Renewable Energy Desalination for Island Communities: Status and Future Prospects in Greece

Abstract

Energy and water are two of the most important components required to ensure prosperity and sustainable development to societies. This paper aims to review the status of renewable energy desalination for Greek islandic communities, deployed in two axes. The first one reviews the desalination systems state of the art technological solutions, their energy needs, how renewable energy may be employed and finally the cost of renewable energy desalination is investigated. The second axis focuses on Greek islands per se, where the current situation is investigated, potential solutions for meeting the water needs are evaluated, all leading to the proposal of a methodology towards designing an appropriate and applicable approach in addressing the water needs. Finally, a discussion takes place on how such options might be further deployed, particularly regarding the impacts they may produce for the livelihood and the future prosperity of the pertinent communities, and at the same time supporting the energy transition towards the EU Green Deal goals.

1. Introduction

Energy and water are two of the most important components required to ensure prosperity and sustainable development to societies. The World Health Organization stated that 785 million people lack even basic drinking-water services and by 2025 half of the world’s population will be living in water-stressed areas or conditions [1]. At the same time, the predicted climate change may increase the frequency of droughts and floods as well as affect the seasonal variation of river flows [2]. In this context, the EU decided to adopt the European Green Deal, a set of policy initiatives aiming to make the EU countries climate neutral by 2050. The use of renewable energy along with energy efficiency are at the core of the Green Deal initiatives [3].

The present effort aims to review the status of renewable energy desalination for Greek islandic communities, having a two-prong emphasis. The first one reviews the state of the art of desalination systems as far as technological solutions are concerned, their energy needs, how renewable energy may be employed and finally the cost of renewable energy desalination. The second focuses on Greek islands per se, where the current situation is investigated, potential solutions for meeting the water needs are evaluated, leading to the proposal of a methodology towards designing an appropriate and applicable approach in addressing the domestic water needs. Finally, a discussion takes place on how such options might further be deployed, particularly regarding the impacts they may produce for the livelihood and the future prosperity of the pertinent communities, and at the same time supporting the energy transition towards the EU Green Deal goals.

2. Materials and Methods

2.1. State of the Art of Desalination Systems Powered by Renewables

2.1.1. Desalination Technologies

The salinity of water is usually characterized through its total dissolved solids (TDS), which is measured in mg/L or parts per million (ppm). Saline water resources are generally categorized as: brackish water, seawater, and brine. Desalination can be defined as “the process of removing dissolved salts from saline or brackish water thus making it suitable for human use and consumption i.e., domestic, agricultural, and industrial purposes” [4]. It is accepted as a different approach for supplying potable water, particularly in areas facing water scarcity challenges. Seawater and brackish groundwater desalination are commercially available and at the same time research is still strong in improving them. Desalination is already providing fresh water to communities, towns, cities, as well as private sector facilities (e.g., hotels, golf courses, etc.) globally. Furthermore, desalinated water can be utilized for agricultural land irrigation, in spite of its high cost [5,6]

The desalination technologies can be broken into three main process categories: thermal desalination (distillation), membrane desalination, and charge-based desalination (ion exchange processes). Thermal technologies are highly effective at large capacities (several thousand m³/day). Charge-based separation technologies (CST) are a good fit for brackish water sources as well as small scale systems (few hundred m³/day). Membrane desalination technologies are the most utilized commercial desalination technologies. It has been optimized for any range of water production capacity [7].

Thermal technologies are based on the distillation processes for desalinating water. Feed water is vaporized through heat provision, and fresh water evaporates as steam leaving brine. The vapor is condensed through cooling and low salinity potable water is collected. As far as cost is concerned, there is a better fit with higher salinity feed water sources [7]. The major thermal desalination technologies are Multi-Effect Distillation (MED), Multi-Stage Flash distillation (MSF), and Vapor Compression distillation (VCD). In most cases, MED, MSF and VCD are preferable in relation to cost only for capacities beyond a few thousand m³/day, and for feed water sources of increased salinity (≥35,000 ppm) [7]. Thermal desalination technologies have a high energy demand in the range of 23–25 kWh/m³ and are also utilized for large scale plants. Thermal technologies are not suitable for small scale desalination systems or systems in rural communities where decreased cost is sought or in cases of brackish water (3000–10,000 ppm).

CST technologies separate positively charged sodium (Na⁺) and negatively charged chloride (Cl⁻) ions (as well as other salt ions) from water utilizing electrostatic attraction and repulsion. This is achieved through the application of low voltage across two electrodes in the high salinity water. CST is suited to brackish water feeds and also most often presents high water recovery rates (85–90%), higher than Reverse Osmosis (RO) (25–80%) or standard thermal technologies (35%) [8]. Currently, the commercial CSTs are electrodialysis (ED), electrodialysis reversal (EDR) and membrane capacitive deionization (mCDI) [9]. Commercial mCDI and EDR systems are only targeting for very slow salinity water sources (<3500 ppm) and they are not currently suitable for sea water [10].

Membrane methods are based on the process of osmosis. Commercially offered membrane processes include RO, forward osmosis (FO) and nanofiltration (NF) [7]. RO is currently by far the most used membrane process. The other membrane processes have little commercial application [11]. RO is based on the application of pressures above the osmotic ones, resulting in the retention of the brine on the high-pressure side (65–75 bars) and the desalinated water passing through the membrane. RO can be utilized both for seawater (SWRO) and brackish water (BWRO) desalination. From the seawater supplied in a RO process about 40–60% is recovered as desalinated water. This percentage increases to 50–90% when lower salinity feed water sources are used. RO systems require special feed water pre-treatment not used in thermal desalination systems. Of the overall electricity consumption of an RO system, high-pressure pumps consume about 80% [12]. Brine from RO processes is ejected at a high pressure and thus is a prime candidate for energy recovery. Using energy recovery devices (ERDs) and high-efficiency pumps can lead to a total reduction in excess of 50%, e.g., a retrofit with high-efficiency pumps and ERD in the desalination system providing water in Corralejo on Fuerteventura in Spain’s Canary Islands managed a 57% energy consumption reduction [13]. Today, the specific energy consumption (SEC) for SWRO plants can be lower than 3.0 kWh/m³ and this is a major reason why SWRO has become more competitive than thermal desalination technologies [14]. A key design benefit of RO is its modularity. This allows the effective and efficient design of a desalination system able to meet desalinated water needs ranging from less than 1 m³/day up to industrial scale installations. RO fits well with small size systems and as such its global market share in this segment (capacities of the order of 1000 m³/day) exceeds 90% [15]. RO technology is being optimized and has been commercially available for decades, and it is reliable, relatively inexpensive and of low risk. Overall, RO is the most promising option today for islands and coastal communities.

