Large scale desalination: is there enough energy to do it?

Very large scale desalination projects may be needed to meet the world’s future fresh water needs. There’s certainly enough seawater, but how much energy does it take?

Desalination plant, Perth, Australia.	(ABC)

In many areas, fresh water is getting used up faster than it’s being replenished, and demand for water is growing. We face a growing “water crisis” ^[1]. As we start running out of fresh water, there’s an obvious place to look for more. Just 2.5% of the Earth’s water is fresh water, the other 97.5% is saline ^[1]. Desalination could become a very large part of the world’s future water infrastructure. This raises several questions, but I’ll stick to just one of them here: How much energy is that going to take?

The Assumptions

How much desalination capacity will be installed world-wide? I don’t know, my crystal ball’s clouded up. What I’ll do is calculate two sets of numbers to try to bracket the possible options, a “low” scenario which may be a plausible outcome, and an extreme upper bound “maximum” scenario.

How many people will there be? The world population will most probably flatten out and peak at around 9.2 billion soon after 2050 ^[2]. I’ll take this as the maximum world population. How much water do we use? The highest per capita fresh water abstractions among OECD nations are those of the U.S., at 1730 m³ per capita per year, the lowest are Denmark’s at 130 m³ per capita per year ^[3]. I’ll use these two values to bracket the range.

How much energy does desalination require? The most efficient (reverse osmosis based) desalination plants consume about 5 kWh of energy per cubic metre of fresh water produced ^[4]. The fundamental thermodynamic limit for desalinating seawater is 0.86 kWh m^{−3  [4]}, so there’s plenty of room for improvement. I’ll use the 5 kWh m⁻³ figure for the “maximum” scenario, and assume a two-fold improvement for the “low” scenario. For the “maximum” scenario I’ll assume that all fresh water must come from desalination (i.e. that no other source of fresh water is available), for the “low” scenario I’ll adopt the more plausible assumption that 10% of the world’s fresh water will come from desalination.

The Energy Required for Large Scale Desalination

Table 1 summarises the two sets of assumptions, and the resulting calculations.
The “maximum” scenario takes the world population at its peak, assumes they all have U.S. levels of water consumption, and assumes that all that water comes from desalination. This is, of course, very unrealistic. It’s there to provide an upper bound to the set of possible outcomes. The “low” scenario goes for a more plausible possibility, with Denmark’s level of per capita water use, and 10% of the world’s water coming from desalination.

Table 1.  Energy requirements for global scale desalination

		Low scenario	Maximum scenario
Peak world population (circa 2050) [2]		9.2 bn	9.2 bn
Water consumption [3] (a)	(m3/capita/yr)	130	1730
Desalination energy [4] (b)	(kWh m−3)	2.5	5.0
Fraction of water from desalination		10%	100%
Power for desalination (globally)	(GW)	34	9100
Additional price of water per capita (c)	(UKP/capita/yr)	£3.25	£865
Mean electricity generation in 2005 (for comparison) was 2100 GW. Notes: (a)  The “low” value is water abstraction for Denmark; “maximum” value is water abstraction for U.S. (b)  The units are kilowatt hours of electricity per cubic metre of fresh water. (c)  Energy cost assumes an electricity price of £0.10 per kWh.

The “low” scenario uses 34 GW of electrical power worldwide, to provide 10% of the world’s water needs by desalination. For comparison, global electricity generation in 2005 averaged 2100 GW ^[5]. That is, the entire world population can have 10% of its water needs met by desalination with less than 2% of the world’s present day electrical energy. The “maximum” scenario requires over four times the world’s electrical power generation, but this is an extreme upper bound.

How much would that cost? For the “low” scenario and an electricity cost of £0.10 per kWh, the cost of the energy for desalinated water is £3.25 per person per year. For the “maximum” scenario, it’s £865. The cost of desalination is sometimes described as “prohibitively expensive”. The need for water is absolute; if there’s no cheaper alternative, the word “prohibitive” isn’t particularly accurate.

Conclusions

I’ll cover the caveats first. Large scale desalination would require extensive infrastructure changes, such as aqueducts from coastal areas leading inland, and so on. I haven’t considered any of that here, I’ve just focussed on a single fundamental question, the energy requirement of this arrangement.

The energy requirement is large, but well within the range of existing energy infrastructure. The price is high, but in the same ballpark as existing utility costs. There are no fundamental showstoppers to desalination on a massive scale. Very large scale desalination is a viable way to extend fresh water resources. There’s a phrase about “eating oil” to describe agricultural fertilizer use. We may end up “drinking energy”.

References

1. The Global Water Crisis, Encyclopedia of Desalination and Water Resources (DESWARE), UNESCO Encyclopedia of Life Support Systems (EOLSS)
2. World Population Prospects, The 2006 Revision Population Database, United Nations Population Division (2007)
3. Water Consumption, OECD Factbook (2005)  (WebCite cache)
4. Energy Requirements Of Desalination Processes, Encyclopedia of Desalination and Water Resources (DESWARE), UNESCO Encyclopedia of Life Support Systems (EOLSS)
5. Key World Energy Statistics, International Energy Agency (2007), (page 26)  (WebCite cache)