The Middle East and North Africa (MENA) region provides excellent conditions for the development of Concentrated Solar Power (CSP),[1] notably much irradiation and unused flat land[2] in close proximity to road networks and some transmission lines. Hence, a number of initiatives are underway to scale-up several donors are jointly launching a program to scale-up CSP in the MENA to several gigawatt (GW) over the next decade.[3]

CSP deployment on this scale[4] would bring substantial advantages to participating countries, including: leveraging investments into CSP plants, thereby almost tripling current global investments in this technology; providing massive investments in MENA countries; supporting MENA countries to achieve their development energy goals; and assisting Europe to meet its greenhouse-gas emissions reduction commitments.

However, CSP scale-up is not exempt from challenges, which comprise: the readiness of Europeans to purchase produced power; affordability of the produced electricity for MENA countries versus the decision to instead purchase less climate-friendly natural gas; the readiness of transmission infrastructure; and the availability of clean water for CSP requirements, along with environmental and social impacts. This article examines the latter.

Water Availability as a Central Challenge

Water use presents particular challenges in the MENA due to the region’s water scarcity: the region’s challenges from potential water scarcity are among the greatest in the world, with only 1,110 m3 of renewable water resources per person per year in 2007, far below the global average of 6,617 m3 p/c.[5] Environmental problems resulting from water issues cost MENA countries between 0.5–2.5% of GDP every year.[6] Future possible problems arising from water scarcity include food security issues[7] and possible conflicts over water.[8]

Comparative Water Use: CSP Versus Traditional Energy Technologies

The water needs of CSP (depending on design[9]) are similar in volume to “standard” energy technologies employed in the region (notably thermal power). Yet, one of the challenges to CSP roll-out in the region is that the technology’s water requirements (mainly for cooling) are nonetheless substantial.[10]

Data from CSP plants built in other parts of the world (e.g., the United States) so far suggest that parabolic trough CSP systems use ~700–800 gallons of water/MWh, compared to an average of ~500 gallons/MWh for coal and nuclear plants.

However, even though conventional power plants work at higher efficiencies (due to their ability to achieve higher temperatures and pressure), such comparisons may exaggerate their advantages in water use terms because:

the comparisons use figures from highly efficient US-based conventional plants, versus desert located CSP plants; and

if water use for scrubbing/ash handling in coal power plants is included in the calculations, overall their water use is similar.


Furthermore, conventional plants have very serious non-water environmental issues: local and global pollution (coal); scarcity (gas); or waste storage issues (nuclear). CSP arguably has less serious non-water environmental impacts, mostly from the risk of toxic fluid leakage — which hardly compare to the risks mentioned for coal or nuclear.

Water Impacts from CSP

The construction and operation of CSP projects lead to a variety of environmental and social impacts that need to be identified, assessed, monitored, and mitigated. This environmental due diligence is site-specific[11] and important at all stages of the project.[12] Table 1, below, shows the key possible CSP impacts related to water.

Estimating Water Needs of CSP in the MENA with Different Cooling Technologies

Generally, decision makers must choose between wet/evaporative cooling, air cooling, or hybrid cooling technologies for each prospective CSP project:[13]


Wet/evaporative cooling: efficient at moderate investment costs but high water consumption (approximately 574 gal/MWh or between 2–3 m3/MWh).

Air/dry cooling: less efficient[14] and more expensive than wet cooling but less than 10% of the water consumption compared to wet cooling (between <0.04–0.3 m3/MWh). Air cooling is restricted as an option if there are frequent overlaps of necessary highest plant output with highest ambient air temperature or “hot days” (above 37.7°C).[15]

Hybrid cooling: combination of wet and air cooling technologies. Air cooling with water during times of peak ambient temperatures. It is less efficient and more expensive than wet cooling but limits water consumption and is generally less expensive than air cooling. Water savings over the wet system can be 50% on average (ranging between 0–95%) since the system can be used in any combination depending on the design and actual usage of the cooling devices. Hybrid systems are especially useful in regions with many “hot days.”[16]