2.1.2. Environmental Considerations for Desalination

Any alteration to the natural environment exerts an impact on it. Desalination is not an exception. Desalination processes include different activities occurring only during the construction phase (e.g., connection to existing electrical grids and water networks), as well as others that are active throughout the whole desalination system operational cycle (e.g., pre-treatment and post-treatment processes and brine disposal). Civil works during construction, interconnection with water and electricity grids as well as road network, erection of required buildings, feed water intake, brine treatment, transport and storage of desalinated water, pre-treatment and post-treatment chemicals, noise and vibration among others alter the local environment and may have adverse impacts [16]. Major environmental impacts may be: (i) adverse effects on soil, air and water as well as other natural resources, (ii) alterations in marine ecosystems and possible community relocation, (iii) chemicals utilization near populated areas, (iv) public health risk due to the nature of effluents, emissions or residues, (v) habitat modification [16].

Brine discharge may have a major environmental impact. Different practices for brine management have been proposed and tested, including minimization to obtain salt products [17]. However, high energy is needed to achieve a (near) zero liquid discharge (ZLD) [18] since energy-intensive thermal methods need to be employed. With minimum liquid discharge (MLD), the energy required for concentrating wastewater is still high, typically 20–25 kWh of electrical energy per m³ of feedwater treated [19]. Novel technologies potentially able to accomplish MLD and ZLD include osmotically assisted RO (OARO) and variations of OARO [20]. However, such technologies are not yet introduced into commercial practice mainly because of the high cost. Other brine management methods have been developed and implemented including release in waste water, evaporation ponds, mixing with surface water, discharging in deep wells, releasing on land [21] and for field production of natural gas hydrate [22]. In more detail [21]):

• Release in waste water: Essentially, brine is mixed in waste water. This can decrease the growth of bacteria in waste water treatment plants. The cost of this method is between 0.32–0.66 USD/m³ brine.
• Evaporation pond: In this method, salt is gathered from the ponds after the water has evaporated. This can result in groundwater pollution and soil salinization. The cost of this method is between 3.28–10.04 USD/m³ Brine.
• Mixing with surface water: In this case, brine is mixed with surface waters. The main challenges of this method is that it can lead to marine environment pollution. The cost of this method is between 0.05–0.30 USD/m³ Brine.
• Discharging in deep wells: In this method, brine is discharged into porous subsurface rock layers. The main challenges associated with this method include groundwater pollution and soil salinization. The cost of this method is between 0.54–2.65 USD/m³ Brine.
• Releasing on land: The brine is used to irrigate tolerant to high salinity plants. This leads to soil salinization. The cost of this method is between 0.74–1.95 USD/m³ Brine.

It is clear that as the complexity of the method increases, so does cost. Moreover, all methods present environmental challenges to an extent.

Membrane technologies (mainly SWRO) produce brine between 60–85 g/L TDS, whereas the salinity of the brine from MSF and MED technologies is between 55–65 g/L TDS. This is the result of higher recovery rates (40–45%) of RO systems. As for the salinity of the treated brine, it depends on the recovery of the processing methods utilized and can exceed 150 g/L TDS [23].

Due to the high salinity of brine (>1.6–2.1 times greater than that of seawater), the most extreme effect is on marine species [24]. However, it has to be noted that numerous species are able to acclimatize to higher than usual water temperature and salinity. In places such as the semi-closed seas (e.g., the Mediterranean), the change in salinity can be significant in the case of large amounts of brine (of the order of hundreds of thousands of cubic meters per day), while the risk is significantly reduced in the case of small plants of a few thousand cubic meters per day [25]. In places with strong currents in the ocean, negligible effects on the marine environment have been reported [26]. The amount of brine discharged from desalination plants must be of sufficient size to minimize the impact, comply with environmental regulations and prevent any environmental impact on water bodies (from rivers to oceans). Brine must be diffused and diluted as efficiently as possible [27] and significant progress has been made in this direction in the recent years to develop new diffusion systems and improve existing ones [28].

The sea or brackish water intakes may also cause environmental impacts. Intakes may be categorized into open sea intakes and subsurface intakes. Intake correct design, erection and operation affects the pre-treatment which in turn affects the overall desalination efficiency.

Effective measures should be applied to minimize the impact of a desalination unit on the local marine environment and aquatic life because the intake and discharge systems are in direct contact with it. Both intakes and brine discharging systems can be designed to minimize the above-mentioned impacts. This applies both to new and existing systems. Design approaches minimizing impact include [29]:

• Intake/outflow to be located in higher depths and areas that are active biologically.
• Deployment of structures preventing the organisms from reaching the intake and discharge systems.
• Deployment of bypassing structures allowing blocked organisms to leave the intake/discharge systems.
• Reduce the flowrate of the feed water intake to 0.15 m/s which helps fish not to get trapped.
• Utilize velocity caps and other devices (e.g., ones that generate light and sound) to prevent organisms from approaching.
• Design the mesh size of the screen effectively to decrease entrainment, impingement, and entrapment.
• Install the inlet/outlet in an active zone with currents and waves.
• Proper design and installation of inlet/outlet wells.
• Reduction of the intake size.
• Use materials with superior corrosion/erosion performance.
• Effective maintenance program.

2.1.3. Powering Desalination

Small-Scale Desalination systems (SSD), combined with appropriate Renewable Energy Sources (RES) technologies, offer a realistic solution for areas with water deficits and also facing electricity provision problems (no grid connection or a weak grid). Furthermore, when a power grid is available, the kWh cost of several renewable energy (RE) technologies competes today with the conventional kWh cost [30]. A plethora of countries all over the world (including Greece) have the potential to utilize renewable energy due to the available high potential of renewable energy resources many times as well as matching power production profile with water demand profile (e.g., high solar radiation during the hot summer period when water demand is high).

