2008) For these reasons, it is difficult to construct apples-to-apples comparisons.
Assuming interest rates of 6%, a capital-payback period of 20 years, a $0.05/kWh energy cost, and a consumption of 4.5 kWh/cu. meter of water, the Committee on Advancing Desalination Technology (2008) estimated a total cost of $0.61 cents per 100,000 cubic meters of water. Electrical energy was the highest cost, representing 38% of the total, and annualized capital costs were second highest, at 25%. The levelized cost of solar power, however, is not $0.05 per kWh as assumed in the model, but $0.08 per kWh. (Energy Information Administration, 2015) Adjusting for this difference, the estimated cost of desalinated water becomes $0.75 per 100,000 cubic meters. The estimated water shortfall was 25.1 million acre-feet. Converting that amount to cubic meters, the total cost of the desalinated water would be $23.2 billion per year.
The above estimate uses a levelized cost of solar power, which includes the capital costs of constructing the solar farm. However, the costs of constructing the reservoirs needed for power storage would not be included. The cost of a reservoir is site specific. However, we saw above that the cost of constructing the Bath County Pumped Hydro Storage Station was $6.2 billion in 2015 dollars, and California would need to construct 4 – 5 of them. I will calculate the estimated cost based on 5 reservoirs: even if 4 were sufficient, they would have to be hardened against earthquakes, and would thus be more expensive. This would be a one-time capital cost, not a yearly cost. There would be operating costs, but they are typically low. Thus, if one amortizes the cost at a 5% interest rate over the anticipated life of such a reservoir and power plant, which I will assume would be 20 years at minimum, then the annual cost would be something like $2.4 billion.
Thus, the combined total estimate of the annual cost of desalination would be $25.6 billion dollars.
How should one put that number in context? Below are some comparisons:
- The annual cost of desalination would be equal to 23% of the 2014-15 California state budget.
- The annual cost of desalination would be equal to 1% of the 2013 California GDP.
- The annual cost of desalination would increase the amount Californian’s spent on transportation and water infrastructure by 41%.
To cover California’s projected water deficit using desalination would involve a massive infrastructure construction project. Many potential obstacles, any of which could derail the project, would have to be overcome. The most contentious would likely be the construction of massive power generating facilities, with associated power storage, to power the desalination plants. Nuclear power and natural gas-fired power were thought to be too objectionable. However, the construction of solar farms to power the plants was thought to be within the realm of possibility, at least conceptually. Whether such a project could secure the needed funding, secure the needed land for reservoirs, secure the needed supply of solar panels, and secure the needed public and private will and consensus, is not known. The annual cost of desalination was projected to be about $25.6 billion, roughly equal to 1% of the current state GDP.
In the next post in this series, I will look at some options California may have to conserve water.
Akhil, Abbas, Georgianne Huff, Aileen Currier, Benjamin Kaun, Dan Rastler, Stella Chen, Andrew Cotter, Dale Bradshaw, and William Gauntlett. 2013. DOE/EPRI 2013 Electricity Storage Handbook in Collaboration with NRECA. Sandai Report SAND2013-5131, Sandia National Laboratories.
California Energy Commission. 2014. “Database of California Power Plants. Energy Almanac. http://www.energyalmanac.ca.gov/powerplants.
California Department of Water Resources. 2015c. Dams Within the Jurisdiction of the State of California. Web page accessed on 6/2/15 at http://www.water.ca.gov/damsafety/docs/Jurisdictional2014.pdf.
California Department of Water Resources. 2015d. Reservoir Information, Sorted by Dam Name. Via the California Data Exchange Center. Web page accessed on 6/2/15 at http://cdec.water.ca.gov/misc/resinfo.html.
California Department of Water Resources. 2015e. Dams Under the Jurisdiction of the State of California (map). Web page accessed 6/4/15 at http://www.water.ca.gov/damsafety/images/Dams2.pdf.
California Environmental Protection Agency. 2013. Greenhouse Gas Inventory Data – 2000 to 2012. Accessed online at http://www.arb.ca.gov/cc/inventory/data/data.htm.
Committee on Advancing Desalination Technology. 2008. Desalination: A National Perspective. Washington, DC. National Academies Press. Accessed online at http://www.nap.edu/catalog.php?record_id=12184.
Congressional Budget Office. 2010. Public Spending on Transportation and Water Infrastructure. Washington, DC: Congress of the United States. Accessed online at https://www.cbo.gov/sites/default/files/11-17-10-Infrastructure.pdf.
Energy Information Administration. “How Much Electricity Does a Nuclear Power Station Generate?” Frequently Asked Questions. Web page accessed 8/10/2014 at http://www.eia.gov/tools/faqs/faq.cfm?id=104&t=3.
Energy Information Administration. 2015. “Levelized Cost and Levelized Avoided Cost of New Generation Resources in the Annual Energy Outlook 2014.” Annual Energy Outlook 2015. EIA. http://www.eia.gov/forecasts/aeo/electricity_generation.cfm.
Kershner, Isabel. (2015). “Aided by the Sea, Israel Overcomes and Old Foe: Drought.” New York Times 5/29/15. Accessed online 5/29/15 at http://www.nytimes.com/2015/05/30/world/middleeast/water-revolution-in-israel-overcomes-any-threat-of-drought.html?ref=earth&_r=0.
Jones, Willie. 4/1/2008. “How Much Water Does It Take to Make Electricity?” IEEE Spectrum. Accessed online at http://spectrum.ieee.org/energy/environment/how-much-water-does-it-take-to-make-electricity.
Jordan, Dirk, and Sarah Kurtz. 2012. Photovoltaic Degradation Rates – An Analytical Review. Accessed online 6/3/15 at http://www.nrel.gov/docs/fy12osti/51664.pdf.
National Renewable Energy Laboratory. Undated. PVWatts Calculator. Accesssed online 6/3/15 at http://pvwatts.nrel.gov/pvwatts.php.
San Diego County Water Authority. 2015. Seawater Desalination: The Carlsbad Desalination Project. http://www.sdcwa.org/sites/default/files/desal-carlsbad-fs-single.pdf.
The Topaz Solar Generating Station as Seen from Space. NASA Earth Observatory. http://earthobservatory.nasa.gov/IOTD/view.php?id=85403&src=eoa-iotd.
Wikipedia. “Bath County Pumped Storage Station.” Web page accessed 6/2/15 at http://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Station.
Wikipedia. “Growth of Photovoltaics.” Web page accessed 6/2/15 at http://en.wikipedia.org/wiki/Growth_of_photovoltaics.
Wikipedia, “Sites Reservoir.” Accessed online 6/3/15 at http://en.wikipedia.org/wiki/Sites_Reservoir.
Wikipedia. “Topaz Solar Farm.” Accessed online 6/3/15 at http://en.wikipedia.org/wiki/Topaz_Solar_Farm.