Initial data suggest that the water savings to be achieved with hybrid and dry cooling are substantial (up to 97% when switching from wet to dry cooling[17]), even though they carry performance and cost penalties (of 1–5% for hybrid cooling, and 5–9% for dry cooling). Performance and costs will likely be further impacted by the number of “hot days.” Thus, water availability, best technology choice, levelized electricity cost (LEC), and “hot days” issues all will require further study with data from the specific project areas.[18]

Based on cost, performance, and water use information, the author obtained data on the capacity and technology of each of the proposed CSP projects in the MENA and established the likely annual generation of each of these projects from average annual generation figures of CSP projects of that capacity level. Subsequently, two data sets were obtained (one from the Clinton Climate Initiative and one from the US Department of Energy) of general figures of estimated water needs for different CSP cooling options at various generation output levels. Project-level water needs were calculated for each project, which confirmed that water savings from air and hybrid cooling systems vis-à-vis wet cooling in CSP are considerable.

The calculations suggest that:


for air cooling and for the lower end range of estimates of hybrid cooling, water use is only 9–18% of the water used for wet cooling.

for hybrid cooling’s higher end range of the estimates, water use is 33–56% of the water used for wet cooling.


However, to make these calculations more accurate, it is important to know the number of “hot days.”[19] This data is currently unavailable (for one full year) but is expected to become available by the end of 2011.[20]

Meeting CSP Cooling and MENA Water Needs: Potential Double-Dividends from Desalination

Given the existing and growing water scarcity problems in the MENA, CSP scale-up should be done in a way that contributes to solving the problems by: serving as a stable electricity source for desalination through processes such as reverse osmosis (and thus partly meeting water needs of the region); and meeting the water needs of the CSP plants.

Since CSP plants are in essence thermal power plants, they can be used for combined heat and power. Thus, they can be coupled with desalination technologies that use both heat and water such as Multi-Stage Flash (MSF) and Multi-Effect Desalination (MED).[21]

An advantage of CSP-powered desalination is that a CSP plant can deliver more stable and constant power capacity than wind or photovoltaic due to its thermal energy storage capacity (the variability found in wind and solar may lead to system inefficiencies).

While a long implementation time frame should be expected (CSP desalination may need some 10–15 years from today in order to reach a significant share in the MENA water supply), CSP desalination has the highest potential to supply the most economic (<0.3/m3) and sustainable water to the region. Given the dearth of alternatives to secure the region’s water supply (e.g., with subsurface water withdrawal having reached over 1,000% of the safe limit in some countries), this option will become increasingly important.

Given the assumptions of water requirement of 2.8–3.4 m3/MWh and capital costs of $1,500–2,000/m3/d for MSF desalination and $900–1700/m3/d for MED, the additional capital costs required for the desalination equipment come to ~$33-37/kW. For the CSP scale-up plan of 1,000MW, this amounts to an additional $33 million, assuming that each plant reaches the economies of scale required for these costs. Since desalination is a mature technology in the region, these costs can be stated with a reasonable degree of accuracy. The additional solar field required would be in the order of 1–3%.

Initial figures researched on the level of water and electricity needed in a combined CSP and desalination scenario suggest that for a 1,000 MW initial configuration of the plan (assumed to run for an average of seven hours a day), the water requirements of the combined plants would be about 19,600 m3/day. Providing all of the needed water for the operation of such 1,000 MW CPS scale-up capacity through desalination would require 0.5–2.8% of the electricity output of the CSP plants, depending on the kind of technology utilized.


CSP scale-up may lead to a number of specific social and environmental issues, though few of these “CSP-specific” issues have so far arisen in actual CSP projects. For the most part, CSP projects are likely to give rise to the “standard” safeguards issues of an infrastructure project of the relevant size; however, special care needs to be taken to avoid replacement of agricultural water use or contamination of water bodies.