The three major costs affecting the final desalinated water production cost is the cost of energy, operational and maintenance cost (O&M) and capital expenditure [31]. The most common technologies in large-scale desalination used to be thermal desalination, however, today the dominant process is RO [32]. Energy costs account for 40–60% of the total desalination cost [31,33] and are the major cost determinants along with capital costs [34].

Table 1. RO desalination operating expenditure by service.

Cost Component	Data from [35]	Data from [11]
Electricity	46.6%	41%
Chemicals	16.8%	11%
Labour	24.4%	13%
Replacements	12.2%
Maintenance		14%
Membranes		5%
Other (Water discharge, Monitoring, Indirect costs)		17%

shows operating costs as divided into main services, i.e., energy, labour, replacements, chemicals, maintenance, etc. [11,35], where it is shown that the higher percentage of the operating expenditures accounts for energy consumption. This highlights the importance of addressing the power provision to the systems not only from an environmental/climate change point of view, but also due to the high share of energy in the overall cost of water.

2.1.4. Renewable Energy and Desalination

Energy produced utilizing fossil fuels causes significant Green House Gas (GHG) emissions. Thermal desalination technologies (e.g., MSF and MED) emit 10× more GHG in comparison to RO. In the 1970s, RO specific energy consumption was ~16 kWh/m³. Currently, this figure has gone down to 2 kWh/m³ [36]. This substantial decrease in the consumed energy by the RO process is mainly the result of the developed ERDs, and also because of advances in membrane materials [14]. Figure 1 presents the consumed energy per desalination technology [36]. With the increase of feed water salinity an increase of the specific energy consumption is observed.

Energy consumption mitigation measures include:

• Use of renewable energy technologies as well as utilization waste heat. Currently, PV use is the most popular application.
• Co-generation can increase the desalination plant’s efficiency so long as electricity and thermal energy are simultaneously utilized [37]. Nevertheless, such co-generation schemes can only be applied to large scale desalination plants of several tens of thousands m³/day and they are not suitable to small scale plants of a few thousand of m³/day.

Conventional energy sources have various shortcomings, especially negative environmental impacts, and this justifies the current trend for using renewable energy technologies and feeding this renewable electricity to desalination plants. The lower energy requirement membrane process powered by RES in comparison to desalination based on thermal processes has made membrane processes more dominant and widespread. This is also the main reason that thermal technologies coupled with renewables have not gained a significant market share since the final cost of water is higher and they are mostly used for systems with production capacities below 1 m³/day [38]. From the technical perspective, solar thermal technologies are commercially available in multiple types [39] and can meet the power requirements of thermal desalination systems [40]. Choosing the most appropriate RES for a desalination method depends on many variables including water demand profile, feed pressure, salinity, area of installation as well as the target water production cost. There are also constraints associated with the use of RES based on their inherent characteristics (e.g., low energy density and intermittency).

These challenges can be addressed to an extent or even eliminated through the use of proper grid integration, hybridizing of multiple RES technologies and the use of energy storage systems such as batteries. Photovoltaics (PVs) with RO is currently the combination with highest technology readiness level and impact [41]. The apparent reason for the domination of RO powered by electricity producing technologies (photovoltaics and wind) is the fact that RO needs only electricity to operate, which makes the combination easier and simpler than other combinations.

There are several technical options available for combining renewables and RO desalination as briefly discussed in the following.

• Autonomous topology (no connection to the grid). The RO system is powered exclusively from the renewable energy system either (a) directly (without battery storage—for very small systems) or (b) through batteries.
• Semi-autonomous topology. In this mode the RO system may be powered from the renewables system (equipped with a battery storage) and any surplus of PV energy is supplied to the grid (or shredded if the grid cannot absorb it and the batteries are full). Any additional power required by the RO plant may be taken from the grid.
• Independent topology, where the renewable energy system is grid connected. In this mode, the power from the PV system is fed to the grid and the required power from the RO plant is drawn from the grid.

The flow diagram presented in Figure 2 outlines the options that the desalination plant developerneeds to take into consideration before deciding on which approach to follow.

Grid connection is the preferred option when it is either directly available on the site of the desalination plant or a grid extension is a viable investment. This can ensure 24 h per day operation of the plant, maximizing its output. Depending on the location and the size of the desalination plant, it has to be ensured that the distribution grid can meet the power demand profile.

If the grid can meet the demand, then the developer could decide to simply purchase electricity from the grid or, depending on the regulatory framework in each country, the desalination plant operator may choose to invest in renewable energy technologies. Such investment may further decrease the plant operational cost and at the same time minimize its environmental footprint regarding energy consumption. In this case, all renewable energy technologies may be considered, but, in reality, photovoltaics are mostly applied due to the ease of deployment, along with wind turbines if the aeolian potential is appropriate. Depending again on the regulatory framework and available financing, energy storage in the form of batteries may also be considered. Investing in hydro power or geothermal power is technically possible, but the more complicated and time-consuming licensing procedures usually deter investors.

There are many cases in the Mediterranean, usually addressed in small islands, where the installed power of the grid is unable to cover the seasonal demand of tourism, a situation which becomes even more challenging when desalination plants are introduced. In those cases, it is up to the Distribution System Operator (DSO) to decide whether it is worth it to upgrade the grid/generation to meet this demand, put a cap on the power it may provide or state the inability to power the desalination plant. The first case is straight forward, while the two other cases are the most challenging.

The challenge stems from the fact that the design process has to change fundamentally if cost-effectiveness is sought. Traditionally, based on the given water needs, a desalination plant is designed aimed at operating constantly and if renewables/autonomy are considered, then a system is designed to meet the specific needs of that desalination plant. Due to the rapid decline of the cost of photovoltaics in the past decade, it makes sense to oversize the desalination plant and operate it for less hours in the day, using the PV power as is produced without storing it for night time operation. This demands that a techno-economic optimization takes place, which sizes at the same time the desalination plant and the power system. Moreover, advanced control schemes may further enhance the use of power leading to lower water cost. The same approach has to be followed in the cases where the DSO can provide limited power to the plant. It has to be mentioned that commercial autonomous microgrid solutions can go up to medium voltage and the MW range.

Coupling of RES with desalination can provide fresh water in a sustainable manner in areas having access to brackish water or seawater around the world. Designing a RES-desalination system needs to take into account elements relating to [42]:

• Technological availability and suitability.
• Environmental/Land use.
• Economic issues.
• Location specific characteristics.
• Energy–RES availability and quality.

2.1.5. Cost of Desalinated Water

Large Scale Plants

For large scale plants, the final cost of water is affected by many interrelated factors such as [11]:

• choice of technology.
• cost of energy.
• plant size and configuration.
• feed water and product water quality.
• environmental compliance requirements.

These factors are heavily site-specific. Τhermal desalination technologies in general and MSF-based plants in particular are considerably more capital intensive in comparison to SWRO. For thermal plants, construction and equipment are the major capital expenditures. For SWRO the capital costs are not heavily correlated with the actual design option. The recurrent costs for SWRO plants per unit of fresh water produced are two times higher than those of MSF plants, and three times higher than those of MED plants [11].

Capital costs for MSF plants per each million litres per day (MLD) of capacity are higher than those of MED plants. The energy cost for thermal plants is by a large degree the largest operation expenditure (OPEX). Energy costs account for two-thirds to three-quarters of all OPEX for thermal and between one-third and nearly one-half for SWRO. The total cost in absolute value is always lower for SWRO. The most recent large-scale SWRO plants exhibit water production costs in the range of USD 0.72/m³ to 1.20/m³. In a typical case of a modern SWRO plant, the two major cost items are energy (30% of total per cubic meter cost) and capital recovery costs (44% of total cost). The remaining 26% of costs are comprised by various variable costs including chemicals, membranes and filters, and brine disposal (12% of total cost); and labour, maintenance, and monitoring costs and other operation and maintenance costs (14% of total cost) [11].

It is interesting to note that a recent study [43] estimated that under the high solar irradiation conditions of the Emirates, the adjusted water cost from a PV-SWRO large scale system with a capacity of 90,000 m³/day would range between 0.38 USD/m³ (base line scenario) to 0.28 USD/m³ (best case scenario).

Thermal desalination technologies present higher economies of scale benefits [44]. The range of cost of water production for MSF is 1.02 to 1.74 with an average of 1.44. Contrarily, economies of scale of SWRO tend to diminish as the production capacity increases. For SWRO plants, the optimum size historically has been 100 MLD (100,000 m³/day) to 200 MLD (200,000 m³/day), resulting in costs of just over USD 1.00/m³, which is the lowest documented until recently [11].

Small-Scale Plants

The estimation criteria for the costs of small-scale desalination plants presents many similarities with large-scale desalination plants. The cost of desalinated water powered by conventional sources of energy is generally reported to be lower than that powered by renewable energy sources. In a study of IRENA [15], the levelled cost of desalinated water produced by conventional power was calculated based on the cost of the electrical kWh. The range of capacities of the RO and MED desalination plants considered was 250 to 2000 m³/day. All in all, as stated by various efforts, the higher the cost of energy, the higher the cost of the produced water for both RO and MED plants. The higher the weighted average cost of capital (WACC) and the smaller the RO desalination plant, the higher the cost of water. Additionally, the cost of RO desalinated water is much lower than that produced by MED.

In [15], the cost of desalinated water produced by renewable energy while the system was connected to the grid was also calculated using different scenarios. Regarding PV, it was considered that 100% of the PV energy was supplied to the RO system (that satisfied 30% of its energy demand) while the grid supplied to the RO the remaining 70%. The cost of energy was assumed to range from USD 0.11/kWh to USD 0.15/kWh for PV and USD 0.20/kWh to USD 0.50/kWh from the grid. The resulting energy cost for the RO plant was between USD 0.18/kWh and USD 0.40/kWh and the corresponding water cost was between 1.5 USD/m³ to 2.5 USD/m³, also depending on WACC. In these calculations, it was assumed that the PV electricity is fully consumed by the RO desalination plant. Another option would be to size the PV system much larger and provide electricity to the grid when more electricity is produced than can be consumed by the desalination plant. Since PV electricity is cheaper than grid electricity, this would make the PV-RO system cheaper. That is, if surpluses of PV electricity could be sold to the grid, the investment would be more attractive. It must be commented that today the PV CAPEX is less than 1000 USD/kWp (at that time it was taken equal to 1470 USD/kWp) and the cost of PV kWh today is much less than USD 0.10; for utility scale PV, the cost reported in [30] for 2020 is USD 0.057/kWh. Therefore, today’s cost of water for a similar PV-RO system is expected to be much lower than those reported above.

In the study of IRENA [15] the cost of RO desalinated water was also estimated for an off-grid (autonomous) PV-RO system. In both cases the RO unit was to produce 250 m³/day. The first case regarded use of batteries to store the electricity provided by the PV system, allowing the RO desalination unit to operate at full load 24 h a day. The second case is to operate the RO unit only during daytime and energy is supplied by the PV system. In the case of the off-grid PV-RO system without batteries, the PV system was assumed to provide electricity to the RO plant for 7 h a day. For 7 hours of operation, the cost of water was calculated as USD 2.30/m³ at 5% WACC and USD 3.30 at 10% WACC. For 24 hours of operation, the cost of water was USD 5.40/m³ at 5% WACC and USD 5.80/m³ at 10% WACC. The cost of water for the 24 h plant is much higher because of the capital and operating costs of the batteries needed, as well as a higher capital cost for the larger PV plant that is required. However, the cheaper variable operation is not yet commercially available. For the same reasons as in the case of the PV (grid connected)–RO system (i.e., lower PV CAPEX), today’s water cost from an off-grid PV–RO system is expected to be much lower than the above-mentioned values.

A more recent study presented additional data on water cost for PV–RO systems [45]. In this study, although the cost varies broadly depending on the system topology, the water salinity and the source of energy go along the same path with those in the study of IRENA. In general, brackish water desalination needs less energy than seawater, which understandably has an impact on cost. Moreover, the choice on how to utilize the renewable energy subsystem also plays a significant role. A recent study deployed by Kyriakarakos and Papadakis [4] showed that the most cost effective option for a system meeting a 100 m³/day water demand was a grid-connected system having a PV system interconnected under a net-metering scheme. Different approaches in meeting the same load along with the net present cost of water are presented in

Table 2. Net Present Cost of desalinated water for different power options used.