CSP scale-up throughout the MENA region may require vast amounts of water depending on the selected cooling option. While water availability for CSP is, at the moment, only a side issue in CSP development in the MENA (as CSP is still a technology with a limited geographical scale), a large-scale roll-out of the technology (e.g., as proposed under the DESERTEC initiative) would imply taking a close look at water availability/scarcity and should bring the issue to the forefront of CSP planning.

This article described that the water savings to be had from hybrid cooling systems and air cooling systems vis-à-vis wet cooling are considerable. This is of special importance, given the water scarcity in the MENA region. Additional water savings generated by air cooling over hybrid cooling (at least at the upper end of the estimates range) generally outweigh the rather marginal additional performance and cost penalty associated with air and hybrid cooling. Nonetheless, the decision on the cooling technology ought to be made in each instance taking into account local conditions.[22]

The article also discussed the possible impact on performance during “hot days” (air temperatures above 37.7°C degrees), and based on outside studies, indicated that as long as these “hot days” do not coincide with high plant output/peak demand and high revenues, the performance and costs impact may not be significant. However, to be able to perform a true assessment of the possible performance penalties associated with high temperatures, more data is required on hourly temperatures throughout the year and more information is also needed on the expected loads in the countries under consideration.

Combining CSP electricity generation with desalination technology can circumvent the problem of water availability since the desalination plant can strike a symbiotic relationship where it supplies the requisite cooling water to the CSP plants in return for electricity/heat needed to purify water for various purposes. This is of particular significance for the MENA, where desalination is already a mature technology and CSP has a very high power generation potential.


[1].This article represents the author’s own viewpoints and not that of any institutions with which she is or has been affiliated. The author would like to thank Georg Caspary and Nishesh Mehta for their valuable contributions. One of the most promising renewable energy technologies.

[2]. Both major types of CSP technologies discussed in this paper require approximately 4 ha/MW for collectors and heliostats (UN Environment Programme [UNEP], 2003).

[3]. Commitment investments for this development amount to nearly $5 billion at present.

[4]. Note that the European industry consortium, DESERTEC, intends to invest $400 billion in CSP and other renewable technologies in North Africa, making it perhaps the most ambitious climate change mitigation effort ever.

[5]. WRI Earthtrends Database, Water Resources and Fresh Water Ecosystems (2007).

[6]. World Bank, “Making the Most of Water Scarcity: Accountability for Better Water Management in the Middle East and North Africa” (2007).

[7]. Hang Yang and Alexander Zehnder, “Water Scarcity and Food Import: A Case Study for Southern Mediterranean Countries” (2002).

[8]. Conflicts over water have occurred in other countries, with a recent example leading to the death of 15 Somalians — a country where land and access to water conflicts are frequent (See Daily News Egypt, March 6–7, 2010). Furthermore, it has been recorded that of the 37 actual military water conflicts since 1950, 32 took place in the Middle East (30 of which involved Israel and its Arab neighbors in conflicts over the Jordan River and its tributaries, which supply millions of people with water for drinking, bathing, and farming. See “National Parting the Waters,” National Geographic (April 2010), See also Joyce Starr, “Water Wars,” Foreign Policy, No. 82 (Spring 1991), pp. 17–36.

[9]. The key types of CSP design considered in this report (and being used in MENA) are parabolic trough and power tower. Power tower systems are one of the three types of concentrating solar power (CSP) technologies in use today. Some power towers use water/steam as the heat-transfer fluid. Other advanced designs are experimenting with molten nitrate salt because of its superior heat-transfer and energy-storage capabilities. Power towers also offer good longer-term prospects because of the high solar-to-electrical conversion efficiency (see US Department of Energy [USDOE] website, 2010). Parabolic troughs are a type of linear concentrator and the most commercially available technology (e.g. they have been performing reliably at a commercial scale in the US for more than 15 years). In such a system, the receiver tube is positioned along the focal line of each parabola-shaped reflector. The tube is fixed to the mirror structure and the heated fluid — either a heat-transfer fluid or water/steam — flows through and out of the field of solar mirrors to where it is used to create steam (or, for the case of a water/steam receiver, it is sent directly to the turbine) and drives a generator to produce electricity (USDOE website, 2010). Note that another type of CSP is dish/engine systems, which use the Stirling thermodynamic cycle to directly produce electricity and therefore are air-cooled and only require water for mirror washing. These systems use sunlight to power a small engine at the focal point and the engines typically use hydrogen as the working fluid. These are not widely used yet and are currently designed to provide electricity only when the sun is shining (low possibilities for thermal storage) (USDOE, 2008). Since this is a disadvantage to utility scale production and when the peak load period lasts past sunset, this technology is not discussed in this paper.