Type of Interconnection	RE Technology Used	Net Present Cost (EUR/m3 of Water)
Grid-connected	No RE	0.69
Grid-connected/Net-metering	PV	0.51
Autonomous	PV—Batteries	0.96
Autonomous	Wind turbine—Batteries	1.12
Autonomous	PV-Wind turbine—Batteries hybrid	0.78
Autonomous	Geothermal ORC—Batteries	0.52

. Specific conclusions can be drawn from this study:

• Energy storage increases the cost of the system. The grid connected system with net-metering provides the cheapest water.
• Geothermal energy is, as expected, the renewable energy resource with the minimum cost even at low temperatures near 100 °C. Even in a fully autonomous scenario well, a minimum size battery bank needs to be included [46] and the cost is comparable with the grid connected PVs.
• Hybrid systems provide lower cost of water in comparison to either only PVs or only wind turbine systems. The wind potential of the area needs to be explored.

Further to the above, a new calculation was made using the exact same parameters and assumptions, increasing the generation cost of electricity of the previous study to 219.37 EUR/MWh, which was the average price based on the Greek energy exchange for the period starting on 1 May 2022 and ending on 15 June 2022. The Net Present Cost of water rose from 0.69 EUR/m³ (value of the original study) to 0.89 EUR/m³, a 29% increase. Due to planned subsidies by the government towards end users, the actual increase in electricity bills might be lower, but the difference will be covered by state budget. With electricity prices rising, the case of self-production further strengthens under net-metering/net-billing schemes as well as feed-in tariffs (where still available).

In summary, important conclusions from the above-mentioned studies regarding small-scale renewable-energy-driven RO systems are that:

• Grid-connected PV-RO may be considered the most cost-effective option (in cases where the cost of land is low, which is not always the case in the Greek Aegean islands).
• If the same amount of water is to be produced by an RO plant which is operated by PV without grid backup, the cost rises. Notice that PV off-grid desalination without (or with minimal) electricity storage and RO operating at variable conditions have not yet become a commercial practice.
• Autonomous PV-ROs are expensive mainly because of the need of battery storage and the high cost associated to them.
• The benefits of using self-produced electricity are increasing as the cost of grid electricity is increasing. In Spring 2022, electricity cost in Europe has been increasing due to the situation caused by the conflict in the Ukraine.

2.2. Desalination in Greece

2.2.1. Water Needs in Greek Islands and Existing Water Supply Options

The need for fresh water in the Greek Islands have been rising in the past years, creating deficits that have strong adverse impacts in the daily lives of the local populations as well as in relation to economic development activities in those areas. A multitude of different possible solutions have been utilized in Greece aiming to meet these urgent needs for fresh water.

The western Greek islands in the Ionian Sea Water Sector are having one of the highest rainfalls in the country, thus requiring mostly sustainable planning efforts of their adequate water resources. To the east of the Greek mainland, the Aegean Archipelago comprises another one of the 14 water resources management sectors in Greece according to Law 3199/2003, consisting of about 3000 islands and thus offering a plethora of “closed” island systems. The islands form three sub-clusters: the eastern one, the Cyclades cluster in the middle and the Dodecanese to the southeast. At the same time, they produce an overall very similar locale in the center of the eastern Greek territories. The map of the islands area portrayed in Figure 3 presents the three islandic clusters, as well as illuminates the area’s dimensions having an average north to south axis of about 500 km, and a west to east one of about 300 km. The islands present an exceptional diversity in their morphological characteristics, geology, flora and fauna. However, they present significant common traits, as they share common climatic conditions with rainy mild winters and dry hot summers. They are also semiarid to arid, having the lowest recorded precipitation in Greece (Anafe Island in Cyclades with an average of about 200 mm/yr) and exposed at large to the same economic, social and environmental challenges. Hence, vulnerability to water stresses is also related to overall ecosystem resilience, especially in relation to droughts, floods and increasing anthropogenic disruptions.

The population was 507,764 inhabitants (2011 census) and tourist arrivals reached more than one million annually in 2021 on the most developed islands only, while overall the population during the summer months usually increases up to five times on many islands. As reported in 2002, the total net water resources in the Aegean islands sector are about 1.25 × 10⁹ m³/yr, of which approximately 250 × 10⁶ m³/yr or 20% constitutes groundwater and the remaining 1.0 × 10⁹ m³/yr was considered as the net surface water resources [46]. According to the 2002 estimates, the Aegean islands water sector used approximately 80 × 10⁶ m³/yr for irrigation, 33 × 10⁶ m³/yr for domestic water supply, and only about 1 × 10⁶ m³/yr for industrial purposes [47]. According to the Management Plan of the River Basins of the Aegean islands Water Sector (2nd Revision, 2019, ΕL14), the mean annual runoff for the Aegean Islands is estimated at 560.61 × (10⁶ m³), reflecting the estimate only for Lemnos, Lesvos, Chios, Samos, Ikaria, Tinos, Andros, Naxos, Kos and Rhodes islands; by adding the remaining ones, the 2002 estimate may reached. The mean annual groundwater recharge in the majority of the Aegean Islands is estimated at 885.86 × (10⁶ m³). The mean annual water needs for urban areas is 90.66 × (10⁶ m³), 111.36 × (10⁶ m³) for irrigation, for livestock 2.39 × (10⁶ m³) for livestock and 0.05 × for industrial uses (2nd Revision, 2019, ΕL14).

Finally, the largest house building increase during the 1971–2011 period was in Tinos (130%), Antiparos (128%) and Mykonos (114%). About 50% of the houses in Antiparos are new constructions (compared to 43% in Kea and 42% in Mykonos). Finally, the islands with the highest birth rate are Mykonos (+8.2%), Thera (+7.7%), Rhodes (+5.98%), Kos (+5.9%), Kastellorizo (+4.05%) and Kalymnos (+5%).

Water resources management efforts in islands should also acknowledge both the short- and long-term impacts of planned efforts and not only those of already in-place traditional/existing water management approaches [47,48,49,50,51]. Environmental impacts of existing options not related to wastewater and haphazard urban and touristic development are predominantly clustered on seawater intrusion in the usually sensitive, and with limited water storage, coastal aquifers. Furthermore, such seawater intrusion is exacerbated by groundwater overexploitation and the increasing water deficits due to the small volumes of stored water in the aquifers, as well as to the slow rate of the natural recharge with the existing low annual rainfall and non-favorable geologic conditions. The resulting impacts created significant deterioration of an already fragile ecosystem. Moreover, low water quality produces serious social and economic consequences, as well as low efficiency operation of water supply systems (salt deposits in closed conduits, proliferating maintenance costs, material damages and/or failure, etc.). In addition to that, using as a patch up response, ship water hauling with escalating high-water cost in the recent decades results in extremely high water prices of debatable quality. Hence, holistic water supply options should not only focus on the identification of the least economic cost water supply options. They should also include environmental, technical and social considerations, as well as their pertinent costs, in order to satisfy the water demand particularly in a long planning horizon. Potential options such as surface water dams and desalination both present high investment costs, but they may guarantee the needed water supply security and overall impacts minimization.

In the past decades, the most prominent solution to meet the needs was the water hauling using water tankers. The Greek government, since more than 20 years ago, has subsidized water hauling to the islands, a practice that continues even today. Thus, a significant proportion of the water demand in the smaller Greek islands is still fulfilled using water hauling by tankers. However, water hauling is a costly solution, which may also result in additional environmental impacts, due to carbon dioxide emissions from the pertinent ships, with most of the fleet being quite antiquated. Overall, the cost of water hauling in Greece is high and reached 12.77 EUR/m³ for 2017 and 2018 (including VAT) [52] and the transported water usually needs further treatment to achieve urban water quality standards. The cost to the local or regional governments is quite large and requires a significant amount of their annual budget, which is mostly ranged between 6 to 12 million EUR/year [52], while water hauling is by no means a permanent solution to the scarce water conditions.

Rainwater harvesting (RWH) is a water supply practice utilized for over 4000 years on the islands. RWH may enhance water supply to meet part of domestic needs and it may be a feasible alternative for domestic water use. The quantities saved range from 12% to 100% of the overall demand in a few cases in other parts of Greece [53]. Nevertheless, the high and continuous touristic development in the last decades has not included the construction of cisterns or various types of tanks in the pertinent infrastructure, since it was predominantly oriented towards constructing central water supply systems, using mostly the limited groundwater resources. Contributing to that is the changing lifestyle of both the permanent population and the tourists, having an average water consumption of about 180 l/d per capita. In this regard, a four-member family would require quite a large cistern for its minimum annual needs, making the whole structure technically difficult and quite expensive, apart from the necessary space requirements and rainfall availability. Concluding, notwithstanding the potential advantages, RWH use is still low in many countries, including Greece, since RWH investments present long payback periods [54], while at the same time there is a lack of relevant appropriate legislation for RWH, especially for the regions suffering from water scarcity such as the Aegean islands.

2.2.2. Desalination Systems in Greece

In order to address the challenges faced in providing potable water to island communities, one of the solutions investigated was desalination. The efforts for introducing desalination in the Greek islands started in the 1960s with the first RO plant for public water supply built on Mykonos island between 1981 and 1982. Since then, RO has dominated desalination applications in Greece. It has been utilized extensively for over 40 years both for public water supply as well as providing water to hotels and luxury villas.

More than 160 desalination plants operate in Greece with a total production of more than 150,000 m³/day [55]. In terms of supply water, 56% is seawater, while 41% is brackish water. As far as the use of the produced desalinated water in Greece is concerned, 48% is the supply of municipalities, 31% covers industrial needs, 16% covers tourist requirements, with the remaining 5% covers the needs of energy production plants and also the needs of the Greek military [55]. Reverse osmosis is the most popular desalination process in Greece, as 75% of the desalinated water is produced by reverse osmosis desalination units [52]. Desalination units in the islands are of small capacity ranging from a few hundred cubic meters per day to a few thousand cubic meters per day [56]. The largest part of the desalination systems are operating on islands serviced by isolated electricity grids. These grids in the vast majority of cases are powered by diesel/oil generators. The rest are installed in interconnected islands. The interconnection is realized using submerged cables.

The desalination plants on Chios, Koufonisi and Chalke islands present an SEC of 5 kWh/m³. In Thirassia, Agathonisi and Akrotiri, the specific energy consumption is estimated higher due to the lower installed capacity, which is below 250 m³/day increasing the share of electricity consumption by auxiliary subsystems. The specific energy consumption of newly installed RO units is as low as 2.6 kWh/m³, even for small RO desalination units of a capacity of a few hundred m³/day [4].

The cost of desalinated water in Greece is in the range of 0.5–3.5 EUR/m³ with most cases presenting a cost of above 1.2 EUR/m³ [55]. This is higher when comparing it with large desalination plants, which usually present costs below 1 EUR/m³. The main reason for this is the smaller size of the Greek plants as well as their age [11]. A number of demonstration desalination plants which are powered by renewables have been deployed in Greece, some of which are still in operation. A list is presented in

Table 3. A list of renewable energy desalination systems in Greece.

Location	Desalination Method	Renewable Energy Type	Comments
Syme island	VCD 450 m3/day	750 kW wind turbine	Demo system. Installed in 2009 but stopped operating in 2013
Kimolos island	MED 80 m3/day	Geothermal low enthalpy ~61 °C	Installed in 2000 in the framework of an EU project and stopped operating after the end of the project
Keratea, Attica	RO 3 m3/day	Hybrid PV 4 kWp and wind turbine 900 W	Demo PV-wind-RO system, still in operation
Melos island	RO 3360 m3/day	Wind turbine 600 kW	Installed in 2007. In operation
Herakleia island	RO 80 m3/day	Floating wind turbine 30 kW and a small PV system	Demo system. Installed in 2007. Not in operation
Strogyle island	RO 0.85 m3/h	PV 10 kW	Installed in 2014. In operation

[52].

2.2.3. Powering Desalination in Greece

In Greece, fossil fuels dominate the electricity mix (61.9%) according to the Greek Renewable Energy Sources and Guarantees of Origin Operator for 2020. Greece is experiencing a shift to renewable energy, which began about 10 years ago and continues to make significant progress. The large Greek island complexes have their own generators, which are organized in Autonomous Power Stations, while the smaller islands are either electrified by connection from the larger islands, the mainland or by smaller diesel generators, some of which are occasionally transferred from one island to another [57].