[10]. A smaller portion is for cleaning of mirrors, while a certain amount of water is also needed for steam generation, if this is the chosen method of heat transfer.

[11]. The regional nature of this program may well necessitate a Strategic Impact Assessment for both environmental and social effects.

[12]. Three main stages may be defined as follows (based on UNEP, 2003): (1) environmental regulatory framework for the project; (2) environmental appraisal of the project; and (3) monitoring of environmental aspects during operation.

[13]. There is also a cooling option in which water is drawn from a body of water and then returned to that source. Termed once-through cooling, this option is not discussed here as it is now highly disregarded as an alternative due to its impacts on aquatic life. The losses from evaporation can also be significant and the option requires an average of 25,000 gal/MWh or 94.6 m3/MWh (USDOE 2008).

[14]. Water cooling has higher thermal efficiency and maintains consistent efficiency year round (USDOE, 2008).

[15]. This would be exacerbated if it coincided with highest electricity demand and revenues. Juergen Dersch and Christoph Richter, “Water-Saving Heat Rejection for Solar Thermal Power Plants,” Institute of Technical Thermodynamics (2007).

[16]. During hot days the performance of air-cooled systems has been shown to drop significantly. In such cases, the hybrid cooling system’s wet unit can reduce the load of the air unit and thus bring the condensing steam temperature closer to the design condenser temperature.

[17]. According to the German Institute for Thermodynamics.

[18]. Initial data from the German Institute for Thermodynamics give insight on the impact of local water costs on LEC in CSP projects in Spain and California, focusing on a comparison between wet and air cooling. The results indicate that wet cooling is only preferable to air cooling at very low water cost levels.

[19]. According to the US Department of Energy data, the performance of air cooled systems is reduced up to 5% for air-cooled trough systems during hot (above 37.7°C) summer days when compared to wet cooling.

[20]. As mentioned earlier, if high temperatures coincide with the highest electricity demand and possibly highest revenues, air cooling may become a less or the least attractive option. Data available: Egypt (Cairo Airport station — the Kureymat plant is 90km south): 22 hot days/yr on average; Morocco (Ouarzazate, project site): 50 hot days/yr on average; Algeria Hassir Mel: 6 hot days/yr on average. However, it has been indicated that providing mean air temperatures for the day is not enough and that hourly temperatures may be more necessary. It should also be noted that: 1) the performance drop is not instantaneous at a certain temperature value but it is continuous; 2) during “hot days” only few hours tend to be above 37.7°C; and 3) the actual extent of the performance decrease depends on the design of the heat rejection system. Furthermore, some of these “hot hours” are even in the late afternoon when the DNI is decreasing and thus the plant may run in part load, which means that the performance penalty from high ambient temperature might be reduced by the performance gain of the air-cooled system running in part load (Dersch 2010 Personal Communication); and that 2) increasing the solar field can allow for higher steam production to offset the higher backpressure during high ambient temperature periods (according to a study performed in California, USDOE 2008).

[21]. German Aerospace Center 2007, “Concentrating Solar Power for Seawater Desalination,”…. There are two broad categories of desalination technologies: thermal technologies and membrane technologies. Thermal technologies have been commonly used for seawater desalination in the MENA region. These involve heating the saline water and collecting condensed vapour to produce pure water.

[22]. E.g., water availability/cost, hourly ambient temperature, electricity costs, topography, population density, as well as the quality of water output needed.