The mass adoption of renewable energy sources began in Greece in 2006, with the installation of photovoltaics. Another relatively new but widely popular energy resource in Greece is the wind. A common feature of most of the Aegean islands is the large wind and solar potential, while in some islands there is a considerable geothermal potential. Nevertheless, despite the abundance of solar and wind energy potential in the Aegean Sea, the progress noted in the field of RES power generation has not been the one expected.

This is due to the local grids’ technical limitations (dynamic penetration limits, thermal units’ technical minima constraints and grid stability) that discourage the installation of new RES plants which will face increased curtailment. The Greek legislation also sets RES power limitations for grid safety reasons, according to the thermal power capacity on each island and in many islands the percentage of RES installed has already reached the max capacity. Therefore, the only available option to install RES-powered desalination on these islands is the autonomous systems.

According to the Preliminary Study of the Ten-Year Development Plan 2022–2031 of the Independent Electricity Transmission Operator (IPTO), the interconnections of Crete, the Cyclades and the Dodecanese islands will create the technical conditions for the creation of new RES installations with a total capacity of more than 2 GW (IPTO announcement, January 2021). This will create a completely new picture in the Aegean Sea, and it will boost RES investments. The interconnection of the islands presents major challenges due to their large number, their dispersion in the Aegean Sea as well as the substantial seasonal irregularity of electricity demand, because of the significant increase of population during the summer months (tourism). Since April 2018, the islands of Syros, Mykonos, Tinos and Paros have been interconnected with the main grid [58].

To date, the electrification of the islands has relied on subsidies that ensure that all the population pays comparable tariffs for electricity. The isolated electricity grids of the islands are powered mostly by diesel and oil-based generators. The central government budget covers the cost of the fuels purchased as well as the transportation cost to the islands. A large part of this cost is covered by a levy paid by all mainland electricity consumers through the electricity bills. When this policy was first introduced, the cost of oil as well as the relevant transportation costs were low. With the decreasing prices of renewable energy technologies, the isolated grids can be hybridized, decreasing fossil fuel consumption and considerably lowering the needed subsidies.

The average total electricity production costs in non-interconnected islands may be multiple times higher than the mainland production cost of electricity [59].

Table 4. Real cost of electricity in non-interconnected islands.

Place	Cost (EUR/MWh) 2012	Cost (EUR/MWh) 2021
Agathonisi	733.99	723.80
Agios Eustratios	444.00	486.34
Amorgos	410.37	436.80
Anafe	480.60	612.66
Antikythera	1748.07	1343.68
Arkioi	893.90	1479.76
Astypalaia	424.15	476.08
Donousa	766.92	1178.32
Ereikousa	878.13	783.08
Thera	204.88	192.37
Ikaria	363.70	447.75
Megiste	492.94	458.12
Melos	263.11	Interconnected with mainland.
Mykonos	360.54
Paros	188.22
Sifnos	381.45	476.38

presents the real cost of electricity in non-interconnected islands for the year 2012 where the Brent oil price averaged 111.63 USD and 2021 where the Brent oil price averaged 70.68 USD [60]. As the oil prices are currently exceeding the 115 USD per barrel price (May 2022), the cost of electricity is expected to further rise. Therefore, it is understood that the real cost of desalinated water in the islands that use the local grid power is much higher than the cost calculated with the subsidized electricity price. Therefore, the cost of water produced by RES-powered desalination should be compared to the real cost of desalinated water that uses the local island electricity grid. For all non-interconnected islands, it makes sense to promote through the regulatory framework PV-RO systems in order to decrease the subsidies that would be required in the case these systems were connected to the main grid.

In the past years, many experimental and demonstration plants have been implemented in several Greek islands for investigating RES-powered desalination. However, today there is only one real commercial desalination unit based on renewable energy, the one on the island of Melos (see Figure 4). Nevertheless, the fact that the kWh cost of several renewable energy technologies, especially wind and photovoltaics, is now much cheaper than the conventional kWh cost prescribes an increased uptake of renewables. Such options along with the prospects of interconnection of many of the Aegean islands with the mainland grid in the coming years offer excellent opportunities for developing RES-powered desalination units to confront the water shortage problem in a sustainable manner.

2.2.4. Strategies for Addressing Water Needs of Islands

Paradigms for island water management were introduced and investigated by Karavitis et al. [51], Voivontas et al. [48] and Assimakopoulos [49]. Figure 5 presents graphically potential solutions for addressing the water needs in islands. SWRO is considered to be a possible solution for addressing permanent and seasonal water needs. It has to be noted, though, that meeting the entire demand especially during the summer months with touristic activity is hard to accomplish. Groundwater use through boreholes and wells is the most common water supply option, which has already created significant short- and long-term negative impacts to the quality and quantity of the supplied water due to salinization. Water hauling by ships is considered only as a last resort option, but increases considerably the already high cost of water in such areas. Renewable energy desalination is a sustainable solution provided that appropriate measures would be taken to stimulate their implementation by introducing innovative business models such as Public Private Partnerships. Demand-side management is another approach to be utilized through an appropriate water tariff pricing framework increasing the cost of water as consumption increases. It has to be noted that the modularity of the RO systems allows modules to be turned off seasonally without affecting their operational lifetime to follow water consumption and extra RO modules can be added to meet future increased demand.

3. Results and Discussion

To determine appropriate water supply options to the islands’ water scarcity conditions, several parameters should be considered. Some of the most crucial ones are the water demand profile throughout the year, existing infrastructure, island area and topography, population and its seasonal variation, economic activities, and the characteristics of the water shortages observed (time and quantities). This has been extensively investigated in the past years, e.g., [50,51,61,62], and various short-, medium- and long-term options for increasing the water supply have been proposed. Some of them have already been implemented such as:

• Constructing new off stream water storage infrastructure (tanks and reservoirs), combined with desalination plants, as well improvement of the operational efficiency of the existing reservoirs.
• Constructing new dams resulting in new water reservoirs.
• Constructing new desalination plants which has become more cost-effective due to the technological progress.
• Water hauling from other neighboring islands or from the mainland—this has to be considered only as a last resort option.
• Maintenance and upgrading of the island’s infrastructure in order to minimize water losses, e.g., extensive maintenance of faulty sections of the water supply networks, complete replacement of antiquated water supply systems, closed conduits water quality management, etc.
• RWH and storage in private or public water reservoirs, wherever this is possible.
• Constructing an additional water supply network in parallel for secondary uses able to utilize lower-quality water.
• Wastewater treatment and reuse for secondary uses.
• Utilization of demand management approaches (i.e., pricing policies, water conservation incentives, public participation and involvement, etc.).

The various aforementioned options to the challenges of water scarcity differ in difficulty, suitability and cost. A central factor that should be considered is the reliability of the proposed options for guaranteed and uninterrupted water supply with an extended operational horizon on the island. One of the obvious options for the Aegean Sea islands would be the erection of large water reservoirs combined with wastewater treatment plants. Such large water reservoirs have low O&M costs and are a long-term, sustainable solution to the problem of water scarcity, which, however, needs considerable rainfall and fitting soil and topography [62].

A sustainable solution to water scarcity ought to ensure an adequate water supply with an extended time horizon, without dependence on external conditions. The realization of non-traditional water supply options, such as desalination plants integrated with appropriate renewables installations, may significantly contribute towards sustainable water resources planning for water scarcity on the islands. Such an effort, in addition to the prerequisite sustainability, has to be economically competitive with any other option of producing fresh water of appropriate quality, thus allowing the dependable water supply to local populations at socially acceptable costs. In this way, it would be feasible to accelerate the sustainable and inclusive economic development of small towns and village communities in the Aegean islands, while improving life quality.

Water supply augmentation by using desalination in the water-scarce Greek islands is adopted and has been widely used already in the past decades. The prospects of desalination as a complementary source (and in many cases as the main source) of fresh water have increased due to desalination technological advancement, as well as the urgent need to address the demand in rapidly growing touristic destinations in the Aegean Archipelago [48]. At both commercial and demonstration levels, research is strong in improving the operational efficiencies of the technologies while decreasing costs at the same time. Such efforts have not always been successful and further reduction of the energy required by the desalination systems still remains a major technical challenge. As energy is the most important cost factor in desalination, methods to reduce specific energy consumption will significantly decrease costs. It is also possible to utilize energy resources of lower cost such as waste heat and geothermal energy in the locations these are available. RES offer a sustainable pathway towards the reduction of the pressure on local electricity systems, and thus the development of desalination plants with renewable energy sources is considered a viable alternative to be pursued meeting energy and water sectors needs in a sustainable manner in Greece, and many incentives should be undertaken in this direction.

Energy storage in the form of batteries still involves high investment costs, but fortunately the operational lifetime of newer technologies (e.g., lithium-ion) is increasing. Other novel energy storage technologies have also been proposed in literature for desalination such as compressed air storage [63]. Hybrid systems, in combination with performance optimization schemes and energy management systems, can also provide viable solutions [64,65]. Wind parks frequently face strong social opposition, a reality exhibited at large in the Greek islands with high touristic activity, as the local population believes that the aesthetic value of the landscape is declining. It has to be highlighted that sustainable planning of wind parks needs to take place, because in some cases in the past vast wind parks were proposed in comparison to the size and population of the islands and this needs to be avoided in the future. The European Commission’s Joint Research Centre report titled “The social acceptance of wind energy” provides a good starting point in this direction. Environmental impacts and land availability may also prevent the use of desalination systems combined with renewable energy technologies. More research and demonstration plants ought to be funded with the aim of achieving an optimal design and decreased cost operation of desalination systems based on renewable energy.

Water shortages on several arid and semi-arid islands are still being dealt with today by transporting water from the mainland. This is financed by the central government for social reasons. At the same time, in many cases the cost of O&M desalination plants is subsidized; it is understandable that subsidies would be in place for social reasons. Still, subsidies would have to be designed appropriately in order to assist the populations in actual need and not to support everyone horizontally, including, for example, profitable investments which can bear the cost of desalination on their own (e.g., luxurious resorts). Various forms of private sector involvement should be considered (e.g., Melos Island plant). In addition, for the sustainable management and development of water and energy resources that are under increasing pressure in the coming decades, it is vital to properly assess compensation and encourage cross-sectoral planning.

The past Greek experience highlights that the construction and operation of desalination plants is feasible with subsidies [56]. Notwithstanding the fact that the desalination energy intensity is high as well as the fact that RO is still expensive, it is at the same time a robust technology and is able to cover the requirements in drinking water. At the same time, technological advancements and decreasing costs of RES technologies are currently driving the final water costs down. Consequently, for the Greek islands, certain technical, implementation and policy challenges including innovative financing mechanisms need to be deployed so that RO desalination powered by renewables may become part of a national sustainable integrated water management policy.

An additional important factor to be discussed is the current status of the water supply networks, which are in poor condition on most islands or in some areas not existing at all. This reality constitutes a barrier to the widespread uptake of desalination solutions. A flexible and independent public service could be set up to oversee all public procurement desalination plants, to ensure the implementation of good practice at all steps of the public water supply, to ensure the quality of water provided and to employ incentives to reduce energy requirements of existing systems and consequently cost. Desalination powered by renewables’ success stories may be also used as good examples for promotion.

Concluding, water supply is a socially and politically sensitive issue for the island populations. Although the public is now familiar with desalination, the state and local authorities need to raise the awareness of the local people about the long-term benefits of water desalination, and system operators must utilize high standards to ensure optimal and cost-effective operation, while protecting the local environment. Social acceptance is key in deploying wind and geothermal energy (where available) and can be an enabler of ensuring cost-effective energy and water provision paving the path towards sustainable development.

4. Conclusions

Water and energy security will be top political priorities for Greece in future, following EU policy. Detailed reviews of utilized solutions for the provision of fresh water may shed light on the advantages and disadvantages of the various technologies deployed on the islands and reveal alternative management options (i.e., appropriate operation/practices) for a more efficient production. The ongoing research in improving the desalination technologies powered by renewables aims at improving operational efficiencies and the overall cost reduction. Nevertheless, interregional and international cooperation will be required, as the challenge of providing fresh water in a sustainable manner may be too great for any country, region, and development funding body to address on its own. Water supply, in the nearest future in the islands of the central and south-eastern Aegean Sea, is expected to use desalination combined with solar and/or wind energy, instead of fossil fuels, in full alignment with the EU Green Deal.